Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
HIV-1 Viral protein R ( Vpr ) induces a cell cycle arrest at the G2/M phase by activating the ATR DNA damage/stress checkpoint . Recently , we and several other groups showed that Vpr performs this activity by recruiting the DDB1-CUL4A ( VPRBP ) E3 ubiquitin ligase . While recruitment of this E3 ubiquitin ligase complex has been shown to be required for G2 arrest , the subcellular compartment where this complex forms and functionally acts is unknown . Herein , using immunofluorescence and confocal microscopy , we show that Vpr forms nuclear foci in several cell types including HeLa cells and primary CD4+ T-lymphocytes . These nuclear foci contain VPRBP and partially overlap with DNA repair foci components such as γ-H2AX , 53BP1 and RPA32 . While treatment with the non-specific ATR inhibitor caffeine or depletion of VPRBP by siRNA did not inhibit formation of Vpr nuclear foci , mutations in the C-terminal domain of Vpr and cytoplasmic sequestration of Vpr by overexpression of Gag-Pol resulted in impaired formation of these nuclear structures and defective G2 arrest . Consistently , we observed that G2 arrest-competent sooty mangabey Vpr could form these foci but not its G2 arrest-defective paralog Vpx , suggesting that formation of Vpr nuclear foci represents a critical early event in the induction of G2 arrest . Indeed , we found that Vpr could associate to chromatin via its C-terminal domain and that it could form a complex with VPRBP on chromatin . Finally , analysis of Vpr nuclear foci by time-lapse microscopy showed that they were highly mobile and stable structures . Overall , our results suggest that Vpr recruits the DDB1-CUL4A ( VPRBP ) E3 ligase to these nuclear foci and uses these mobile structures to target a chromatin-bound cellular substrate for ubiquitination in order to induce DNA damage/replication stress , ultimately leading to ATR activation and G2 cell cycle arrest .
HIV-1 encodes several proteins termed accessory that have been implicated in the modulation of host cell environment to promote efficient viral replication and evasion from innate and acquired immunity [1] . One of these accessory proteins , viral protein R ( Vpr ) , is a small amphipathic protein of 96 amino acids . In addition to being expressed in infected cells , Vpr is packaged into virions through an interaction with the p6 domain of the Gag polyprotein precursor [2] , [3] , [4] . The molecular structure of Vpr was recently resolved and found to consist of a hydrophobic core comprising three interacting alpha helices flanked by N- and C-terminal flexible domains [5] . Of note , the third alpha helix includes a leucine-rich region essential for the stability of the core and the flexible C-terminus comprises a functionally important stretch of positively charged arginine residues [6] . Several biological functions have been attributed to Vpr including transactivation of the viral long terminal repeat ( LTR ) , enhancement of infection in macrophages , induction of apoptosis , and promotion of a cell cycle arrest at the G2/M phase [7] . Vpr-mediated G2 arrest likely plays an important role in vivo for viral replication or pathogenesis given that this activity is highly conserved among primate lentiviruses [8] , [9] and since abnormal accumulation of cells in G2/M can be observed in HIV-infected individuals [10] . Indeed , recent studies reported that Vpr upregulated the expression of ligands for the activating NKG2D receptor and promoted natural killer ( NK ) cell-mediated killing by a process that relied on Vpr ability to induce a G2 arrest , thus suggesting an immunomodulatory role for Vpr that may not only contribute to HIV-1-induced CD4+ T-lymphocyte depletion but may also take part in HIV-1-induced NK cell dysfunction [11] , [12] . Several investigators have reported that Vpr-induced cell cycle arrest involves the activation of the ATR ( ataxia telangiectasia-mutated and Rad3-related; NM_001184 ) -mediated G2/M checkpoint [10] , [13] , [14] . ATR is a kinase of the phosphatidylinositol 3 kinase-like family and is involved in the activation of the G2/M checkpoint and in the coordination of DNA repair following the occurrence of DNA damages or DNA replication stress . Activation of ATR by exogenous DNA damaging agents such as UV leads to phosphorylation of several effector molecules , including Chk1 and H2AX ( histone 2A , variant X; NM_002105 ) , inducing the formation of DNA repair foci containing γ-H2AX ( phosphorylated H2AX ) , MDC1 ( mediator of DNA damage checkpoint 1 ) , 53BP1 ( p53 binding protein 1; NM_001141979 ) , BRCA1 ( breast cancer 1 ) , as well as the RPA ( replication protein A ) , 9-1-1 ( Rad9-Hus1-Rad1 ) , and Rad17 complexes on the sites of DNA damage [15] , [16] . Activation of ATR by Vpr similarly leads to phosphorylation of Chk1 and to the formation of DNA repair foci containing γ-H2AX , 53BP1 , RPA , Hus1 , Rad17 , and BRCA1 [13] , [14] , [17] , [18] . The immediate cause of the activation of ATR following Vpr expression has remained elusive but implicates in part the recruitment by Vpr of the host DDB1 ( damage DNA binding protein 1; NM_001923 ) -CUL4A ( cullin 4A; NM_003589 ) E3 ubiquitin ligase complex via a direct binding to the substrate specificity receptor VPRBP ( Vpr-binding protein , also known as DCAF1; NM_014703 ) [19] , [20] , [21] , [22] , [23] , [24] , [25] . Specifically , RNA interference-mediated depletion of VPRBP or mutations in the hydrophobic leucine-rich core domain of Vpr impaired association to the E3 ligase complex and induction of G2 arrest . In contrast , G2 arrest-defective mutants of Vpr in the C-terminal arginine-rich domain , which maintained their association to the E3 ligase , nevertheless failed to induce G2 arrest [19] , [20] , [21] , [22] , [23] , [24] , [25] . These results indicate that association of Vpr to the E3 ligase complex is required but not sufficient to induce G2 arrest , thus supporting a model in which Vpr would act as a connector between a ubiquitin ligase complex and a yet-unknown cellular protein . We recently provided evidence that Vpr-induced K48-polyubiquitination and proteasomal degradation of this protein ( s ) would lead to DNA damage/stress , activation of ATR , and ultimately G2 cell cycle arrest [26] . HIV-2 and some species of simian immunodeficiency virus ( SIV ) encode a paralog of Vpr , called Vpx , which does not induce G2/M arrest but instead counteracts a putative restriction factor expressed in macrophages and dendritic cells that affects infection at a post-entry step [1] . Interestingly , Vpx also interacts with DDB1-CUL4A ( VPRBP ) via its hydrophobic leucine-rich core domain . This association is required for the inactivation of the restriction factor and probably leads to its proteasomal degradation [27] , [28] , [29] . The subcellular localization of Vpr and its importance for the induction of G2 arrest has remained a source of controversy . Several investigators reported that Vpr expressed in absence of any other viral proteins primarily localized to the nucleus in a diffuse pattern [30] , [31] , [32] , [33] , [34] , [35] , [36] while others observed a significant accumulation at the nuclear envelope [37] , [38] , [39] , [40] , [41] , [42] . Of note , Sherman et al . showed that Vpr shuttles between the cytoplasm and nucleoplasm [43] . Moreover , Vpr has been shown to form punctuate structures in the nucleus [17] as well as to induce and co-localize with nuclear membrane herniations [44] . C-terminal mutations impairing G2 arrest did not alter localization of Vpr whereas other mutations , predominantly in the first alpha-helix , impaired both nuclear localization and G2 arrest , implying that nuclear/nuclear-envelope localization of Vpr would be required but not sufficient for this activity [33] , [38] . In agreement with this model , Lai et al showed that nuclear punctuate structures formed by Vpr were associated to chromatin and partially co-localized with γ-H2AX , suggesting that Vpr might target host cell DNA and interfere with DNA replication [17] . In contrast , the F34I , V57L , R62P , L68S , and I70S mutations in Vpr caused a re-localization of the protein to the cytoplasm without significantly affecting the induction of G2 arrest [30] , [36] , [37] , [41] . Although inconsistent results were also obtained for some of these mutants [38] , these data would suggest instead that Vpr does not induce G2 arrest from the nucleus but from an extra-nuclear compartment . Therefore , the spatial prerequisites for the induction of Vpr-mediated G2 arrest remain unclear . Additionally , while recruitment of the DDB1-CUL4A ( VPRBP ) E3 ubiquitin ligase complex has been shown to be critical for G2 arrest , the subcellular compartment where this association occurs and becomes functionally relevant is still unknown . We thus sought to locate the Vpr-VPRBP interaction and to determine the relevance of this localization for the induction of G2 arrest with the goal of furthering our understanding of the mechanism underlying Vpr activation of ATR and providing important information on the potential substrate targeted by Vpr . Herein , we show that Vpr forms nuclear foci that contain VPRBP and that partially co-localize with DNA repair foci components , such as γ-H2AX , RPA32 ( replication protein A2 , 32kD; NM_002946 ) and 53BP1 . Moreover , we provide evidence that formation of these Vpr nuclear foci constitute a critical early event in the induction of G2 arrest . We also show that Vpr associates to chromatin via its C-terminal domain and that it binds VPRBP on chromatin . Finally , we observed that Vpr foci were highly mobile nuclear bodies . Our results suggest that Vpr recruit the DDB1-CUL4A ( VPRBP ) E3 ubiquitin ligase complex within mobile nuclear structures to target a chromatin-bound substrate whose ubiquitination and proteolysis would activate ATR and induce G2 arrest .
The interaction between Vpr and VPRBP was previously revealed to be required for the induction of a G2 cell cycle arrest [19] , [20] , [21] , [22] , [23] , [24] , [25] . However , the subcellular localization where this event might take place still remains to be determined . To this end , we performed laser-scanning confocal fluorescence immunohistochemistry to identify the respective subcellular localization and potential co-localization of Vpr and VPRBP . HeLa cells were transduced with a lentiviral vector co-expressing HA-tagged Vpr ( HA-Vpr ) and GFP or a control lentiviral vector expressing GFP alone . Forty-eight hours after transduction , cells were fixed , permeabilized and stained with antibodies against HA , VPRBP , and nucleoporin . The localization of HA-Vpr was mostly diffuse in the nucleus at standard amplification gain ( data not shown ) . However , when the gain was reduced , we could observe that HA-Vpr formed small circular nuclear structures of variable relative sizes that we refer to thereafter as Vpr nuclear foci ( VNF ) ( Figure 1A ) . The number of Vpr nuclear foci varied from cell to cell and from experiment to experiment but generally averaged 35 foci ( SD±10 ) per cell . Formation of these foci was not due to the HA tag because we observed that native Vpr could also form nuclear foci ( Figure S1A ) . Endogenous VPRBP was found to be mostly localized to the nucleus in a punctuate pattern ( Figure 1A ) . We observed that HA-Vpr colocalized with endogenous VPRBP in the nucleus . Strikingly , a significant fraction but not all of Vpr nuclear foci co-localized with VPRBP foci , suggesting that Vpr might be able to recruit the E3 ubiquitin ligase complex to these discreet structures . Of note , in presence of Vpr , we also observed some nuclear membrane perturbations reminiscent of the previously described Vpr-induced membrane herniations [44] . Importantly , transduction of activated primary CD4+ T-lymphocytes with a lentiviral vector expressing HA-Vpr also resulted in the formation of Vpr nuclear foci that co-localized with VPRBP ( Figure 1B ) , indicating that these foci are not solely the result of overexpression of Vpr in transformed cell lines but that their formation also occurs in a physiological cellular host of HIV . Infection of HeLa cells with a VSV-G- pseudotyped virus expressing HA-Vpr ( HxBru HA-Vpr ) also resulted in the formation of Vpr nuclear foci in a minor fraction of cells ( Figure S1B ) . However , the majority of cells displayed a relocalization of HA-Vpr to cytoplasmic compartments ( Figure S1B ) , suggesting that formation of these foci would be a dynamic process , regulated over time during the infection cycle . Indeed , preventing Vpr interaction with Gag and subsequent packaging of the protein into virions using mutations in the p6 domain of Gag ( LF/PS ) or in Vpr ( L23F ) [2] , [32] , resulted in the accumulation of Vpr nuclear foci ( Figure S1C ) . These results provide evidence of the dynamic interplay between Vpr nuclear foci and Gag during infection . To show that the observed co-localization between Vpr and VPRBP foci was not fortuitous and that Vpr foci truly contained VPRBP , we used an in situ proximity ligation assay ( PLA ) [45] . This assay is based on the ligation of antibody-coupled DNA molecules when these are in close proximity ( when secondary antibodies are less than 400 angströms apart ) . Amplification of ligation products and hybridization with fluorochrome-labelled probes allow the detection of physiological interaction in situ without the need to overexpress proteins fused to fluorescent markers . Using this technique , HA-Vpr was found in close proximity to endogenous VPRBP in dense nuclear foci ( Figure 1C ) . We did not observe similar interactions in mock-transfected cells ( Figure S2 ) , in cells expressing a Vpr mutant ( Q65R ) impaired for its interaction with VPRBP [19] , [20] , [22] ( Figure S2 ) , or when any of the primary antibodies where omitted ( data not shown ) . These results therefore suggest that Vpr forms nuclear foci containing VPRBP . To investigate the nature and composition of these Vpr nuclear foci , we first evaluated whether these would correspond to known well-defined nuclear bodies with similar sizes and numbers . We did not however find any significant co-localization with the canonical nuclear speckle marker SC35 ( also known as SFRS2 ) or with PML ( promyelocytic leukemia ) bodies ( Figure S3 ) . Lai et al . previously reported formation and partial co-localization of Vpr nuclear foci with γ-H2AX [17] . We thus evaluated if the Vpr nuclear foci described herein where the same foci that Lai et al . reported . Interestingly , we observed a partial co-localization between HA-Vpr nuclear foci and 53BP1 ( Figure 2A ) . Indeed , expression of HA-Vpr induced the re-localization of 53BP1 from its sites of residence in the nucleus to DNA repair foci , some of which were positive for HA-Vpr foci . We also observed a partial co-localization between some HA-Vpr nuclear foci and phosphorylated RPA32 ( Figure 2B ) . Similar results were obtained for γ-H2AX ( Figure S4A ) . Co-localization of Vpr with components of DNA repair foci suggest that formation of Vpr nuclear foci might represent an early event in the induction of G2 arrest that would be responsible for the generation of DNA replication stress or DNA damage . Conversely , those might simply reflect the re-organization of the nuclear compartment following the activation of the ATR checkpoint by Vpr . To distinguish between these two possibilities , we transduced HeLa cells with a lentiviral vector expressing HA-Vpr and concomitantly treated the cells with caffeine , a non-specific inhibitor of ATR and ATM ( ataxia telangiectasia mutated ) . In these conditions , the addition of caffeine inhibited Vpr-induced cell cycle arrest ( data not shown; [12] ) . However , we did not detect significant changes in the number of Vpr nuclear foci ( Figure 3A , 33±10 for non-treated cells vs 32±9 for caffeine-treated cells ) , suggesting that formation of these foci would take place independently of the activation of ATR . Moreover , consistent with the observation that not all Vpr nuclear foci co-localized with VPRBP ( Figure 1A ) , depletion of VPRBP by siRNA ( 95%±3 . 5% knockdown relative to scrambled siRNA ) in HeLa cells ( Figure 3B ) did not significantly alter the number of foci ( 36±10 for control siRNA vs 33±8 for VPRBP siRNA ) ( Figure 3C ) , suggesting that VPRBP is dispensable for the formation of Vpr nuclear foci . Similar results ( data not shown ) were obtained in a HEK293T monoclonal cell line stably depleted of VPRBP [26] . Moreover , in contrast to its absence of effect on Vpr foci , knockdown of VPRBP abrogated Vpr-induced formation of DNA repair foci containing γ-H2AX and 53BP1 ( Figures S4A and S4B ) . These results indicate that Vpr forms nuclear foci prior to and independently of the activation of ATR and suggest that it is Vpr that recruits VPRBP to these foci and not the inverse . To evaluate the potential role of these Vpr nuclear foci in the induction of G2 arrest , we monitored the capacity of several G2 arrest-defective Vpr mutants to form these foci . HeLa cells were transfected with plasmids expressing HA-tagged Vpr mutants and formation of nuclear foci was evaluated by fluorescence immunohistochemistry and confocal microscopy ( Figure 4 ) . We found that Vpr ( R80A ) , which still interacts with the E3 ligase but is strongly attenuated for the induction of G2 arrest , was defective for the formation of nuclear foci ( 2 . 5±1 . 1 ) , even though its subcellular localization was nuclear . Deletion of the C-terminus of Vpr ( Vpr 1–78 ) , which also maintains the association with the E3 ligase [22] but impairs the induction of G2 arrest [46] , similarly resulted in a defect in the formation of nuclear foci ( Figure 4 ) . Similar results were also obtained with the C-terminal mutants Vpr ( S79A ) and Vpr ( 1–86 ) ( data not shown ) . Vpr ( Q65R ) , which is unable to associate with the E3 ligase and is consequently defective for G2 arrest , was found to be defective for the formation of nuclear foci and also accumulated in cytoplasmic aggregates . Similar localization phenotypes where observed for Vpr ( H71R ) , a mutant of Vpr also defective for its interaction with VPRBP [21] ( data not shown ) . The results obtained with the Q65R and H71R mutations are in contrast with the siRNA-mediated depletion of VPRBP which did not block the formation of Vpr nuclear foci , suggesting that these mutant proteins might have additional defects besides an impaired interaction with VPRBP ( see below ) . These results thus suggest that the C-terminal domain of Vpr , which is required for the induction of G2 arrest , is also critical for the formation of Vpr nuclear foci . The observation that C-terminal G2 arrest-defective mutants of Vpr are compromised in their capacity to form nuclear foci suggests that these nuclear foci might constitute an important early event in the induction of G2 arrest by Vpr . To directly address this possibility , we first evaluated the functional effect of artificially sequestering Vpr in the cytoplasm by overexpression of Gag-Pol . Co-transfection of HeLa cells with HA-Vpr- and Gag-Pol-expressing plasmids produced a sequestration of HA-Vpr in p24-positive cytoplasmic compartments ( Figure 5A ) . This sequestration abrogated Vpr nuclear foci formation ( Figure 5A ) . Similar results were obtained in HEK293T cells ( data not shown ) . To evaluate the functional effect of this cytoplasmic sequestration of Vpr , HEK293T cells were co-transfected with plasmids expressing HA-Vpr and Gag-Pol or with adequate empty plasmid controls . Two days later , the cell cycle and expression profiles of transfected cells were evaluated by flow cytometry and western blot ( Figures 5B and 5C ) . Expression of HA-Vpr alone produced an accumulation of cells in G2/M ( G2+M:G1 = 1 . 81 vs 0 . 66 for mock-transfected cells ) . Interestingly , overexpression of Gag-Pol completely abrogated HA-Vpr-induced G2 arrest ( G2+M:G1 = 0 . 67 ) in absence of any significant effect on the cell cycle when expressed alone ( G2+M:G1 = 0 . 77 ) . Inhibition of Vpr-induced G2 arrest by overexpression of Gag-pol was dependent on the Gag-Vpr interaction and was not the result of some non-specific effects on the cell cycle since Vpr ( L23F ) , a mutant of Vpr unable to bind the p6 domain of Gag [2] , could form nuclear foci ( Figure S5A ) but was impervious to the effect of Gag-Pol on Vpr nuclear localization ( Figure S5B ) and induction of G2 arrest ( Figure S5C ) . Although overexpression of Gag-Pol led to a reduction of the affinity between HA-Vpr and endogenous VPRBP , the overall increase in the expression of HA-Vpr resulted in an increase in the levels of Vpr-bound VPRBP ( Figure 5D ) , excluding the possibility that overexpression of Gag-Pol inhibited G2 arrest by preventing the Vpr-VPRBP interaction . The observed inhibition of G2 arrest by overexpression of Gag-Pol is however unlikely to have a significant role at physiological levels of expression given that infection with a wild type virus led to a G2 arrest that was as efficient as the one obtained with a virus unable to relocalize Vpr from the nucleus because of a mutation in the P6 domain of Gag ( LF/PS ) ( Figures S1C and S5D ) . Overall , these results imply that nuclear localization of Vpr and possibly the formation of nuclear foci would be required for the induction of G2 arrest . To further show that the formation of these Vpr nuclear foci is critical for the induction of G2 arrest , we evaluated the capacity of SIV Vpr and its paralog Vpx to form these foci . Both of these proteins are able to associate with the E3 ligase complex but in contrast to Vpr , Vpx does not induce G2 arrest but counteract a putative restriction factor in macrophages and dentritic cells [27] , [28] , [29] . HeLa cells were transfected with plasmids expressing either HA-tagged sooty mangabey Vpr ( HA-Vpr sm ) or Vpx ( HA-Vpx sm ) . Two days after transfection , cells were fixed , permeabilized , and stained with antibodies against HA and nucleoporin ( Figure 6 ) . Consistent with its ability to induce G2 arrest ( data not shown and [9] ) , we found that Vpr sm could accumulate into nuclear foci ( 16±4 foci per cell ) in contrast to the G2-arrest incompetent Vpx that did not form any foci despite being present in the nucleus ( Figure 6 ) . Taken together , these results indicate that formation of Vpr nuclear foci is an early event that is required to induce G2 arrest . These results also indicate that nuclear localization of Vpr is not sufficient to induce formation of foci . Given that these foci constitute an early event in the induction of G2 arrest , we sought to determine how they would form . These foci are likely the results of a local observable accumulation of Vpr either through oligomerization of the protein or following its recruitment by a locally abundant tethering factor . To distinguish between these two possibilities , we first monitored the dimerization efficiency of the Vpr mutants Q65R and R80A , which are defective for foci formation . HEK293T cells were co-transfected with plasmids expressing enhanced yellow fluorescence protein ( eYFP ) fused to the N-terminus of wild type Vpr and renilla luciferase ( Rluc ) fused to the N-terminus of wild type Vpr and mutants . Two days after transfection , self-affinity was assessed by bioluminescence resonance energy transfer ( BRET ) . Figure 7A reveals that all Vpr fusion proteins were efficiently expressed . In this system , we observed a specific energy transfer between eYFP-Vpr ( WT ) and Rluc-Vpr ( WT ) ( Figure 7B ) . The maximum energy transfer at saturation ( BRETmax ) was 0 . 983 and the concentration of acceptor at 50% of BRETmax ( BRET50 ) was 0 . 397 . In contrast , co-expression of eYFP and Rluc-Vpr did not lead to any significant energy transfer , demonstrating the specificity of the eYFP-Vpr/Rluc-Vpr interaction . The Q65R mutant , showed a significant decrease in its affinity for wild type eYFP-Vpr ( BRET50 = 0 . 791 , 50% self-affinity ) as well as a drastic decrease in BRETmax ( 0 . 314 for Q65R vs 0 . 983 for wild type Vpr ) , suggesting that in addition to a reduction in dimerization efficiency , formation of higher-order complexes ( multimerization ) would also be synergistically decreased . In contrast , the R80A mutant displayed an affinity for wild type Rluc-Vpr that was at least comparable to wild type Vpr ( BRET50 = 0 . 326 , 121% self-affinity relative to wild type ) . Similar results were obtained when eYFP-Vpr R80A and Rluc-Vpr R80A were co-expressed ( data not shown ) . Thus , these results suggest that the ability of Vpr to oligomerize does not directly correlate with nuclear foci formation and does not explain the defect in foci formation observed in the context of C-terminal mutants . To determine if oligomerization of Vpr could still be involved in formation of Vpr nuclear foci , we performed trans-complementation experiments in HeLa cells and monitored formation of Vpr foci by immunofluorescence confocal microscopy . Trans-complementation of HA-Vpr R80A with eYFP-Vpr could rescue the defective phenotype of the R80A mutant by re-localizing the protein to eYFP-Vpr foci ( Figure 7C ) . In contrast , eYFP-Vpr was unable to re-localize the HA-tagged Q65R mutant ( Figure 7C ) , suggesting that oligomerization of Vpr , although not sufficient to induce formation of Vpr foci , may however contribute to the process to some degree . Since oligomerization does not fully account for the ability of Vpr to form foci , Vpr could thus be tethered to specific sites by a cellular co-factor . Co-localization of Vpr nuclear foci with chromatin-bound factors detected at DNA repair sites suggests that this tethering co-factor could be a chromatin-bound protein or structure or DNA itself . To assess this possibility , HeLa cells were first transiently transfected with an empty plasmid or a plasmid expressing HA-Vpr and cells were lysed with 0 . 5% Triton X-100 , resulting in the release of soluble proteins . Treatment of Triton-insoluble cellular material , including chromatin , with microccocal nuclease resulted in the solubilization of chromatin-bound cellular proteins including RPA70 ( replication protein A1 , 70 kDa ) ( data not shown ) and histone 3 ( Figure 8A ) . These proteins were not detected when cell extracts were incubated in buffer without microccocal nuclease . Importantly , chromatin extracts were not contaminated with cytoplasmic proteins as revealed by the absence of GAPDH ( glyceraldehyde-3-phosphate dehydrogenase ) ( Figure 8A ) . Using this system , we found that a fraction of HA-Vpr was released in extracts treated by microccocal nuclease but not with buffer alone , indicating that Vpr associates with chromatin directly or indirectly via other proteins ( Figure 8A ) . A specific association of a fraction of endogenous VPRBP with chromatin was also observed in presence and in absence of Vpr ( Figure 8A ) . To determine whether the defects of foci formation observed with C-terminal mutants of Vpr would correlate with a defect in chromatin association , we analyzed the capacity of several Vpr mutants to associate to chromatin in HeLa cells . Interestingly , both Vpr ( R80A ) and a C-terminal deletion mutant ( Vpr 1–78 ) showed a drastic reduction in their association to chromatin ( Figure 8B ) . Of note , Vpr ( Q65R ) ( Figure 8B ) and Vpr ( H71R ) ( data not shown ) also failed to associate with chromatin , possibly explaining their unexpected incapacity to form foci . In contrast , knockdown of VPRBP did not significantly alter the affinity of Vpr for chromatin ( Figure 8C ) , suggesting that VPRBP does not contribute to this association and that the absence of chromatin association with the Q65R and H71R mutants is not due to its impaired binding to VPRBP . Therefore , the ability of Vpr to form foci correlates with its ability to associate with chromatin . Co-localization of Vpr nuclear foci with VPRBP and the association of both proteins to chromatin suggest that they might interact on chromatin . To evaluate this possibility , we transfected HeLa cells with an empty plasmid or a plasmid expressing HA-Vpr and performed anti-HA immunoprecipitations on proteins released from chromatin by microccocal nuclease ( Figure 9A ) . Interestingly , we could detect co-immunoprecipitation of endogenous VPRBP specifically in cells extracts containing HA-Vpr , in the soluble fraction as well as in the chromatin fraction ( Figure 9A ) . Deletion of the C-terminal domain of Vpr abrogated its interaction with VPRBP on chromatin but not in the soluble fraction ( Figure 9B ) , demonstrating the specificity of these interactions . These data suggest that Vpr interacts with VPRBP on chromatin . Importantly , histone 3 did not co-immunoprecipitate with HA-Vpr in the chromatin fraction ( Figure 9A ) . Moreover , treatment with high concentrations of ethidium bromide during the immunoprecipitation , a treatment that displace proteins from DNA [47] , did not disrupt the Vpr-VPRBP interaction in the soluble fraction as well as on chromatin ( Figure 9B ) , thus excluding the possibility that the observed Vpr-VPRBP interaction was mediated by incompletely digested chromatin fragments . Nuclear bodies stably or transiently associating with chromatin are generally dynamic structures , either in mobility or in stability . For instance , PML bodies display varying levels of mobility in the nucleus . Conversely , DNA repair foci show limited mobility but can rapidly form in response to genotoxic stress and can disassemble following checkpoint recovery [48] , [49] , [50] . To investigate the possible dynamic nature of Vpr nuclear foci , we performed time-lapse confocal microscopy in living HeLa cells expressing eYFP-Vpr . Strikingly , observation of eYFP-Vpr foci for two minutes ( at two-second intervals ) revealed that these were highly mobile structures ( Figure 10A; Videos S1 and S2 ) . Software-assisted tracking of over fifty Vpr foci ( Figure 10B and data not shown ) revealed rates of displacement ranging from 0 . 05 µm/min to 8 . 30 µm/min for an average of 0 . 73 µm/min ( SD = 1 . 00 µm/min; median = 0 . 30 µm/min ) . The mobility of Vpr foci was not dependent on the presence of VPRBP since its knockdown ( Figure 10C ) did not significantly alter their dynamic behavior ( average rate of displacement of 0 . 85 µm/min for VPRBP siRNA vs 0 . 92 µm/min for scrambled siRNA; P = 0 . 78 ) . Because some eYFP-Vpr foci seemingly appeared and disappeared during the course of these observations , we performed time-lapse spinning-disk microscopy analyses to evaluate whether these foci were translating in and out of the focal plane or instead assembling and disassembling . Tracking of eYFP-Vpr nuclear foci for 15 minutes at intervals of 5 seconds highlighted translational movements in the three axes ( Figure S6A ) . Moreover , these analyses did not reveal any apparition or disappearance of nuclear foci ( data not shown ) , suggesting that these are structurally stable . Similar results were obtained from observations over longer periods of time ( 30 minutes ) . Monitoring of the mean fluorescence of eYFP-Vpr in foci showed relatively stable signal over time ( Figure S6A , right panel ) . Some transient fluctuations in fluorescence were however detected . To determine if these fluctuations could be the result of quick exchange of Vpr molecules in and out of nuclear foci , we performed fluorescence recovery after photobleaching ( FRAP ) analyses on eYFP-Vpr foci ( Figure S6B ) . Photobleaching of eYFP-Vpr foci did not however lead to any fluorescence recovery even after an extensive period of time ( 350 seconds ) , suggesting that the inter-exchange of Vpr molecules is minimal . Overall , our results suggest that Vpr would associate to chromatin-bound nuclear foci via its C-terminus . These would serve as a mobile scaffold to recruit the DDB1-CUL4A ( VPRBP ) E3 ubiquitin ligase to induce the ubiquitination and degradation of a chromatin-bound substrate , resulting in DNA damage or replication stress .
Our results show that Vpr mainly localizes to the nucleus in transformed epithelial cells , such as HeLa and HEK293T cells , as well as in primary CD4+ T-lymphocytes ( Figure 1 and data not shown ) . We noticed that the localization of Vpr in HeLa cells closely resembles that observed in primary CD4+ T-lymphocytes , prompting us to select this cellular model for most of our study . Moreover , we found that ectopically expressed HA-tagged Vpr had a subcellular localization similar to that of the native protein ( Figure S1A ) . In infected cells , the nuclear localization of Vpr appears transient because Gag interacts with Vpr to package the protein into assembling viral particles ( Figures S1B and S1C ) . Our localization data show that Vpr can form nuclear punctuate structure that we termed Vpr nuclear foci ( Figure 1 ) , as was reported previously by Lai and colleagues [17] . It is noteworthy that these foci are not readily apparent and require careful calibration of gain to be observed ( data not shown ) . Importantly , we observed a strong co-localization of Vpr with VPRBP in the nucleus , particularly in these foci . In situ proximity ligation assays confirmed the close proximity of the two proteins in these foci ( Figure 1C ) , suggesting that Vpr interacts with the E3 ubiquitin ligase at the levels of these punctuate structures . In contrast to the observations of other investigators [37] , [38] , [39] , [40] , [41] , [42] , we did not observe a significant accumulation of Vpr at the nuclear membrane in these cell types . Several technical reasons might explain these discrepancies , including cell types , levels of expression , fixation and permeabilization conditions , or the tag used . Indeed , we did observe an enrichment of eYFP-Vpr at the nuclear membrane of Hela cells ( Videos S1 and S2 ) . We obtained several lines of evidence demonstrating that Vpr nuclear foci are involved in Vpr-mediated G2 arrest . First , we observed a partial co-localization between these foci and RPA32 , 53BP1 and γ-H2AX , which are usually detected at DNA repair sites ( Figures 2 and S4 ) . Similar results were obtained by Lai and colleagues with γ-H2AX [17] . Secondly , C-terminal mutants of Vpr defective for G2 arrest failed to induce formation of Vpr foci despite their nuclear localization ( Figure 4 ) . Thirdly , cytoplasmic sequestration of Vpr by overexpression of Gag inhibited G2 arrest as well as foci formation ( Figure 5 ) . Fourthly , only Vpr from sooty mangabey SIV but not its G2 arrest-defective paralog Vpx was able to form these foci ( Figure 6 ) . Lastly , the reduced number of foci formed by sooty mangabey Vpr in comparison to HIV-1 Vpr correlated with reduced G2 arrest activity in human cells ( data not shown and [9] ) . All these results suggest that formation of foci is linked to G2 arrest . Moreover , these results also suggest that nuclear localization of Vpr is required but not sufficient to induce formation of these foci . Our results and conclusions are in contrast with previous reports , including one of ours , describing cytoplasmic mutants of Vpr that retain their G2 arrest activity [30] , [36] , [37] , [41] . We had reported over a decade ago that the V57L and R62P mutations induced the relocalization of Vpr to the cytoplasm , while these mutants were still able to induce G2 arrest [36] . However , careful re-examination of the localization of these mutants showed that both mutants could localize to the nucleus to some degree . While , the V57L mutant had a reduced capacity to form foci , the R62P mutant was completely defective for foci formation ( Figure S7A ) . The reduced capacity of V57L mutant and the defect of the R62P mutant in foci formation correlated , respectively , with attenuation and abrogation of G2 arrest ( Figure S7B ) . These differences between our present localization data and our previously published results can probably be explained by improved imaging sensitivity , whereas the discrepancies in G2 arrest activity are unclear . Nevertheless , these results highlight an important technical limitation in these types of localization experiments: lack of detection in a subcellular compartment does not necessarily indicate an absence of protein . Correlation between G2 arrest and formation of Vpr nuclear foci implies that the formation of these foci could either be an early event leading to G2 arrest or could be a consequence of this G2 arrest . We observed that treatment with the ATR/ATM inhibitor caffeine ( Figure 3A ) did not abrogate formation of Vpr foci , thus indicating that these foci likely constitute an early event in the induction of G2 arrest by Vpr . In fact , formation of Vpr foci was not affected by an almost complete knockdown of VPRBP suggesting that their formation is independent of the recruitment of the E3 ligase complex and would therefore precede ubiquitination and degradation of the putative G2 arrest substrate ( Figures 3B and S4 ) . In contrast , we found that the Q65R mutant of Vpr was unable to form foci . In addition to a reduced affinity for VPRBP [19] , [20] , [22] , this mutation also leads to other defects including accumulation of Vpr in the cytoplasm ( Figure 4 ) , reduced dimerization efficiency ( Figure 7 ) , and absence of binding to chromatin ( Figure 8B ) , indicating that the Q65R mutation has pleiotropic effects on the functions of Vpr . Yet , this mutation did not prevent efficient packaging of Vpr into virions [12] . Cautions should thus be used when interpreting results obtained with this mutant . Despite these pleiotropic defects , we cannot completely exclude the possibility that , in addition to the C-terminal domain , binding to VPRBP would also contribute to foci formation and chromatin association . Given that Vpr foci containing VPRBP partially co-localize with chromatin-bound protein such as RPA32 and that Vpr associates with DNA in vitro [51] and in vivo ( Figure 8A and [17] ) , we propose that Vpr might be able to target a chromatin-bound cellular factor . In support of this hypothesis , Lai et al . showed that in situ nuclease treatment of Vpr-expressing cells eliminates Vpr nuclear foci [17] , suggesting that Vpr nuclear foci are anchored to chromatin . Deletion of the C-terminal domain of Vpr drastically reduced foci formation ( Figure 4 ) and its chromatin association ( Figure 8B ) . Similar results were obtained by Lai and colleagues [17] . Moreover , mutation of the arginine at position 80 did not affect direct binding to nucleic acids in vitro [51] but nevertheless impaired association to chromatin in vivo ( Figure 8B ) , implying that a cellular factor rather than a direct binding to DNA would be implicated in association to chromatin . This cellular factor does not appear to be VPRBP since its knockdown did not significantly reduce the binding of Vpr to chromatin ( Figure 8C ) . Moreover , we also observed protein-protein interaction between Vpr and VPRBP on chromatin ( Figure 9 ) , suggesting that Vpr would be able to recruit the E3 ligase DDB1-CUL4A ( VPRBP ) onto chromatin . Strikingly , analysis of Vpr nuclear foci by time-lapse microscopy ( Figures 10 and S5 , Videos S1 and S2 ) revealed that these foci moved rapidly in the nucleus ( average of 0 . 73 µm/min ) . As a comparison , passive diffusion of chromatin-bound DNA repair foci was estimated at 1–2 µm per 6 hours [52] . These results suggest that instead of stably interacting with chromatin , Vpr nuclear foci would do so in a dynamic manner , allowing movement of the foci along chromatin strands . One possible model to integrate all our results is that Vpr could interact with its putative substrate via its C-terminus in these chromatin-bound nuclear foci and could recruit the DDB1-CUL4A ( VPRBP ) E3 ligase to degrade the substrate , thus preventing its potential role in DNA replication or DNA repair . This model implies that Vpr would initially require binding with the substrate to localize in these nuclear bodies and that the subsequent degradation of this substrate would not exclude Vpr from these structures nor would it disrupt them . Another possibility is that Vpr would interact with a nuclear foci-associated co-factor via its C-terminus and would utilize these mobile structures to scan chromatin for its putative substrate . This second model requires that either Vpr possesses an additional functional domain mediating the interaction with the substrate or that Vpr targets VPRBP's own natural substrates . Irrespective of the above models , as was recently documented , the substrate would be covalently modified with classical K48-linked polyubiquitin chains in a DDB1-CUL4A ( VRPBP ) -dependent manner and degraded by the proteasome [26] . Moreover , multiple units of the putative substrate/co-factor are probably required in these nuclear bodies in order for Vpr to accumulate in these structures . Even though Vpr multimerization was shown to occur in these foci ( Figure 7 ) , it is unlikely that it would play a major role in this process given that the L23F mutation was previously shown to block dimerization [40] , [53] but did not significantly affect foci formation and induction of G2 arrest ( Figure S5 ) . Similar conclusions were also previously obtained with the I70S mutation which was shown to block dimerization without affecting the induction of G2 arrest [30] . It however remains unclear whether Vpr would bind VPRBP before or after localizing to these foci , particularly when considering the important level of interaction observed in the Triton-soluble fraction ( Figure 9 ) . Moreover , the significance of the partial co-localization observed between Vpr and DNA repair foci containing RPA32 , 53BP1 and γ-H2AX ( Figures 2 and S4 ) is also unclear . On one hand it could mean that degradation of the chromatin-bound substrate would induce DNA damage or DNA replication stress in situ and that this partial co-localization would be explained by the high mobility of Vpr foci . On the other hand , we cannot exclude the possibility that degradation of the substrate could induce global genomic instability and that this partial co-localization would only be fortuitous . Overall , our results show that Vpr forms highly mobile nuclear foci containing VPRBP and demonstrate that formation of these structures constitutes a critical early event in the induction of DNA damage/stress and G2 arrest by Vpr . The characterization of these chromatin-bound nuclear foci hijacked by Vpr will likely contribute to better delineate the mechanism by which Vpr activates ATR and induces G2 arrest . Importantly , our results further suggest that the putative cellular substrate targeted by Vpr is likely to be a chromatin-associated protein .
Peripheral blood samples were obtained from adult donors who gave written informed consent under research protocols approved by the research ethics review board of the Institut de recherches cliniques de Montreal . HeLa and HEK293T cells were cultured as previously described [54] . Primary CD4+ T-lymphocytes were isolated and cultured as previously described [26] . The development of the HEK293T cell line stably depleted of VPRBP was described previously [26] . Caffeine and DAPI ( 4′ , 6-Diamidino-2-phenylindole ) were purchased from Sigma-Aldrich ( St . Louis , MO , USA ) . SiRNA targeting VPRBP ( siGenome SMARTpool M-021119-00 ) and scrambled control siRNA ( non-targeting siRNA #2 ) were obtained from Dharmacon ( Chicago , IL , USA ) . The anti-HA ( clone 12CA5 ) and anti-p24 ( catalog no . HB9725 ) monoclonal antibodies were produced from hybridomas obtained from the American Type Culture Collection ( Manassas , VA , USA ) . The monoclonal antibody against Vpr ( clone 8D1 ) was a kind gift of Dr Y . Ishizaka ( International Medical Center of Japan , Tokyo , Japan ) [55] . The following commercially available antibodies were used: mouse anti-nucleoporin ( Abcam , Cambridge , MA , USA ) , mouse anti-RPA70 ( Abcam ) , rabbit anti-53BP1 ( Abcam ) , rabbit anti-GAPDH ( Cell Signaling Technology , Danvers , MA , USA ) , rabbit anti-H3 antibodies ( Abcam ) rabbit anti-phospho RPA32 ( S4/S8 ) ( Bethyl Laboratories , Montgomery , TX , USA ) , rabbit anti-VPRBP ( Accurate Chemical and Scientific Corporation , Westbury , NY , USA ) , rabbit anti-actin ( Sigma-Aldrich , St . Louis , MO , USA ) , mouse anti-phosphoS139-H2AX ( clone JBW301 ) ( Upstate , Millipore , Billerica , MA , USA ) , mouse FITC-conjugated anti-p24 ( clone KC57 , Beckman Coulter Canada , Mississauga , Ontario , Canada ) , mouse anti-SC35 ( Sigma-Aldrich ) , and mouse anti-PML ( Santa Cruz Biotechnology , Santa Cruz , CA , USA ) . All fluorochrome-conjugated secondary antibodies were obtained from Molecular Probes ( Invitrogen , San Diego , CA , USA ) . SVCMV-Vpr ( WT ) , SVCMV-Vpr ( L23F ) , SVCMV-HA-Vpr ( WT ) , SVCMV-HA-Vpr ( V57L ) , SVCMV-HA-Vpr ( R62P ) , SVCMV-HA-Vpr ( Q65R ) , SVCMV-HA-Vpr ( H71R ) , SVCMV-HA-Vpr ( R80A ) , SVCMV-HA-Vpr ( S79A ) , SVCMV-HA-Vpr ( 1–86 ) , SVCMV-HA-Vpr ( 1–78 ) , and SVCMV-VSV-G were previously described or were constructed by PCR as previously described [19] , [32] , [46] . Plasmids pCDNA3 . 1_eYFP-MCS ( MB ) and pCDNA3 . 1_Rluc-MCS ( MB ) for the expression of eYFP and renilla luciferease ( Rluc ) N-terminal fusion proteins were kind gifts of M . Baril and D . Lamarre [56] . Wild type Vpr was amplified by PCR from SVCMV-HA-Vpr ( WT ) and subcloned into pCDNA3 . 1_eYFP-MCS ( MB ) and pCDNA3 . 1_Rluc-MCS ( MB ) to generate respectively pCDNA3 . 1-eYFP-Vpr ( WT ) and pCDNA3 . 1-Rluc-Vpr ( WT ) . Vpr ( R80A ) and Vpr ( Q65R ) were subcloned into pCDNA3 . 1_Rluc-MCS ( MB ) to generate pCDNA3 . 1-Rluc-Vpr ( R80A ) and pCDNA3 . 1-Rluc-Vpr ( Q65R ) using the same strategy . The lentiviral vector pWPI as well as the packaging plasmid psPAX2 expressing Gag-Pol , Tat and Rev were obtained from Dr . D . Trono ( School of Life Sciences , Swiss Institute of Technology , Lausanne , Switzerland ) . The lentiviral vector pWPI-HA-Vpr ( WT ) transducing HA-tagged Vpr and GFP was generated from the parental vector pWPI using a strategy described previously [19] . The plasmids expressing sooty mangabey HA-tagged Vpr and Vpx were obtained from S . Benichou ( Institut Cochin , Paris , France ) [4] . The infectious molecular clones HxBru ( Vpr- ) , HxBru ( HA-Vpr ) , and HxBru Vpr L23F , were described previously [26] , [32] , [57] . The HxBru VprWT LF/PS molecular clone with mutations ( L44P , F45S ) in the p6 domain of Gag disrupting interaction with Vpr was described previously [58] . The production and titration of VSV-G-pseudotyped HIV particles and lentiviral vectors were performed as described previously [19] , [46] . HeLa cells were transfected using the Lipofectamine 2000 reagent ( Invitrogen Canada , Burlington , Ontario , Canada ) according to the manufacturer's instructions . HEK293T cells were transfected by a standard calcium phosphate precipitation protocol . SiRNA were transfected using Lipofectamine RNAi Max ( Invitrogen Canada , Burlington , Ontario , Canada ) , according to the manufacturer's instructions . HeLa cells were transduced with the lentiviral vectors WPI and WPI-HA-Vpr in presence of 8µg/ml polybrene at a multiplicity of infection of 0 . 5 to 2 . 5 , as indicated for each experiment . Primary CD4+ T-lymphocytes were transduced by spinoculation at a multiplicity of infection of 1 . Briefly , cells were mixed with lentiviral vector particles in presence of 8µg/ml polybrene and centrifuged for 2 hours at 1200g . HeLa cells were infected , in presence of 8 µg/ml polybrene , with VSV-G-pseudotyped HIV-1 viruses at a concentration of 100 cpm/cell or at a MOI of 1 . 0 , as indicated for each experiment . Fifty thousand HeLa cells were seeded on cover slips in 24-well plates . Cells were transfected , transduced , or infected as indicated for each experiment . Two days later , cells were processed for fluorescence immunohistochemistry and laser-scanning confocal microscopy as previously described [59] . For analysis of CD4+ primary T-lymphocytes , 5×105 cells were first adhered on poly-Lysine-treated coverslips for two hours in PBS and then processed as described [59] . Quantification of Vpr nuclear foci was performed in at least 30 randomly selected cells by manual counting . Time-lapse confocal microscopy was performed on living cells in a PeCON environmental chamber maintained at 37°C and 5% CO2 . Images were acquired using a Zeiss LSM 710 system with the ZEN 2009 software . Spinning-disk confocal microscopy was performed on living cells using a Quorum WaveFX-X1 spinning-disc confocal system ( Quorum Technologies Inc , Guelph , Ontario , Canada ) . Cells were maintained at 37°C in 5% CO2 in a Live Cell Instruments Chamlide TC environmental chamber . Images were acquired with a Hamamatsu ImagEM C9100-13 camera using the Metamorph software . FRAP ( fluorescence recovery after photobleaching ) experiments were conducted using the Quorum WaveFX-X1 spinning-disc confocal system equipped with a Photonic Instruments Mosaic 405 nm laser . Images were processed using AxioVision v . 4 . 7 . Videos were generated with the ZEN 2009 software . Software-assisted fluorescence quantification and tracking of Vpr foci was performed with the Volocity software v . 5 . 2 . 1 . Statistical analysis was performed using Sigma Plot software v . 10 . In situ proximity ligation assays ( PLA ) were performed using the Duolink kit 613 ( Olink bioscience , Uppsala , Sweden ) . Briefly , HeLa cells were transfected with a plasmid encoding HA-Vpr or an empty plasmid as negative control . At 48h post-transfection , the cells were cytospun for 7 min at 1 , 100 rpm onto a glass slides and were fixed and permeabilized as described above . The fixed cells were incubated with the following antibodies: mouse monoclonal antibody against HA ( clone 12CA5 ) or Vpr ( a gift from Dr Y . Ishizaka . The antibody was shown to recognize both Vpr WT and Q65R [12] ) and a rabbit polyclonal antibody against VPRBP ( Accurate Chemical and Scientific Corporation ) . The Duolink system provides oligonucleotide-labeled secondary antibodies ( PLA probes ) to each of the primary antibodies that , in combination with a DNA amplification-based reporter system , generate a signal only when the two primary antibodies are in close proximity . The signal from each detected pair of primary antibodies was visualized as a spot ( please see the manufacturer's instructions for more details ) . Nuclei were delineated using Hoechst 33342 . Cell cycle analysis was performed using propidium iodide staining and flow cytometry as previously described [12] , [19] . Immunoprecipitations using anti-HA-conjugated agarose beads were performed as previously described [26] . Analysis of proteins by western blot was performed as previously described [26] . HEK293T cells were transfected in 24-well plates with 10ng of the BRET donor plasmids pCDNA3 . 1_Rluc-MCS ( MB ) , pCDNA3 . 1-Rluc-Vpr ( WT ) , pCDNA3 . 1-Rluc-Vpr ( R80A ) or pCDNA3 . 1-Rluc-Vpr ( Q65R ) and increasing concentration ( 0 to 500 ng ) of the BRET acceptor plasmids pCDNA3 . 1_eYFP-MCS ( MB ) or pCDNA3 . 1-eYFP-Vpr ( WT ) using Lipofectamine 2000 . Two days after transfection , cells were harvested , washed twice in PBS , and aliquoted in two wells of a 96-well plate ( Costar 3917 ) . Total eYFP fluorescence was measured with an excitation wavelength of 485 nm and an emission wavelength at 520±10 nm . BRET was initiated by adding 5µM of the renilla luciferase substrate coelenterazine H ( Prolume Ltd . , Lakeside , AZ , USA ) . Luminescence was then measured 10 minutes later at 475±15 nm and BRET fluorescence was measured at 535±15 nm . All measurements were performed on a PheraStar microplate reader ( BMG Labtech , Cary , NC , USA ) . BRET ratios were calculated using this formula: ( emission at 535 nm/emission at 475 nm ) - ( background emission at 535nm/background emission at 475 nm ) , as previously described [60] . Cells were lysed in triton lysis buffer ( 50 mM Tris pH 7 . 5 , 150 mM NaCl , 0 . 5% Triton X-100 , and complete protease inhibitors cocktail ( Roche ) for 15 minutes . Insoluble cell debris , including chromatin , was pelleted by centrifugation ( 2500g for 10 minutes ) . The supernatant was harvested and represented the soluble input control . Pellets were washed once with nuclease buffer ( 50 mM Tris pH 8 . 0 , 5 mM CaCl2 , and 100 µg/ml BSA ) , split in two , and resuspended in nuclease buffer alone or nuclease buffer containing 200 U/ml microccocal nuclease ( New England Biolabs , Ipswich , MA , USA ) . Pellets were incubated for 30 minutes on ice and then centrifuged at 12000g for 10 minutes . The supernatant was harvested and represented the chromatin-bound fraction . The corresponding supernatant obtained in absence of nuclease was used to control for non-specific release . For immunoprecipitation experiments , soluble and nuclease-treated fractions were incubated with 25 µl of anti-HA-conjugated agarose beads ( Sigma-Aldrich ) for 2h at 4C . In some experiments , immunoprecipitations were supplemented with 25 µg/ml ethidium bromide to displace proteins from DNA [47] . | HIV-1 , the causative agent of AIDS , encodes several proteins termed accessory , which play a critical role in viral pathogenesis . One of these accessory proteins , viral protein R ( Vpr ) , has been found to block normal cell division . This impairment of cell division by Vpr is thought to increase viral replication and to trigger immune cell death . However , how Vpr is able to block cell growth remains unknown . We and other investigators recently showed that Vpr was performing this activity by interacting with a cellular protein complex involved in ubiquitination . Ubiquitination is characterized by the conjugation of a small protein called ubiquitin to various other proteins to regulate their degradation or activities . In this report , we demonstrate that Vpr forms mobile punctuate structures called foci on the DNA of host cells . We also show that formation of these foci by Vpr is required to block cell division . We propose that Vpr recruits the ubiquitination complex to these nuclear foci and uses these mobile structures to target a DNA-bound cellular protein for degradation , resulting in the activation of a host cell response leading to a cell division block . Identification of the unknown cellular factor targeted by Vpr will contribute to the understanding of the role of Vpr during HIV infection and AIDS pathogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/virulence",
"factors",
"and",
"mechanisms",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses",
"cell",
"biology/cell",
"signaling",
"virology/immunodeficiency",
"viruses",
"biochemistry/replication",
"and",
"repair"
] | 2010 | Formation of Mobile Chromatin-Associated Nuclear Foci Containing HIV-1 Vpr and VPRBP Is Critical for the Induction of G2 Cell Cycle Arrest |
In networks of excitatory and inhibitory neurons with mutual synaptic coupling , specific drive to sub-ensembles of cells often leads to gamma-frequency ( 25–100 Hz ) oscillations . When the number of driven cells is too small , however , the synaptic interactions may not be strong or homogeneous enough to support the mechanism underlying the rhythm . Using a combination of computational simulation and mathematical analysis , we study the breakdown of gamma rhythms as the driven ensembles become too small , or the synaptic interactions become too weak and heterogeneous . Heterogeneities in drives or synaptic strengths play an important role in the breakdown of the rhythms; nonetheless , we find that the analysis of homogeneous networks yields insight into the breakdown of rhythms in heterogeneous networks . In particular , if parameter values are such that in a homogeneous network , it takes several gamma cycles to converge to synchrony , then in a similar , but realistically heterogeneous network , synchrony breaks down altogether . This leads to the surprising conclusion that in a network with realistic heterogeneity , gamma rhythms based on the interaction of excitatory and inhibitory cell populations must arise either rapidly , or not at all . For given synaptic strengths and heterogeneities , there is a ( soft ) lower bound on the possible number of cells in an ensemble oscillating at gamma frequency , based simply on the requirement that synaptic interactions between the two cell populations be strong enough . This observation suggests explanations for recent experimental results concerning the modulation of gamma oscillations in macaque primary visual cortex by varying spatial stimulus size or attention level , and for our own experimental results , reported here , concerning the optogenetic modulation of gamma oscillations in kainate-activated hippocampal slices . We make specific predictions about the behavior of pyramidal cells and fast-spiking interneurons in these experiments .
Mechanisms underlying the formation of gamma-frequency ( 25–100 Hz ) rhythms in networks of excitatory and inhibitory neurons ( E- and I-cells ) have been investigated extensively [1]–[11] . However , mechanisms underlying the loss of rhythmicity , as parameters change , have been given less attention . Here we consider the loss of gamma rhythmicity as the number of participating cells decreases . We focus on gamma rhythms resulting from the synaptic interaction of E- and I-cells , thinking of pyramidal cells and fast-spiking interneurons interacting via AMPA- and -receptor-mediated synapses . The E-cells spike intrinsically , driving and synchronizing the I-cells , which in turn gate and synchronize the E-cells . Rhythms of this kind are called PING ( Pyramidal-Interneuronal Network Gamma ) rhythms [10] , [11] . We distinguish between “strong PING” and “weak PING” . In strong PING , there is strong tonic ( i . e . , temporally constant ) drive to some or all E-cells , and those E-cells that participate at all typically participate on every population cycle . In weak PING , drive to the E-cells is stochastic , and typically each individual E-cell participates only on a fraction of population cycles [12] . We think of weak PING as a reduced model of the kainate-induced persistent gamma rhythm in slice [13]–[15] . Of course , real gamma oscillations might also be a mixture of “strong” and “weak” PING , with a stochastically fluctuating drive added to largely tonic baseline excitation . In most of the simulations of this paper , we omit any stochastic drive for simplicity , and assume that some or all E-cells receive strong constant drive . When the driven ensemble is large and synaptic interactions are strong , a strong PING rhythm will often arise in the driven ensemble [12] , [16] , [17] . We refer to an ensemble of cells of this sort , firing in synchrony at gamma frequency , as a “cell assembly” . The breakdown of gamma rhythms , as the number of tonically driven cells is decreased , is the combined effect of weak synaptic interactions and heterogeneity . Therefore the modeling part of the Results section of this paper begins with a study of how strong PING rhythms break down as synapses are weakened in heterogeneous E/I-networks of fixed size , assuming that all E-cells are driven . Using a combination of numerical simulations and mathematical analysis , we show that the breakdown of the rhythm can often be understood well by studying highly reduced , homogeneous networks . In a realistically heterogeneous network , the rhythm breaks down when the E-to-I-synapses become so weak that a single excitatory spike volley is no longer sufficient to prompt a response from the I-cells , or when the I-to-E-synapses become so weak that in a homogeneous network , convergence to tight synchrony would take several gamma periods . A surprising conclusion of our analysis is that a slow , graduate slide into PING , which is possible in a homogeneous network , is not possible in a realistically heterogeneous network; the PING rhythm is either established rapidly , or not at all . We then apply this analysis to understand how the rhythm breaks down as the number of driven cells is reduced . We conclude that for given synaptic strengths and heterogeneities , there is a ( soft ) lower bound on the possible size of cell assemblies . As explained in detail in the Discussion , our modeling results suggest possible theoretical explanations for several recent experimental findings: ( 1 ) Gieselmann et al . [18] , observed no gamma rhythm in primary visual cortex when the spatial extent of the driving stimulus was too small . ( 2 ) Chalk et al . [19] found that attention can weaken gamma oscillations in primary visual cortex . ( 3 ) In this paper , we report that in kainate-bathed hippocampal slices , strong optogenetic drive to the pyramidal cells elicits fast oscillations , whereas weak drive not only fails to elicit fast oscillations , but also abolishes the slower kainate-induced background gamma oscillations [14] , [15] . Our results lead to specific predictions concerning the behavior of pyramidal cells and fast-spiking interneurons in these experiments .
Lentivirus carrying the light-activated cation channel ChIEF [20] , an enhanced-performance version of channelrhodopsin-2 [21] , [22] , under the control of the CaMKII promoter was injected in the CA3 region of C57BL/6 mice . After 3–5 weeks , the animals were sacrificed , and 450 -thick horizontal hippocampal slices were cut . Bath application of 400 nM of the glutamatergic agonist kainic acid ( Cayman Chemicals ) induced 25–50 Hz oscillations in the local field potential ( LFP ) , as recorded in the CA3 stratum radiatum . Light pulses were delivered via a DG-4 optical switch with a 300 W xenon lamp ( Sutter Instruments ) and GFP filter set ( Chroma ) . For further details on the experimental methods , see Text S1 , ection A .
In kainate-activated mouse hippocampal slices , expressing the light-activated cation channel ChIEF under the control of the CaMKII promoter allowed us to selectively drive CA3 pyramidal cells , without optically affecting other cell classes present in the network . We explored the effects of modulating the light intensity , focusing on two conditions: “weak” ( 1–3 , 470 nm ) and “strong” ( 10–30 , 470 nm ) stimulation . Thus there were two sources of excitation in these experiments . The first was the kainic acid in the bath , which induced a persistent gamma ( 25–50 Hz ) oscillation in the LFP as reported by others , e . g . , [14] , [15] . We model the effect of the kainic acid as stochastic drive to the E-cells – see Methods and Text S1 , Section B , as well as refs . [12] and [16] . The second source of excitation was optogenetic drive . We model it as tonic drive . Weak optogenetic drive reduced the power of ongoing oscillations ( Fig . 1 ) . Energy in the low gamma frequency band during weak stimulation was of the pre-stimulation baseline ( 94 trials , 8 slices , ) , with no significant change in peak frequency ( ) . It is important to note that the stimulation did not simply result in a shift to a faster oscillation frequency , as the energy in all higher frequency bands was also decreased . In contrast , strong optogenetic drive resulted in the emergence of fast oscillations with a peak frequency of 86 . 8 Hz and a standard deviation of 18 . 4 Hz ( 40 trials , 4 slices ) , significantly different from the baseline oscillation frequency of Hz ( 60 trials , 5 slices ) ( ) . The fast oscillations appeared to replace , rather than superimpose onto , the baseline rhythm , as energy in the 25–50 Hz frequency band was of baseline ( ) . As has been measured by others ( e . g . , Huber et al . [24 , Fig . 1G] ) , the intrinsic variation in the amount of channelrhodopsin expressed in one cell vs . another means that stronger light will elicit spiking in more cells than weaker light . We therefore use computational simulation and mathematical analysis to study how the number of driven cells governs whether or not a PING rhythm forms in model networks . The total number of active cells determines the strength of synaptic input per target cell . A more fundamental question is , therefore , how weakening synaptic connections leads to the disintegration of PING rhythms . We study this question first , then apply the conclusions to understanding how gamma rhythms break down when reducing the number of tonically driven E-cells , or reducing , in a spatially structured network , the size of the region in which the E-cells are driven , or making synaptic connectivity more local . As one gradually weakens the E-to-I-synapses in an E/I-network , there comes a point at which the PING mechanism fails . Where the breakdown occurs depends on network heterogeneity to some degree , but we show below that it can be predicted , with good accuracy , by studying homogeneous networks: In a homogeneous network , weakening of E-to-I-synapses leads to a sudden switch from 1∶1 entrainment of I-cells by E-cells to more complicated patterns , such as 2∶1 entrainment . The synaptic strength at which this happens approximately equals the synaptic strength at which rhythmicity breaks down , somewhat more gradually , in a heterogeneous network . We will also show that for significantly stronger E-to-I-synapses , even fairly substantial heterogeneities in synaptic connections and external drives to the I-cells do not have strong effects on I-cell synchrony . Weakening of the I-to-E-synapses eventually results in breakdown of the PING mechanism . As for the E-to-I-synapses , the point of breakdown can be predicted well by studying homogeneous networks . However , here the breakdown of the rhythm in the heterogeneous network is not signaled by downright breakdown in the homogeneous network , but by lengthening of the time needed to establish the rhythm in the homogeneous network when starting from asynchronous initial conditions . To demonstrate and explain this point , we will first give a computational analysis of the response of a single cell to an inhibitory pulse , then present results of network simulations . Finally , we will give a detailed analysis for a simplified problem , elucidating the reason for the link between slow convergence to synchrony in a homogeneous network and failure to synchronize at all in a heterogeneous network . We now connect the previous results to the size of cell assemblies . We start with a spatially unstructured network that is sparsely and randomly connected , and consider the behavior as the number of tonically driven E-cells is reduced . Our expectation for what should happen comes from our earlier discussion of weakening E-to-I-coupling; reducing the number of active E-cells reduces the total excitatory current received by I-cells . Thus we expect the rhythm to be destroyed if the number of participating E-cells gets too small . Fig . 13 shows the breakdown of the rhythm . In contrast with other simulations in this paper , in the simulation of Fig . 13 , all E-cells receive a background of stochastic input; see Text S1 , Section B for details . As the number of E-cells receiving tonic drive is reduced , there is a transition from PING ( A and B ) , through an arrhythmic regime in which both E- and I-cells continue spiking ( C ) , to a background weak PING rhythm [12] driven by the stochastic input to the E-cells , akin to a kainate-induced persistent gamma rhythm in a hippocampal slice . We hypothesize that panel D of Fig . 13 is an analogue of what happens in Fig . 1 before and after the stimulus . As discussed earlier , we think of driving fewer E-cells as the analogue of using weaker optogenetic stimulation; therefore panel C is an analogue of what happens in Fig . 1 during the stimulus . We show the time interval from 200 ms to 400 ms , not the initial time interval from 0 ms to 200 ms , in Fig . 13 in order to demonstrate that in panel C , the rhythm is not just slow to arise , but does not arise at all . The mean frequency of the I-cells is approximately 54 Hz in panel A , 48 Hz in panel B , and 29 Hz in panel C . Panel D of Fig . 13 shows a simulation in which no E-cells receive tonic drive; the results would look very similar if , for instance , 5 E-cells received tonic drive . When a significant number of the E-cells receive specific drive ( panels A–C ) , the weak PING rhythm in the other E-cells is suppressed either by fast rhythmic activity of the I-cells ( panels A and B ) , or by sparser , but arrhythmic and therefore still powerful [2] activity of the I-cells ( panel C ) . With fewer E-cells driven tonically , the I-cells receive less synaptic excitatory input; therein lies the main connection between the simulation results in Fig . 13 and the earlier ones concerning the effect of weakening E-to-I-synapses ( Fig . 3 ) . However , effective heterogeneity in excitatory synaptic input per I-cell also becomes greater when the input originates from a smaller number of E-cells , for statistical reasons: When the excitatory synaptic input into the I-cells originates primarily from a small number of tonically driven E-cells , the heterogeneity resulting from the randomness of the connectivity is substantial . This effect is reduced by the Law of Large Numbers as the number of tonically driven E-cells increases . In Fig . 13C , the I-cells inhibit each other . However , the I-to-I-interactions alone are not sufficient to create synchrony , i . e . , there is no ING [11] rhythm , because of the heterogeneity of the drive from the E-cells [29] . Our reasoning suggests that in Fig . 13C , the rhythm should be restored if the E-to-I-synapses are strengthened: The loss of rhythmicity occurs simply because E-to-I-connectivity is so weak that more than 50 E-cells are required to sustain the rhythm . Indeed , if the strength of excitatory synapses is tripled in Fig . 13C , the rhythm does return; see Fig . 13E . We now consider a network in which the connection probability decays with increasing distance between neurons ( see Text S1 , Section B ) . Fig . 14 shows the placement of neurons in the unit disk . ( Distance is non-dimensionalized here . ) Strong drive is given to the E-cells in a smaller circle of radius , also indicated in Fig . 14 . As is reduced , the rhythm eventually breaks down , as shown in Fig . 15 . In plotting the spike rastergrams , each of the two cell populations was ordered in such a way that smaller indices correspond to positions closer to the center of the disk . As the size of the driven patch gets smaller , the I-cells receive less excitatory input . This is one way in which the results of Fig . 15 are related to the earlier ones on weakening E-to-I-synapses ( Fig . 3 ) and reducing the number of cells driven ( Fig . 13 ) . In addition , however , as the driven patch decreases in size , fewer I-cells receive enough excitatory synaptic input to participate in the rhythm , since the probability of synaptic interaction decays with distance here . Thus , in fact , the loss of rhythmicity in Fig . 15 is also related to a reduction in the total amount of inhibitory input per E-cell; compare Fig . 7 . As in the earlier Figs . 3 , 7 , and 13 , the loss of rhythmicity in Fig . 15 results from a combination of weak effective synaptic interactions and heterogeneity . There are two sources of effective heterogeneity here . The first is statistical: When fewer E-cells give synaptic input to the I-cells ( or vice versa ) , there are more statistical fluctuations , because the Law of Large Numbers does not wash them out . If we increased the size of the model network , while proportionally reducing the strength of each individual synapse and leaving all other parameters fixed , the statistical heterogeneity would be reduced . However , here there is a second , geometric source of heterogeneity: Cells that lie near the edge of the driven patch receive fewer synaptic inputs than ones that are far from the edge . For a smaller patch , the percentage of cells significantly affected by this effect is greater . If we increased the size of the model network , while proportionally reducing the strength of each individual synapse and leaving all other parameters fixed , the geometric heterogeneity would not disappear . Fig . 16 shows the same numerical experiment as Fig . 15 , but with twice as many E- and I-cells , and with the strengths of individual synapses halved . Inspection of the rastergrams suggests that synchrony is sharper in Fig . 16 than in Fig . 15 . This is confirmed by Fig . 17 , which shows the synchrony measure ( see Text S1 , Section B ) as a function of the radius of the driven patch of E-cells . The enhanced synchrony for the larger network is likely due to the reduction in statistical fluctuations in the stochastic input per cell as the network is enlarged . Fig . 17 also shows that for both networks , there is a gradual but marked deterioration of gamma power as is reduced , setting in at approximately the same value of . We consider the same kind of spatially structured networks as before . However , we now fix the radius of the circular patch in which the E-cells receive specific drive , and vary the length constant characterizing the decay of the connection probability with distance ( see Text S1 , Eq . S12 ) . The effect of a decrease in is a reduction in total synaptic input per cell , and therefore eventually the breakdown of the PING rhythm , for the reasons discussed earlier; see Fig . 18 , panels A and B . Panel C of Fig . 18 shows the result of making connectivity more local , as in panel B of the figure , but compensating by strengthening the synapses: The rhythm returns . Thus panel C illustrates that the rhythm breaks down in panel B because there is too little overall synaptic input per cell , not because connectivity is too local per se .
We have examined how the PING mechanism breaks down as excitatory and/or inhibitory synapses are weakened , as the numbers of participating excitatory and/or inhibitory cells get too low , or as synaptic connectivity becomes too local . Although the breakdown of the rhythm results from the interaction of weakness of synaptic inputs with network heterogeneity , the point at which it occurs can be predicted , with good accuracy , by studying homogeneous networks . The effects of heterogeneity in synaptic or external drive to the I-cells disappear rapidly as the strength of excitatory synapses increases , and is substantial only if the excitatory synapses are marginal , i . e . , nearly so weak that in a homogeneous network , 1∶1-entrainment of E- and I-cells would be replaced by a more complicated pattern such as 2∶1-entrainment . In contrast , the effects of heterogeneity in synaptic or external inputs to the E-cells remain sizable even in the limit as the strength of the inhibitory synaptic input per E-cell tends to infinity . They increase greatly , and quickly lead to a complete breakdown of the rhythm in a heterogeneous network , when inhibitory synaptic inputs into the E-cells become so weak that , in a homogeneous network , the rhythm would only be established after multiple gamma cycles . In a realistically heterogeneous network , a PING rhythm is either established rapidly , within a small number of gamma cycles , or not at all . As the number of driven cells is reduced , the synaptic input per cell is reduced as well . At the same time , the effective heterogeneity in synaptic connections becomes more significant . The combination of effectively weaker synaptic interactions with greater heterogeneity eventually leads to the breakdown of the rhythm . There are two reasons why a reduction in the number of driven cells can lead , in effect , to more significant heterogeneity . First , when connectivity is sparse and random , different cells receive different numbers of synaptic inputs . When many cells participate in the rhythm , this effect is largely erased by the Law of Large numbers . However , when only a small number of cells participate , it can be substantial . Second , when only the neurons in a certain spatial domain are driven strongly , and when the probability of synaptic connections decreases with increasing spatial separation , those neurons near the edge of the spatial domain receive less synaptic input than those near the center . As the size of the driven domain is reduced , the fraction of cells that are close enough to the edge for this effect to matter increases . Our results imply that for given synaptic strengths , cell assemblies cannot be arbitrarily small . This is complementary to recent work by Oswald et al . [30] , who have pointed out a reason why there may be an upper bound on the possible size of cell assemblies . The lower bound on the size of cell assemblies substantially depends on the strength of E-to-I- and I-to-E-coupling; stronger synapses allow smaller assemblies . To some extent , it also depends on network heterogeneity , with less heterogeneity allowing smaller assemblies . It would be very interesting to give a specific , numerical answer to the question “How many cells must an assembly include to oscillate at gamma frequency ? ” Unfortunately , such an estimate would have to remain highly speculative at this point . We would , for instance , have to know how many synchronous excitatory synaptic inputs a fast-spiking parvalbumin-positive interneuron must receive for a spike to be elicited , what are the density and spatial reach of synaptic connections from pyramidal cells to fast-spiking interneurons and vice versa , and how many inhibitory synaptic inputs are required to create the synchronizing “river” in the pyramidal cells . None of these data are known with any degree of certainty , and they are surely different in different parts of the brain . These uncertainties render any attempt to give a numerical estimate futile . In those of our simulations that involved space-dependent connectivity probabilities , we assumed that excitation and inhibition reached equally far . While this assumption is in agreement with what Adesnik and Scanziani [31] have found in their recent work for horizontal connectivity in layer 2/3 of mouse somatosensory cortex , it is not crucial for the results of the present paper . In this study , we have focused on strong PING , i . e . , PING driven by tonic excitation of the E-cells . ( The only exception is Fig . 13 , where we added stochastic drive to the E-cells in order to demonstrate that the weak PING rhythm created by such drive is abolished when a small , but not extremely small number of E-cells receive tonic drive . ) However , we do not see a reason why similar results should not hold for stochastically driven PING rhythms . Spencer [32] has previously studied the effects of weakening synaptic connections in E/I networks of integrate-and-fire neurons , and found rapid deterioration of gamma rhythms with the weakening of fast synaptic connections . The loss of gamma rhythmicity in [32] appears to be faster than in our numerical experiments; compare for instance [32 , Fig . 3] with Fig . 4 of this paper . This is a quantitative ( not qualitative ) difference that may be accounted for by several differences in details between the model of [32] and ours . For instance , external drive in [32] fluctuates stochastically in time , whereas in our model the primary drive is tonic , but often with significant heterogeneity . We note that Fig . 1 of [32] does appear to show a gradual slide into PING , contrary to our conclusion that in a heterogeneous network , PING is formed either rapidly , or not at all . However , note that the rhythm in [32 , Fig . 1] relies on recurrent , NMDA-receptor-mediated excitation among the E-cells . This excitation has to build up before the rhythm begins . Here we have only considered rhythms sustained by external drive to the E-cells . There is a considerable body of work on the stability of asynchronous states in neuronal networks [33]–[36] . Our focus here has not been on this bifurcation , in which oscillations are created , but rather on the gradual tightening of synchronization , and resulting rise in oscillation amplitude , as synaptic strengths are increased beyond the level necessary for the asynchronous state to loose its stability . ( Of course , in a heterogeneous E/I network , synchrony will never get entirely tight – there is no strictly synchronous state in such a network . ) There is also previous literature on the dependence of synchronization on connectivity , e . g . , [37] , [38] . Our focus here is different , however; we point out that making synaptic connectivity more local , without compensating by strengthening individual synapses , can lead to a loss of rhythmicity simply because synaptic interactions get too weak , before they get too local ( independently of their strength ) ; see Fig . 18 . Bartos et al . [39] have suggested that the inhibitory synapses relevant to gamma oscillations should be faster and less hyperpolarizing than the ones used here . There is controversy over which are the biologically realistic parameter choices . For example , Cobb et al . [40] reported -receptor-mediated inhibition in hippocampus to be hyperpolarizing , not shunting . Traub et al . [6] reported IPSCs with decay time constants around 10 ms during ongoing gamma oscillations . Gamma rhythms are possible in our model networks with the parameters used by Bartos et al . , but many of their properties , including the mechanisms by which they are lost when synaptic interactions are weakened , are different . The fact that the PING rhythm eventually breaks down as synaptic interactions become weaker is , of course , obvious , but it suggests explanations for several seemingly unrelated recent experiments . First , in our own experiments in kainate-bathed slices of area CA3 of mouse hippocampus , strong light activation of pyramidal cells elicits a strong , fast rhythm . Weak light activation not only fails to elicit such a rhythm , but also abolishes the kainate-induced persistent gamma rhythm . We hypothesize that weaker light activation in these experiments results in activation of a smaller number of cells . These results are then closely analogous to our Fig . 13 , where we examined the loss of the gamma rhythm as the number of E-cells with specific drive becomes too small . Second , in macaque primary visual cortex , gamma rhythms are elicited by spatially extended stimuli , but not by ones that are too small [18 , Fig . 1]; this is expected from our Fig . 15 . We note that gamma rhythms have been associated with binding [41]–[43] , and may therefore not be needed for highly spatially focused neuronal activity . Third , attention has been found to reduce gamma power in primary visual cortex [19] . Our Fig . 18 , which shows the loss of gamma rhythms as synaptic connections become too local , offers a possible theoretical explanation of this result , since the cholinergic modulation associated with attentional processing is thought to reduce the efficacy of lateral synaptic connections [44] . Our simulation results predict that in all of these experiments , when the rhythm disappears , vigorous activity should continue in the driven E-cells and in nearby I-cells . In summary , the strengths of the reciprocal synaptic interactions between principal cells and those inhibitory interneurons participating in the gamma rhythm play a crucial role in determining the possible sizes of cell ensembles oscillating at gamma frequency , with stronger synapses allowing smaller ensembles . | Gamma-frequency ( 25–100 Hz ) oscillations in the brain often arise as a result of an interaction between excitatory and inhibitory cell populations . For this mechanism to work , the interaction must be sufficiently strong , and connectivity and external drives to participating neurons must be sufficiently homogeneous . As the interactions become weaker , either because the neuronal ensembles become smaller or because synapses weaken , the rhythms deteriorate , and eventually break down . This fact , by itself , is not surprising , but details of how the breakdown occurs are subtle . In particular , our analysis leads to the conclusion that in realistically heterogeneous networks , gamma rhythms must arise quickly , within a small number of oscillation periods , if they arise at all . Our findings suggest explanations for recent experimental findings concerning the minimal spatial extent of stimuli eliciting gamma oscillations in the primary visual cortex , the modulation of gamma oscillations in the primary visual cortex by attention , as well as our own experimental results , reported here , concerning the minimal light intensity below which optogenetic drive to pyramidal cells in a kainate-activated hippocampal slice results in disruption of an ongoing gamma oscillation . Our analysis leads to experimentally testable predictions about the behavior of the excitatory and inhibitory cells in these experiments . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biology",
"neuroscience"
] | 2012 | Minimal Size of Cell Assemblies Coordinated by Gamma Oscillations |
Cerebrospinal fluid flow is crucial for neurodevelopment and homeostasis of the ventricular system of the brain , with localized flow being established by the polarized beating of the ependymal cell ( EC ) cilia . Here , we report a homozygous one base-pair deletion , c . 1193delT ( p . Leu398Glnfs*2 ) , in the Kinesin Family Member 6 ( KIF6 ) gene in a child displaying neurodevelopmental defects and intellectual disability . To test the pathogenicity of this novel human KIF6 mutation we engineered an analogous C-terminal truncating mutation in mouse . These mutant mice display severe , postnatal-onset hydrocephalus . We generated a Kif6-LacZ transgenic mouse strain and report expression specifically and uniquely within the ependymal cells ( ECs ) of the brain , without labeling other multiciliated mouse tissues . Analysis of Kif6 mutant mice with scanning electron microscopy ( SEM ) and immunofluorescence ( IF ) revealed specific defects in the formation of EC cilia , without obvious effect of cilia of other multiciliated tissues . Dilation of the ventricular system and defects in the formation of EC cilia were also observed in adult kif6 mutant zebrafish . Finally , we report Kif6-GFP localization at the axoneme and basal bodies of multi-ciliated cells ( MCCs ) of the mucociliary Xenopus epidermis . Overall , this work describes the first clinically-defined KIF6 homozygous null mutation in human and defines KIF6 as a conserved mediator of neurological development with a specific role for EC ciliogenesis in vertebrates .
The delicate balance of cerebrospinal fluid ( CSF ) production and flow is important for the morphogenesis and function of the brain during development and homeostasis . CSF circulation in human is largely due to gradients established by the secretion of CSF from the choroid plexuses , and its resorption at the arachnoid granulations [1] . The clinical significance of CSF stasis includes hydrocephalus and intracranial hypertension . Moreover , severely diminished CSF flow combined with increased intracranial pressure can secondarily cause ventriculomegaly , cognitive impairment , as well as degenerative and age-related dementias [2] . For these reasons , the identification of genetic risk factors involved in the pathogenesis of CSF stasis is critical for the development of genetic diagnostics and early interventions for these disorders . One element for circulation of CSF is the multiciliated ependymal cells ( ECs ) , which are specialized glial cells covering the ventricular walls of the brain and spinal canal [3] . In contrast , to primary cilia which are single , immotile cellular organelles extending from most cell types , ECs contain dozens of apically-arranged motile cilia , which beat in a polarized fashion to generate localized or near-wall CSF flows [4] . Defective differentiation or alterations in their stereotyped synchronous , polarized beating leads to alterations of localized CSF flow contributing to increased intracranial pressure , dilation of ventricles , and hydrocephalus in mice [5–8] . Importantly , this EC cilia-driven CSF flow is vital for regulating brain function and neurogenesis during adult development [4 , 9] . Impaired motility of cilia due to disruptions of the key kinesins , dyenins , or intraflagellar components , are associated with a syndromic condition known as primary ciliary dyskinesia ( PCD ) in humans [10 , 11] . While hydrocephalus can occur in some PCD patients , it is a less common manifestation of the disease in humans [11] . In contrast , genes implicated in PCD or mutations which disrupt the structure or motility of all motile cilia are strongly correlated with hydrocephalus in mouse [8] . Alternatively , some hydrocephalus in mice with dysfunctional cilia may be the result of altered function of the choroid plexus , prior to the onset of cilia-driven CSF flow [7] . KIF6 ( Kinesin family member 6 , OMIM: 613919 ) encodes a member of the kinesin-9 superfamily of microtubules motor proteins which act predominately as "plus-end" directed molecular motors that generate force and movement across microtubules [12] . Kinesins are critical for numerous cellular functions such as intracellular transport and cell division , as well as for building and maintaining the cilium in a process known as intraflagellar transport [13] . During this process , kinesins have been shown to transport cargo within the ciliary axoneme [14] , establish motility and compartmentalization of the axoneme [15] , or to facilitate plus-end directed microtubule disassembly and control of axonemal length [16] . As such , multiple kinesins have shown to be associated with monogenic disorders affecting a wide-spectrum of tissues , with several modes of inheritance ( www . omim . org ) . Interestingly , KIF6 has previously been proposed as locus for susceptibility to coronary heart disease [17] , while other studies did not substantiate this association [18] . We previously reported that kif6 mutant zebrafish are adult viable exhibiting larval-onset scoliosis without obvious heart defects [19] . Because of these conflicting results , and a lack of relevant mouse models , the role of KIF6 in human disease remains an open question . Here , we present a patient with consanguineous parents , presenting with abnormal neurological morphologies and intellectual disability . Homozygosity mapping followed by whole-exome sequencing ( WES ) identified a novel homozygous frameshift mutation in KIF6 which is predicted to result in the truncation of the C-terminal cargo-binding domain of the kinesin motor protein . We generated an analogous frameshift mutation in mouse and found that these mutant mice displayed progressive , postnatal-onset hydrocephalus with cranial expansion , coupled with an obvious defect in EC cilia formation . In addition , we observed that kif6 mutant zebrafish also display dilation of the ventricular system , coupled with reduced EC cilia . We failed to observe cilia defects in other multiciliated tissues in Kif6 mutant mouse or zebrafish models . Together these results demonstrate that KIF6 has a specific function for the formation of EC cilia . Finally , we propose that KIF6 represents a novel locus for understanding mechanisms of neurological development and intellectual disability in humans .
We identified a Thai boy with intellectual disability and megalencephaly . His parents were first cousins . He was born at 34 weeks gestation with a head circumference of 34 cm ( 97th centile ) . APGAR scores were 7 and 9 at 1 and 5 minutes , respectively . Neonatal hypoglycemia ( blood sugar of 11 mg/dL ) and neonatal jaundice were treated promptly . In the first few months of life , he was found to have delayed neurodevelopment and central hypotonia . He was able to hold his head at 5 months , rolled over at 8 months , walked and had first words at 2 years old . At the age of 9 years and 9 months , an IQ test by Wechsler Intelligence Scale for Children: 4th edition ( WISC-IV ) revealed that his full-scale IQ was 56 , indicating intellectual disability . The patient had possible seizure activity at age 10 described as parasomnias , was found to have intermittent bifrontocentreal rhythmic theta activity , and the spells resolved after valproic acid therapy . His height and weight followed the curve of 50th centile , but his head circumference remained at 97th centile ( 53 . 5 cm and 55 cm at 6 and 9 years old , respectively ) . Physical examination was generally unremarkable except macrocephaly and low-set prominent anti-helical pinnae ( Fig 1A ) . Eye examination , hearing tests , thyroid function tests , chromosomal analysis , and nerve conduction velocity were normal . Both brain CT scans at 4 months and 8 years old and brain MRI at 7 months old showed a slight dolichocephalic cranial shape ( cephalic index = 75 ) , without overt structural brain abnormalities ( Fig 1B–1D ) . X-ray analysis of the spine showed no obvious scoliosis at 10-years-old ( Fig 1E ) . To elucidate the genetic etiology , we performed homozygosity mapping , whole genome array comparative genomic hybridization ( CGH ) , and whole exome sequencing ( WES ) . We identified 83 homozygous variants , which had not been reported as SNPs in dbSNP137 ( S1 Table ) . We then selected only those located within the 63 homozygous regions found by homozygosity mapping ( S2 Table ) . Seven candidate variants ( one frameshift and six missense mutations; Table 1 ) were identified . Of the six missense , five were predicted to be either benign by Polyphen-2 or tolerated by SIFT prediction programs . The remaining variant , c . 235G>A; p . V79M of the Carboxypeptidase E ( CPE ) gene , was not evolutionarily conserved among diverged species ( S1 Fig ) . We , therefore , decided to further our study on the only candidate truncating mutation , a homozygous one base-pair deletion , c . 1193delT ( p . Leu398Glnfs*2 ) in exon 11 of Kinesin family member 6 ( KIF6 ) ( NM_001289021 . 2 ) . KIF6 is located on human chromosome 6p21 . 2 and comprises 23 exons . The 2 . 4-kb KIF6 cDNA encodes a canonical N-terminal kinesin motor domain ( amino acid positions 3–353 ) and three coiled-coil regions ( amino acid positions 358–385 , 457–493 , and 633–683 ) , predicted by SMART server [20] . Segregation of the homozygous sequence variant with the disease phenotype was confirmed by Sanger sequencing ( Fig 1F ) and by restriction fragment length polymorphism ( RFLP ) analysis of the pedigree ( Fig 1G ) , while his parents and his unaffected brother were heterozygous for the deletion ( Fig 1F and 1G ) . The deletion was not observed in our 1 , 600 in-house Thai exomes , the 1000 Genome Database , and the ExAC Database . The pedigree combined with the novelty of the mutation in KIF6 presented here , strongly suggest this C-terminal truncating mutation in KIF6 may be etiologic for neurological developmental defects . To test the functional consequence of the C-terminal truncating p . L398fsX2 mutation ( Fig 1H ) , we generated an analogous frameshift mutation in exon14 of the mouse Kif6 ( ENSMUST00000162854 ) gene , which is ~150bp downstream of the frameshift mutation found in the patient ( Fig 2A ) . After backcross of founder mice to the C57B6/J strain , we identified a nonsense allele with scarless insertion ( c . 1665ins ) of a 3-stop donor cassette -providing integration of an ochre termination codon in all three reading frames into the endogenous Kif6 locus ( S2 Fig ) . This endonuclease-mediated insertional frameshift mutation ( Kif6em1Rgray ) is predicted to truncate the C-terminal cargo-binding domain of the kinesin motor protein ( p . G555+6fs ) . This novel mutant allele of Kif6 ( hereafter called Kif6p . G555fs ) is predicted to encode a C-terminal truncated KIF6 protein 168 amino acids longer than is predicted for the human p . L398fsX2 variant ( Fig 2A ) . Real time qualitative-PCR analysis of several Kif6 exon-exon boundaries found no evidence for non-sense mediated decay in Kif6p . G555fs mutant mice ( Fig 2B ) . Intercrossing Kif6p . G555fs/+ heterozygous animals gave offspring with the expected Mendelian ratios , with typical appearance at birth . However , beginning at postnatal day ( P ) 14-onwards , 100% ( n = 7 ) of Kif6p . G555fs homozygous mutant mice displayed classic indications of hydrocephalus including doming of the cranium ( Fig 2C ) , a hunched appearance , and with decreased open field activity . We observed apparent megalencephaly and hemorrhaging in older ( P21-P28 ) Kif6p . G555fs mutant brains ( Fig 2D ) , which likely results from increased intracranial pressure and swelling of the ventricles causing damage to the neural tissue against the cranium . At P14 , the body weights were not significantly decreased in Kif6p . G555fs mutants ( 5 . 8±1 . 3 ( g ) rams ) compared with littermate controls ( 7 . 0±1 . 2g ) ( n = 5/genotype; p = 0 . 17 ) . However , at P28 mutant mice showed decreased weight on average ( 12 . 67±1 . 53 g ) compared to littermate controls ( 15 . 33±1 . 15g ) , although this trend was not statistically significant ( n = 3/genotype; p = 0 . 07 ) . At P28 , extracted whole brain sizes appear to be larger in Kif6p . G555fs mutants compared to non-mutant littermate controls ( Fig 2D ) . Due to increased morbidity in these mutant animals no experimental observations were made past P28 . To determine whether a more N-terminal truncated Kif6 mutation would result in a more severe hydrocephalus phenotype , we isolated a conditional-ready Kif6 allele , where exon 4 is flanked by LoxP sites ( Kif6tm1c ) ( KOMP repository , see Methods and Materials ) . Recombination of the Kif6tm1c allele is predicted to generate a frameshift mutation , which should generate a severely truncated , 89 amino acid , KIF6 protein ( p . G83E+6fs ) with a non-functional N-terminal motor domain . We generated a whole body conditional knockout by crossing the Kif6tm1c mouse to the CMV-Cre deleter mouse [21] . We observed postnatal-onset , hydrocephalus in CMV-Cre; Kif6tm1c/tm1c conditional mutant mice ( n = 12 ) analogous to our observations in Kif6p . G555fs mutant mice ( S3 Fig ) . Interestingly , we find no evidence of non-sense mediated decay in these mutant mice despite the generation of an early premature termination codon ( Fig 2B ) . Because the onset and progression of hydrocephalus was equivalent comparing the whole-body conditional CMV-Cre;Kif6tm1c/tm1c and Kif6p . G555fs mutant mice strains we suggest that any KIF6 protein encoded by these mutant mouse strains is likely non-functional . Given its relevance to the human mutation , the majority of experiments were all done using the p . G555fs allele . Mouse brains were analyzed histologically by hematoxylin and eosin ( H&E ) stained coronal sections . Our analysis of coronal sectioned brain at P14 failed to find significance when comparing the total area in section ( 499 . 2+39 . 9μm ( Control ) vs . 552 . 5+50 . 8μm ( Kif6p . G555fs ) ; n = 7/genotype; p = 0 . 42 ) . However , lateral and third ventricles ( LV and 3V respectively ) were obviously enlarged in both Kif6p . G555fs and CMV-Cre;Kif6tm1c/tm1c mutant mice ( Fig 2F and S3 Fig ) . Quantitation of LV area normalized to total brain area confirmed ventricular expansion in Kif6p . G555fs ( n = 7/genotype; p≤0 . 05; Fig 2G ) and in CMV-Cre;Kif6tm1c/tm1c mutant animals ( n = 5/genotype; p≤0 . 05; S3 Fig ) . No obvious defects of the cortex or dysmorphology of other regions of the brain were apparent in these mice ( Fig 2E and 2F and S3 Fig ) . Together these data suggest that Kif6 mutant mice display postnatal-onset , progressive hydrocephalus , without obvious overgrowth of neural cortex . To determine the endogenous expression patterns of Kif6 in the mouse , we also isolated a Kif6-LacZ reporter mouse ( Kif6-LacZtm1b ) ( KOMP repository , see Methods and Materials ) . Hemizygous Kif6-LacZtm1b/+ mice appeared unremarkable and exhibited no evidence of hydrocephalus . Intercrosses of Kif6-LacZtm1b/+ hemizygous mice failed to generate litters with Kif6-LacZtm1b/tm1b homozygous mice , suggesting that the homozygosity of the lacZ expressing allele is embryonic lethal ( 0/13 Kif6-LacZtm1b/tm1b homozygous mutant mice from 3 independent litters ) . At P10 and P21 , Kif6-LacZtm1b/+ transgenic mice showed lacZ expression in the ECs of the ventricular system ( red arrows; Fig 3A” ) and the central canal ( S4 Fig ) . However , no lacZ expression was detected in the choroid plexus or in other regions of the brain ( Fig 3A and 3A’ and S4 Fig ) , with the exception of a small population of cells , possibly the hypothalamic nuclei , flanking the third ventricle ( arrows , S4 Fig ) . Interestingly at P10 , other multi-ciliated tissues such as the oviduct or trachea were not labeled in these transgenic mice , despite the clear presence of cilia observed by oblique lighting ( red arrows Fig 3B’ and 3C’ ) as well as by IF using acetylated-tubulin to label axonemes in adjacent sections of these tissues ( Fig 3B” and 3C” ) . No obvious changes to oviduct or trachea cilia were observed in Kif6 p . G555fs mutant mice at P21 ( S5 Fig ) , suggesting that Kif6 expression and function are tightly restricted to the multiciliated ECs in mouse . Taken together these data suggested a cellular mechanism centered on defective ECs underlying the development of hydrocephalus in Kif6 mutant mice . Defects of EC cilia are known to cause hydrocephalus in mouse [8] . To assay this we first utilized scanning electron microscopy ( SEM ) to directly visualize the LV en face . Heterozygous Kif6p . G555fs/+ mice displayed a high-density of regularly spaced EC multiciliated tufts along the LV surface ( Fig 4A and 4A’ ) , typical at P21 in mouse development [22] . In contrast , homozygous Kif6p . G555fs mutant mice displayed a marked reduction of multiciliated tufts across the LV wall ( p = 0 . 029; Avg . # ECs /sectionhet = 225; Avg . # ECs /sectionmutant = 129 . 3 ) , coupled with a reduction in the density of ciliary axonemes within these ciliary tufts ( Fig 3B and 3B’ ) . This phenotype was even more obvious at P28 in Kif6 mutant mice ( p = 0 . 001; Avg . # ECs /sectionhet = 146; Avg . # ECs /sectionmutant = 0 ) ( S6 Fig ) . Together , these data suggested that hydrocephalus may result from either a reduction in EC differentiation and/or defects in EC cilia during postnatal development . To address the differentiation status of the ECs , we utilized IF in coronal sectioned brain tissues to image known proteins components of the EC and their cilia . At P21 , we observed the expression of the ependymal cell-marker S100B [22] throughout the epithelium lining luminal surface of the ventricles , as well as , the presence of apically localized γ-tubulin-positive basal bodies within these ECs in both WT ( Fig 4C and 4C'' ) and Kif6p . G555fs mutant mice ( Fig 4D and 4D'' ) . Conversely , we observed a obvious reduction in the density of CD133-positive EC axonemes [23] extending into the ventricular lumen in Kif6p . G555fs mutant mice ( Fig 4C and 4C‴ ) , compared with WT ( Fig 4D and 4D‴ ) . Quantitation of binned mean fluorescence intensity from of CD133-positive axonemes confirmed a severe reduction of EC axonemes in Kif6 mutant mice ( n = 9 mice/genotype , p≤0 . 001 ) ( Fig 4E ) . In order to address whether Kif6 is required for EC cilia formation or ciliogenesis , we performed IF at P14 , the time point at which the ECs are fully ciliated across the ventricular system in mouse [22] . We observe significant alterations in the extension of EC ciliary axonemes ( CD133-positive ) ( n = 5 mice/genotype , p≤0 . 01 ) , without an alterations in the specification of ECs or their apical polarity ( S7 Fig ) . The reduction of EC cilia was also observed at both P14 and P21 in Kif6p . G555fs mutant mice using an acetylated tubulin antibody , which labels both primary and motile cilia as well as neuronal cell types in the cortex ( S8 Fig ) . Taken together , these results suggest that the onset of hydrocephalus in Kif6 mutant mice is primarily due to a general defect of cilia formation ( ciliogenesis ) and not the result of defects in differentiation of the ECs or a loss of EC cilia as the result of the onset of hydrocephalus . Alterations of the motility or synchronicity of EC cilia beating are known to cause hydrocephaly in mouse [8] . In order to address if Kif6 is required for normal EC cilia beating we performed live ex vivo imaging of lateral wall explants taken from both Kif6 mutant and littermate control mice . We consistently observed multiple tufts of EC cilia in WT animals that beating in a synchronous fashion at P21 , with obvious flow generation , demonstrated by the movement of fortuitous particles within the media ( S1 Movie ) . As expected , we observed an obvious reduction in formation of EC cilia coupled with reduced particle flow in explants from Kif6 mutant mice ( S2 Movie ) . Interestingly , on the rare occasions where we did observed EC cilia in these mutant mice , the motility of these EC cilia appeared typical . These results in conjunction with our IF and SEM analysis ( Fig 4 , S6–S8 Figs ) suggest that the onset of progressive hydrocephaly in Kif6 mutant mice is the result of defective ciliogenesis of ECs leading to a reduction in near-wall CSF flow . Defects in EC cilia in the central canal leads to impaired CSF flow and hydrocephalus in zebrafish embryos [24] . Our previous studies in kif6sko mutant zebrafish observed late-onset scoliosis in larval zebrafish , without defects of CSF flow , EC cilia in the central canal , or hydrocephalus during early embryonic development [19] . However , in contrast to the defined period of EC development in the mouse , zebrafish demonstrate continuous differentiation of ECs within the ventricular system throughout adult development [25] . Interestingly , recent studies in a variety mutant zebrafish demonstrate that reduced CSF flow , ventricular dilation , and loss of EC cilia may underlie the onset late-onset scoliosis in larval zebrafish [26] . In order to determine if kif6 mutant zebrafish might also display changes in the ventricular system in adults , we used iodine contrast-enhanced , micro computed tomography ( μCT ) [27] to generate high-resolution ( 5 μm voxel size ) images of aged matched , 3-month-old WT and kif6sko homozygous mutant zebrafish ( Fig 5A and 5B’ ) . After reconstruction and alignment of 3D tomographic datasets in the coronal plane , we utilized visualization software ( Avizo Lite v . 9 . 5 ) for 3D-reconstruction and segmentation of a virtual endocast to represent the ventricular volume in wild-type ( WT ) and kif6sko mutant zebrafish ( Fig 5A–5D’ ) . Analysis of the ventricles in the endocast of the adult zebrafish brain highlighted several dysmorphophic regions in kif6sko mutants including dilation of the diencephalic ventricle ( DiV ) ( Fig 5D ) and dilation of regions of the central canal ( red arrows; Fig 5D and 5D’ ) . At the same time , we observe that while some ventricles patent in wild-type zebrafish , were completely obstructed or less open in kif6sko mutants ( asterisks; Fig 5C’ and 5D’ ) . To further describe and quantify alterations of the ventricular system we utilized individual transverse optical slices from these contrast-enhanced μCT datasets , using stereotyped landmarks of the zebrafish brain and spinal cord to compare equivalent axial sections . At distinct axial levels of the brain ( Fig 5E ) , we observed consistent dilation of the ventricular system and central canal in kif6sko homozygous mutant zebrafish ( yellow arrows; Fig 5G , 5I and 5K ) , compared to the stereotyped anatomy described for the adult zebrafish brain ( Fig 5F , 5H and 5J ) [28] . Multiple regions of kif6sko mutant zebrafish brain were found to be structurally abnormal in kif6sko mutants compared to WT zebrafish ( S3 , S4 Movies ) . We next quantified the areas of two anatomically distinctive ventricles in our tomographic datasets: ( i ) the medial tectal ventricle ( TecV ) at the medial division of valvula cerebelli ( Vam ) ( Fig 5F and 5G ) and ( ii ) a region of the rhombencephalic ventricle ( RV ) just posterior to the lobus facialis ( Fig 5H and 5I ) [28] . We observed a significant increase in the area ( dashed yellow line , Fig 5F–5I ) of the medial TecV and the posterior RV in kif6sko mutant zebrafish comparing several optical sections from independent aged-matched zebrafish ( n = 3 fish/genotype; p<0 . 0001 ) . The central canal was also clearly dilated in kif6sko mutant zebrafish ( yellow arrow , Fig 5K ) . However , we were unable to reliably quantify this area in WT samples at the current imaging resolution . Our previous observations in kif6sko mutant zebrafish embryos failed to find phenotypes that are characteristic of cilia defects , such as hydrocephalus , situs inversus , or kidney cysts [19] . Moreover , we observed normal development and function of motile EC cilia within the central canal in embryonic mutant zebrafish [19] . These data , coupled with our new observations of ventricular dilation in adult kif6 mutants ( Fig 5 ) , suggest that Kif6 is required for the post-embryonic , robust development of the EC cilia within the ventricles of the brain as was reported in other zebrafish mutants displaying similar late-onset scoliosis , as observed in kif6 mutant zebrafish [19 , 26] . In order to assay whether EC cilia were affected during adult development in zebrafish , we isolated a stable transgenic allele , Tg ( Foxj1a:GFP ) dp1 previously reported to label multiciliated Foxj1a-positive multi-ciliated cells ( MCCs ) , including ECs , with cytoplasmic EGFP in zebrafish [26] . Using this transgenic approach , we observed no differences in the specification of Foxj1a:GFP-positive ECs comparing adult ( 90dpf ) heterozygous kif6sko/+ phenotypically wild-type and homozygous kif6sko mutant fish ( Fig 6A and 6B ) . Cytoplasmic GFP can freely diffuse through the transition zone of cilia and label the axoneme [29] . As such , we were also able to observe GFP-filled EC axonemes projecting into the ventricular lumen in Tg ( Foxj1a:GFP ) dp1; kif6sko/+ heterozygous fish ( red arrowheads; Fig 6A ) . In contrast , these GFP-filled EC axonemes were reduced or absent in Tg[Foxj1a:GFP] dp1; kif6sko mutant fish ( Fig 6B ) . To further support our model that EC cilia were affected in kif6 mutant zebrafish , we generated a novel-transgenic fish line Tg ( Foxj1a:Arl13b-GFP ) dp22 to allow for the direct fluorescent-labeling of ciliary axonemes with mouse Arl13b-GFP specifically in fox1a-expressing lineages . Analysis of kif6sko; Tg ( Foxj1a:Arl13b-GFP ) dp22 transgenic adult mutant zebrafish ( 90dpf ) demonstrated an obvious reduction in Arl13b-GFP labeled axonemes ( Fig 6D ) , in comparison to robust labeling of EC cilia in heterozygous kif6sko/+;Tg ( Foxj1a:Arl13b-GFP ) dp22 sibling fish ( Fig 6C ) . Furthermore , SEM imaging of the telencephalic and rhombencephalic ventricles in kif6sko mutant zebrafish demonstrated: ( i ) ventricular dilation ( dashed red line , S9 Fig ) and ( ii ) an obvious reduction in EC cilia in adult zebrafish ( p = 0 . 014; Avg . # ECs /sectionhet = 37 . 5; Avg . # ECs /sectionmutant = 1 ) ( red arrowheads , S9 Fig ) . Akin to our observations in Kif6 mutant mice trachea , we did not observe defects of other multiciliated tissues in kif6sko zebrafish mutants , such as the nasal cilia ( S5 Fig ) . Together , these data suggest that Kif6 functions specifically in the formation of EC cilia , in regulation of ventricular homeostasis during adult development in zebrafish . Fluorescently-tagged Kif6 ( Kif6-GFP ) localizes to the basal body of Kuper’s vesicle motile cilia in zebrafish [30] . Indeed our own attempts to visualize Kif6-EGFP by microinjection demonstrate that Kif6-EGFP localizes to microtubule-rich spindle poles and mitotic spindle during cell divisions in the rapidly dividing blastomeres of the early embryo ( S10 Fig ) . Our attempts to drive Kif6-EGFP within the ECs using the foxj1a promoter in transgenic zebrafish have failed for this purpose , whereas clear signal was obtained for similar foxj1a-driven GFP and Arl13bGFP transgenic constructs ( Fig 6 ) . Unfortunately , IF and Westerns using several commercially-available KIF6 antibodies were unsuccessful to report on endogenous KIF6 in both mouse and zebrafish tissues ( S11 Fig ) . For these reasons , we turned to the muco-ciliated Xenopus laevis epidermis in order to address Kif6 localization in an analogous MCCs lineage . The Xenopus mucociliary epithelium is analogous to the airway epithelium of the mammalian trachea [31] . Importantly , this system provides a robust model system for both genetic analysis [32] and robust visualization of fluorescently-tagged proteins in MCCs in vivo [33 , 34] . By microinjection of synthetic Xenopus Kif6-GFP RNA , we observed Kif6-GFP localization at the basal bodies of MCCs ( Fig 7D and 7F ) and co-localized with a basal body marker; Centrin-BFP [35] ( Fig 7E and 7F ) . Similarly , we observed Kif6-GFP puncta within the axoneme ( co-labeled with pan-membrane bound RFP ) ( Fig 7A–7C ) . Interestingly , similar punctate localization at the basal bodies and within the axoneme has been previously been shown in Xenopus MCCs using various fluorescently-tagged IFT proteins [36] . Together these data suggest a model in which Kif6 may act as a component of IFT trafficking uniquely and specifically in EC cilia , which is dispensable for the function of other MCCs in mouse and zebrafish .
This study demonstrates the importance of KIF6 for EC cilia formation and homeostasis of the ventricular system in vertebrates , and potentially implicates a novel locus for understanding neurological defects in humans . This is supported by several lines of evidence including the discovery of a novel nonsense-mutation of KIF6 in a child with intellectual disability and megalencephaly and underscored by functional analysis in both mouse and zebrafish Kif6 mutant models ( Table 2 ) . We identified a homozygous KIF6 c . 1193delT mutation in a child with macrocephaly and cognitive impairment that segregated with this phenotype in his family , and leads to a loss of the C-terminal second and third coiled-coil regions which are important for dimerization and cargo selectivity of kinesin motors [13] . We engineered an analogous , C-terminal truncating mutation of KIF6 in mouse , which displays severe hydrocephalus and defects of EC cilia providing strong evidence for pathogenicity of the mutation in the child . Other than the case described here , no prior mutation directly attributed to human disease has been described for KIF6 . Taken together , the clinical data reported here suggest that biallelic mutations in KIF6 may underlie some unexplained intellectual disability and neurological developmental defects . Future analyses of KIF6 mutations in these patient groups are warranted . In addition , our analyses of several independent loss-of-function Kif6 mutant animal models found no evidence of obvious heart abnormalities to explain the prior association of the common variant KIF6 p . W719R in some[17] , but not all [18] , studies of coronary heart disease in humans . Because expressed sequence tag clones of KIF6 have not been reported from cDNAs libraries derived from human heart or vascular tissues ( UniGene 1956991—Hs . 588202 ) , any possible functional effects of KIF6 on heart function remains unexplained . However , detailed analysis of coronary function was not explored in our models , therefore it is possible that subtle defects may be present . Previous reports of an ENU-derived Kif6 splice acceptor site mutant mouse strain , predicted to delete the 3rd exon of KIF6 ( Kif6Δ3/Δ3 ) , also did not show cardiac or lipid abnormalities [37] . Of note this mutant mouse was also not reported to have hydrocephalus . Our analysis shows that the loss of exon 3 in Kif6Δ3/Δ3 mutant mouse generates an inframe deletion of only 25 amino acids in the N-terminal motor domain of the KIF6 protein , otherwise generating a mostly full-length KIF6 protein ( Table 2 ) . In contrast , here we report two novel Kif6 mutant mice: ( i ) a C-terminal Kif6p . G555fs/p . G555fs deletion mutant , predicted to truncate 248 amino acids of the C-terminal domain , which are important for cargo binding in Kinesin motor proteins [13]; and ( ii ) a conditional CMV-Cre;Kif6tm1c/tm1c mutant which recombines exon 4 leading to an early frame shift mutation predicted to generate a N-terminal truncated 122 amino acid KIF6 protein ( Table 2 ) , both of which display indistinguishable progressive , hydrocephalus with EC ciliogenesis defects . The most parsimonious explanation for the difference in phenotypes in these mutant mice is that the Kif6Δ3 allele encodes a functional KIF6 protein . Analysis of these mutations in trans or quantitative analysis of these kinesin motor proteins in vitro is warranted to more fully address these conflicting observations . There are noticeable differences in the phenotypes among the human , mouse , and zebrafish associated with mutations in KIF6 . For example , kif6 mutant zebrafish display post-natal onset scoliosis , mirroring adolescent idiopathic scoliosis ( IS ) in humans [38] . The formation of IS-like defects in zebrafish has been shown to be the result of a loss of CSF flow , associated with a loss of EC cilia and ventricular dilation during a defined window of larval zebrafish development [26] . Interestingly , we did not observe scoliosis in the Kif6 mutant mice ( S12 Fig ) , despite being of an appropriate age when IS-like scoliosis can manifest in mouse [39] . Moreover , we do not observe scoliosis in the patient at the age of 10 years , though it is possible that he may yet develop scoliosis during adolescence . The mechanisms behind these differences may reflect distinctions in the functional input of the ventricular system for spine stability amongst teleosts and amniotes . Furthermore , while we observe a clear role for KIF6 in maintaining the ventricular system in mouse and zebrafish , the patient does not have obvious hydrocephalus . However , his relative macrocephaly and slightly enlarged ventricles by MRI ( Fig 1B–1D ) may suggest an element of what is commonly referred to as arrested hydrocephalus [40] . Moreover , the contribution of EC cilia beating to bulk CSF flow might be species dependent . For instance , the majority of CSF flow in humans is thought to occur via the generation of a source-sink gradients; partly from the secretion of the choroid plexus and exchanges of the interstitial fluids , coupled with absorption at the arachnoid villi and lymphatics [41] . In contrast , localized or near-wall CSF flow [4] , generated by polarized beating of EC cilia , are clearly important for the formation of hydrocephalus in rodents [8]; however , there have been limited evidence of EC cilia defects causing hydrocephalus in humans . Regardless there is growing evidence suggesting that EC cilia dependent CSF flow is crucial for the regulation of brain function and neurogenesis [4] , and for adult neural stem cell proliferation [9] . It is possible that a specific loss of EC cilia in humans may only have minor effects on CSF bulk flow and ventricular homeostasis , while causing severe defects of neurogenesis leading to intellectual disability and other neurological disease . It will be important to determine ( i ) if the loss of KIF6 function during adult development in mouse will lead to a reduction in EC cilia; and ( ii ) whether the loss of EC cilia function contributes to ventricular dilation and decline of neurological function . Finally , KIF6 now joins five other kinesin genes , KIF1C , KIF2A , KIF4A , KIF5C and KIF7 that were previously reported to be associated with neurological abnormalities in humans [42–45] . Here we suggest that KIF6 has a uniquely specific function in the EC cilia in vertebrates , resulting in both cognitive impairment and macrocephaly in a child with a homozygous one-base pair deletion . Using a cell biological and transgenic approaches in both mouse and zebrafish , we identified specific loss of EC cilia these Kif6 mutant models suggesting a strong conservation of KIF6 function in ventricular system in vertebrates . Furthermore , we utilized imaging of fluorescently-tagged Kif6 in MCCs of the Xenopus epidermis , which are anatomically and functionally analogous to ECs lining the ventricles of the brain . Using this heterologous system we demonstrated that Kif6-GFP localizes to the basal bodies and as puncta within the ciliary axonemes in these MCCs , which is reminiscent of observations of canonical IFT proteins [36 , 46] . This imaging data coupled with our findings of defective ciliogenesis in ECs in both Kif6 mutant mice and zebrafish suggest a model where KIF6 is acting in concert with one or more anterograde Kinesin-II motors to promote robust ciliogenesis specifically in ECs . Indeed , tissue specific accessory IFT motors have been described in amphid sensory neurons in C . elegans [47] and zebrafish photoreceptors [48 , 49] . Interestingly , a homologue of KIF6 , KIF9B , has been shown to be critical for flagellar motility and for the stabilization of the paraflagellar rod structure which tightly abutted the flagella in the protist , Trypanosoma brucei [15] . It remains to be determined whether KIF6 has an analogous , structural-functional role in ECs which may provide a unique functional role in these cilia , which is not critical for other analogous MCC lineages such as the trachea or oviduct . The confinement of LacZ expression specifically with EC in the Kif6-LacZtm1b gene trap mouse suggest a large part of this specificity of function is simply due to strict regulation of Kif6 expression to the EC lineages , which would implicate the action of a tightly regulated , EC specific-promoter driving expression of the Kif6 locus . Both models will be important to test in future studies to better understand the critical components and pathways important for EC cilia development .
The collection and use of human DNA samples in this study was approved by the Institutional Review Board Faculty of Medicine , Chulalongkorn University , Bangkok , Thailand ( IRB 381/61 ) . All subjects provided written informed consent prior to inclusion in the study . All animal research was conducted according to federal , state , and institutional guidelines and in accordance with protocols approved by Institutional Animal Care and Use Committees at University of Texas at Austin ( AUP-2015-00185; AUP-2015-00187; and AUP-2018-00225 ) . The patient’s genomic DNA of patient was extracted from peripheral blood leukocyte using AchivePure DNA Blood Kit ( 5 Prime Inc . , Gaithersburg , MD ) . The sample was sent to Macrogen , Inc . ( Seoul , Korea ) for whole exome sequencing . The 4 ug of DNA sample was enriched by TruSeq Exome Enrichment Kit and was sequenced onto Hiseq 2000 . The raw data per exome was mapped to the human reference genome hg19 using CASAVA v1 . 7 . Variants calling were detected with SAMtools . The sample was sent to Macrogen , Inc . ( Seoul , Korea ) for genotyping . The DNA sample was genotyped by HumanOmni 2 . 5-4v1 DNA BeadChip ( Illumina ) which contain 2 , 443 , 177 SNPs . The experiment was performed by the array protocol . PLINK was used to analyze for the homozygous regions . We performed resequencing of KIF6 pathogenic region in patient and patient’s family . Primers for the amplification of the candidate variant were designed using Primer 3 software ( version 0 . 4 . 0 ) . Primers KIF6-1193delT-F 5’-CAGCTTGAACATGGCTGAAA-3’ and KIF6-1193delT-R 5’-TTCTGTAAAGAGGTGGGAACAA-3’were used to amplify . The 20 ul of PCR reaction contained 50–100 ng of genomic DNA , 200 uM of each dNTP , 150 nM of each primer , 1 . 5 mM MgCl2 and 0 . 5 unit of Taq DNA polymerase ( Fermentas Inc . , Glen Burnie , MD ) . The PCR condition was started with 95 oC for 5 min for pre-denaturation following with the 35 cycles of 94 oC for 30 sec , 55 oC for 30 sec and 72 oC for 30 sec . The product size of these primers is 276 bp . For sequencing , PCR products were treated with ExoSAP-IT ( USP Corporation , Cleveland , OH ) , and sent for direct sequencing at Macrogen Inc . ( Seoul , Korea ) . Bi-directional sequencing was done by using KIF6-1193delT F and R primers . Analyses were performed by Sequencher 4 . 2 ( Gene Codes Corporation , Ann Arbor , MI ) . One hundred chromosomes and patient’s trio were genotyped by PCR-RFLP . Primer KIF6 MfeI F 5’-TGGCTTCACTATAAATTTCACTTTGTCAATG-3’ and mutagenic primer KIF6 mutagenic MfeI R 5’-TCCTGGTCTTCCAAAAAGGATGCAAT-3’were used to amplify KIF6 T-deletion . The 20 ul of PCR reaction contained 50–100 ng of genomic DNA , 200 uM of each dNTP , 150 nM of each primer , 1 . 5 mM MgCl2 and 0 . 5 unit of Taq DNA polymerase ( Fermentas Inc . , Glen Burnie , MD ) . The PCR condition was started with 95 C for 5 min for pre-denaturation following with the 35 cycles of 94 C for 30 sec , 60 C for 30 sec and 72 C for 30 sec . The product size of these primers is around 223 bp . The PCR product was incubated with 10U of Mfe-HF ( New England Biolabs , Ipswich , MA ) at 37 C overnight . Three percent of agarose gel electrophoresis was used to detect the different cut sizes of PCR product . A 196 bp and 26 bp bands were present in one base deletion sample . All mouse studies and procedures were approved by the Animal Studies Committee at the University of Texas at Austin ( AUP-2015-00185 ) . The Kif6p . G555fs mutant mouse were developed using CRISPR-Cas9-mediated genome editing . Using the CHOP-CHOP online tool [50] , we identified a suitable 20-nucleotide site ( GGAGATGTCACTGGGACGCC ) targeting exon 14 of mouse Kif6 ( ENSMUST00000162854 . 1 ) in order to generate a C-terminal truncation allele . The gene specific and universal tracrRNA oligonucleotides ( S3 Table ) were annealed , filled in with CloneAmp HiFi PCR premix , column purified , and directly used for in vitro transcription of single-guide RNAs ( sgRNAs ) with a T7 Polymerase mix ( M0255A NEB ) . All sgRNA reactions were treated with RNAse free-DNAse . We utilized a ssDNA oligo ( S3 Table ) to insert a frameshift mutation in all three reading frames , along with 8-cutter restriction sites for genotyping ( 3-stop donor ) [51] ( S2 Fig ) . The Kif6 ex14 3-stop donor and mKif6-R2-ex14-T7 sgRNA were submitted for pronuclear injection at the University of Texas at Austin Mouse Genetic Engineering Facility ( UT-MGEF ) using standard protocols ( https://www . biomedsupport . utexas . edu/transgenics ) . We confirmed segregation of the Kif6p . G555fs allele using several methods including increased mobility on a high percentage electrophoresis gel , donor-specific primer PCR , or PmeI ( NEB ) digestion of the Kif6 exon14 amplicon ( S2 Fig and S3 Table ) . PCR products in isolated alleles were cloned to pCRII TOPO ( ThermoFisher ) to identify scarless integration of the 3-stop donor at the Kif6 locus using gene specific flanking primers ( S3 Table ) . Kif6-LacZtm1b mice were generated by injection of embryonic stem cell clones obtained from the Knockout Mouse Project ( KOMP ) Repository . Three Kif6tm1a ( KOMP ) Mbp embryonic stem ( ES ) cell clones ( KOMP: EPD0736_3_G01; EPD0736_3_H02; and EPD0736_3_A03 ) all targeting exon 4 of the Kif6 gene with a promoter-driven targeting cassette for the generation of a 'Knockout-first allele' [52] . Pronuclear injections of all clones were done using standard procedures established by the UT MGEF . After screening for germline transmission , we isolated and confirmed a single heterozygous founder male ( Kif6tm1a ( KOMP ) Mbp ) carrier derived from the G01 clone . We confirmed the locus by long-range PCR , several confirmation PCR strategies targeting specific transgene sequences , and Sanger sequencing of the predicted breakpoints ( S3 Table ) . After several backcrosses to the WT C57BL/6J substrain ( JAX ) , we crossed a hemizygous Kif6tm1a/+ mutant male to a homozyogus CMV-Cre female ( B6 . C-Tg ( CMV-cre ) 1Cgn/J ) ( JAX , 006054 ) to convert the Kif6tm1a allele to a stable LacZ expressing Kif6tm1b allele ( Kif6-LacZtm1b ) . Mutant F1 offspring from this cross were backcrossed to WT C57BL6/J mice and the F2 progeny were genotyped to confirm the Kif6-LacZtm1b allele and the presence/absence of the CMV-Cre transgene . A single founder Kif6-LacZtm1b with the desired genotype ( Kif6-LacZtm1b hemizygous , Cre transgene absent ) was used to expand a colony for spatial expression analysis . Kif6tm1c conditional ready mice were generated by outcross of the Kif6tm1a ( KOMP ) Mbp allele described above to a ubiquitously expressed Flippase strain ( 129S4/SvJaeSor-Gt ( ROSA ) 26Sortm1 ( FLP1 ) Dym/J ) ( JAX , 003946 ) . F1 offspring were genotyped and sequenced at several breakpoints to ensure proper flip recombination and a single F1 founder was used to backcross to C57B6/J for propagation of the Kif6tm1c strain . Analysis of recombination of the floxed Kif6tm1c was performed by crossing homozygous Kif6tm1c/tm1c to a compound heterozygous CMV-Cre; Kif6tm1c/+ mouse . Recombination of the exon 4 of Kif6 was confirmed by PCR-gel electrophoresis analysis ( S3 Table ) . Mice were perfused with LacZ fixative and post fixed for 2 hours at RT . Whole brains were then stained in X-gal solution overnight at 37°C followed by post-fixation in 4% PFA overnight at 4°C . The samples were then prepped for cryosectioning in 30% sucrose/OCT and sectioned . Sections were counter stained in Nuclear Fast Red stain ( Sigma ) . Radiographs of the mouse skeleton were generated using a Kubtec DIGIMUS X-ray system ( Kubtec , T0081B ) with auto exposure under 25 kV . All zebrafish studies and procedures were approved by the Animal Studies Committee at the University of Texas at Austin ( AUP-2015-00187 ) . Adult zebrafish of the AB were maintained and bred as previously described [53] . Individual fish were used for analysis and compared to siblings and experimental control fish of similar size and age . Independent experiments were repeated using separate clutches of animals . Strains generated for this study: Tg ( Foxj1a:GFP ) dp1 and Tg ( foxj1a::Arl13b-GFP ) dp15 . Transgenic lines were generated using a Gateway-compatible middle entry cloning containing mouse Arl13b open reading frame[54]was modified to include a C-terminal GFP by megaprimer PCR to generate pME-Arl13b GFP . This clone was recombined with p5E-foxj1aP [26] , p3E-polyA[55] and pDEST pDestTol2pACryGFP to generate a final transgenesis vector . Embryos were injected at the one-cell stage with 25 pg of assembled transgene and 25 pg Tol2 mRNA . Embryos were sorted at 48 hpf for reporter expression ( GFP+ eyes ) and were subsequently grown to adulthood . Individuals were bred to TU wild-type zebrafish to generate a stable F1 line , and subsequently bred into a kif6sko mutant background . pDestTol2pACryGFP was a gift from Joachim Berger & Peter Currie ( Addgene plasmid # 64022 ) . Previously published strains: kif6sko [19] . Xenopus embryo manipulations were carried out using standard protocols [56] . Full length of Xenopus Kif6 cDNA sequence was provided from Xenbase ( www . xenbase . org ) and amplified from Xenopus cDNA library by PCR and inserted in-frame into pCS10R-eGFP . 5’-capped Kif6-GFP RNA was synthesized using mMESSAGE mMACHINE SP6 transcription kit ( Invitrogen Ambion ) . Synthetic 5’-capped RNAs: Kif6-GFP , membrane RFP and Centrin4-BFP [35] were injected into two ventral blastomeres at the 4-cell stage with ~ 40 pg/RNA/injection . Live images were captured with a Zeiss LSM700 laser scanning confocal microscope using a plan-APOCHROMAT 63X 1 . 4 NA oil objective lens ( Zeiss ) . Mice were humanely euthanized by extended CO2 exposure and transferred to chemical hood where the mouse was perfused with buffered saline followed by 4% PFA . Whole brains were placed in 4% PFA 4 hours at RT , then at 4° C overnight . Zebrafish were euthanized by exposure to lethal , extended dose of Tricane ( 8% ) followed by decapitation . Zebrafish brains were extracted and fixed in 4% PFA at 4° C overnight . For paraffin embedding , the fixed brains were embedded and cut using standard paraffin embedding and sectioning protocols . Paraffin sections were stained with standard hematoxylin-eosin solution . For frozen sections both mouse or zebrafish brains were fixed as above and then equilibrated to 30% or 35% sucrose , respectively at 4° C overnight . Whole brains were then placed in O . C . T . Compound ( Tissue-Tek ) and flash in cold ethanol bath . All blocks were stored at -80° Celsius until sectioning on a cryostat ( Leica ) . All sections were dried at RT for ~2hrs . and stored at -80°C until use . Sectioned tissues were warmed at room temperature for ~1 hour , then washed thrice in 1xPBS + 0 . 1% Tween ( PBST ) . Antigen retrieval was hot citrate buffer ( pH6 . 8 ) . Blocking was done in 10% Normal goat serum ( Sigma ) in 1xPBST . Primary antibodies ( S100B at 1:1 , 000 , ab52642 , Abcam; CD133 ( Prominin-1 ) , 134A , 1/500; Gamma Tubulin , sc-17787 , Santa Cruz ( C-11 ) , 1/500; Anti-GFP , SC9996 , Santa Cruz , 1:1 , 000 ) were diluted in 10% NGSS , 1xPBST and allowed to bind overnight at 4°C in a humidified chamber . Secondary fluorophores ( Alexa Fluor 488 ( A-11034 ) ; 568 ( A10042 ) ; and 647 ( A32728 ) , 1:1 , 000 , ThermoFisher ) were diluted in 10% NGS; 1xPBST were allowed to bind at RT for ~1hr . We used Prolong gold with DAPI ( Cell Signaling Technologies , 8961 ) to seal coverslips prior to imaging . Zebrafish specimens were fixed overnight in 10% buffered formalin , washed thrice in diH2O and stained ~48 hours in 25% Lugol’s solution/75% distilled water . Specimens were scanned by the High-resolution X-ray CT Facility ( http://www . ctlab . geo . utexas . edu/ ) on an Xradia at 100kV , 10W , 3 . 5s acquisition time , detector 11 . 5 mm , source -37 mm , XYZ [816 , 10425 , –841] , camera bin 2 , angles ±180 , 1261 views , no filter , dithering , no sample drift correction . Reconstructed with center shift 5 . 5 , beam hardening 0 . 15 , theta -7 , byte scaling [–150 , 2200] , binning 1 , recon filter smooth ( kernel size = 0 . 5 ) . Brains were isolated from freshly euthanized mice and were dissected in cold DMEM/F12 media . The brain was cut into thin coronal slices , promptly placed onto cover slip with PBS , and imaged using oblique lighting with a Keyence BZ-X800 microscope . GraphPad Prism version 7 . 0c for Mac ( GraphPad Software ) was used to analyze and plot data . Images for measurement were opened in FIJI ( Image J ) [57] , and measures were taken using the freehand tool to draw outlines on ventricular area or whole brain area . Statistically significant differences between any two groups were examined using a two-tailed Student’s t-test , given equal variance . P values were considered significant at or below 0 . 05 . | Cerebrospinal fluid flow is crucial for neurodevelopment and homeostasis of the ventricular system of the brain . Localized flows of cerebrospinal fluid throughout the ventricular system of the brain are established from the polarized beating of the ependymal cell ( EC ) cilia . Here , we identified a homozygous truncating mutation in KIF6 in a child displaying neurodevelopmental defects and intellectual disability . To test the function of KIF6 in vivo , we engineered mutations of Kif6 in mouse . These Kif6 mutant mice display severe hydrocephalus , coupled with defects in the formation of EC cilia . Similarly , we observed hydrocephalus and a reduction in EC cilia in kif6 mutant zebrafish . Overall , this work describes the first clinically-defined KIF6 mutation in human , while our animal studies demonstrate the pathogenicity of mutations in KIF6 and establish KIF6 as a conserved mediator of ciliogenesis in ECs in vertebrates . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"fish",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"kinesins",
"nervous",
"system",
"vertebrates",
"animals",
"animal",
"models",
"osteichthyes",
"mutation",
"model",
"organisms",
"molecular",
"motors",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"frameshift",
"mutation",
"cellular",
"structures",
"and",
"organelles",
"research",
"and",
"analysis",
"methods",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"studies",
"proteins",
"hydrocephalus",
"mouse",
"models",
"molecular",
"biology",
"biochemistry",
"zebrafish",
"cytoskeletal",
"proteins",
"eukaryota",
"cell",
"biology",
"anatomy",
"cilia",
"polymerase",
"chain",
"reaction",
"neurology",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"microtubule",
"motors",
"cerebrospinal",
"fluid",
"organisms"
] | 2018 | Mutations in Kinesin family member 6 reveal specific role in ependymal cell ciliogenesis and human neurological development |
UTX ( KDM6A ) and UTY are homologous X and Y chromosome members of the Histone H3 Lysine 27 ( H3K27 ) demethylase gene family . UTX can demethylate H3K27; however , in vitro assays suggest that human UTY has lost enzymatic activity due to sequence divergence . We produced mouse mutations in both Utx and Uty . Homozygous Utx mutant female embryos are mid-gestational lethal with defects in neural tube , yolk sac , and cardiac development . We demonstrate that mouse UTY is devoid of in vivo demethylase activity , so hemizygous XUtx− Y+ mutant male embryos should phenocopy homozygous XUtx− XUtx− females . However , XUtx− Y+ mutant male embryos develop to term; although runted , approximately 25% survive postnatally reaching adulthood . Hemizygous X+ YUty− mutant males are viable . In contrast , compound hemizygous XUtx− YUty− males phenocopy homozygous XUtx− XUtx− females . Therefore , despite divergence of UTX and UTY in catalyzing H3K27 demethylation , they maintain functional redundancy during embryonic development . Our data suggest that UTX and UTY are able to regulate gene activity through demethylase independent mechanisms . We conclude that UTX H3K27 demethylation is non-essential for embryonic viability .
Post-translational modifications of histones establish and maintain active or repressive chromatin states throughout cell lineages . Thus , the enzymes that catalyze these modifications often have crucial roles in establishing genomic transcriptional states in developmental decision-making . Histone methylation can stimulate gene activation or repression depending on which residues are targeted . Methylation of histone H3 on Lysine 4 ( H3K4me ) is an active chromatin modification , while methylation on histone H3 Lysine 27 ( H3K27me ) is associated with repression of gene activity [1] . The polycomb repressive complex 2 ( PRC2 ) methylates H3K27 [2] , [3] , [4] , [5] . Within this complex , enhancer of zeste homolog 2 ( EZH2 ) catalyzes di and tri-methylation of H3K27 . Embryonic ectoderm development ( EED ) and suppressor of zeste homolog 12 ( SUZ12 ) are additional PRC2 core components indispensible for PRC2 activity [6] , [7] , [8] . EZH1 is a secondary , less efficient H3K27 methyl-transferase that shares some overlapping redundancy with EZH2 in ES cells and epidermal stem cells [9] , [10] , [11] , [12] . The PRC1 complex is recruited through H3K27 trimethylation for additional histone modification and chromatin compaction [13] . In embryonic stem ( ES ) cells , PRC2 targets and represses genes essential for developmental events [14] , [15] , [16] , [17] . The promoters of these PRC2 targets typically contain “bivalent” chromatin marks with both active H3K4 and repressive H3K27 methylation [18] , [19] , [20] . Loss of PRC2 activity de-represses these genes but does not alter ES cell pluripotency [14] . However , mouse mutations in any of the three PRC2 core components are early embryonic lethal with gastrulation defects [7] , [21] , [22] . H3K27 trimethylation is reversible as a family of histone demethylases catalyzes the removal of this epigenetic mark [23] , [24] , [25] , [26] . JMJD3 ( KDM6B ) is an autosomal H3K27 demethylase upregulated during specific differentiation events [25] , [27] . UTX ( KDM6A ) is a broadly expressed X-linked H3K27 demethylase that escapes X-inactivation [23] , [24] , [26] , [28] . UTY is the Y chromosome homolog of UTX . Both UTX and JMJD3 demethylate H3K27 di and tri-methyl residues; however , UTY lacks this activity in vitro [26] , [29] . Based on cell culture models , UTX and JMJD3 mediated H3K27 demethylation is vital in a wide array of functions including cell cycle regulation , M2 macrophage differentiation , neuronal stem cell specification , skin differentiation , and muscle differentiation [27] , [30] , [31] , [32] , [33] , [34] , [35] . In contrast , the biological function of UTY remains unknown . Utx and Uty are genetically amenable to delineate H3K27me3 demethylation dependent versus demethylation independent function in mouse development . Comparative amino acid sequence analysis of UTX and UTY reveals 88% sequence similarity in humans ( 83% identity ) and 82% sequence similarity in mouse . Across the annotated JmjC histone demethylase domain , the similarity is at 98% and 97% for human and mouse respectively . In the TPR ( tetratricopeptide repeat ) domain , the similarity is at 94% . So while UTY is reported to have lost H3K27 demethylase activity , it is remarkably well conserved with respect to UTX . Recent discoveries have revealed that JMJD3 functions in macrophage lipopolysaccharide response and lymphocyte Th1 response through H3K27 demethylase independent gene regulation [36] , [37] , suggesting that function of this family of proteins is not limited to histone demethylation . It has been hypothesized that X and Y chromosome homologs will escape X-inactivation in instances where the Y homolog has not lost functional activity and male to female dosage remains balanced [38] . Therefore , it is possible that UTX and UTY have functional overlap in H3K27 demethylase independent gene regulatory processes . A recent publication by Lee et al . characterized heart defects in Utx homozygous embryos [39] . Cell culture experiments suggested that the phenotype resulted from H3K27 demethylase activity . Utx hemizygotes were reported to have a wide range of abnormalities , but it was not clear if any phenotypes overlap with the Utx homozygotes as no comparative data were illustrated . Given that Uty remained intact in these studies , it was not possible to conclude definitively whether Utx demethylase activity was essential for early embryonic development . Furthermore , it is not known whether mouse UTY is capable of H3K27 demethylation . The classification of UTY as having no demethylase activity is based on in vitro assays only . The possibility of in vivo demethylase activity due to other co-factors remains a possibility . Also , mouse UTY has considerable sequence divergence from human UTY . The two proteins are 75% identical overall , and 95% identical in the JmjC demethylase domain . Thus , it is possible that mouse UTY has retained demethylase activity . In our study , we have generated mouse mutations in both Utx and Uty . Hemizygous Utx mutant male mice ( XUtx− Y+ ) were runted at birth with only a small number surviving to adulthood . In contrast , Utx homozygous females ( XUtx− XUtx− ) had severe phenotypes mid-gestation , with developmental delay , neural tube closure , yolk sac , and heart defects . Unlike homozygotes , Utx hemizygotes lack mid-gestational cardiovascular defects and are recovered in Mendelian frequencies at E18 . 5 . Furthermore , compound hemizygous male embryos ( XUtx− YUty− ) carrying mutations of both Utx and Uty phenocopy the Utx homozygotes . Thus , the disparity in hemizygous and homozygous Utx phenotypes is due to compensation by Uty in the hemizygous male embryos . We have utilized an in vivo H3K27 demethylation assay to demonstrate that mouse UTY is not capable of H3K27 demethylation . Additionally , cell culture data indicate UTX and UTY may function in gene activation as both proteins associate with the H3K4 methyl-transferase complex , the BRG1 chromatin remodeler , as well as heart transcription factors . Our results implicate a crucial H3K27 demethylase independent function for UTX and UTY in mouse embryonic development . This is the first ascribed function for UTY , and the first example of developmental redundancy for X and Y chromosome homologous genes . Notably , our data suggest the H3K27 demethylase activity of UTX is not essential for embryonic viability .
We developed mutant mouse lines to assess the contribution of UTX H3K27 demethylase function in mouse development . Two alleles for Utx were obtained from public resources . The BayGenomics gene trap line Kdm6aGt ( RRA094 ) Byg is designated as XUtxGT1 ( Figure 1A ) . RT-PCR and PCR genotyping verified the identity of this allele in both ES cells and mutant mice ( Figures S1 and S2A–S2C ) . Additionally , we obtained the EUCOMM Kdm6a knockout line ( project 26585 , Kdm6atm1a ( EUCOMM ) Wtsi ) , designated as XUtxGT2fl , which inserts a gene trap in intron 2 along with a floxed 3rd exon ( Figure 1A ) . Southern blotting and PCR genotyping verified the identity of this allele ( Figures S1 and S2D–S2F ) . Notably , quantitative RT-PCR comparison of tail RNA from XUtxGT1 YUty+ versus XUtxGT2fl YUty+ mice demonstrated that Utx gene trap 1 is more effective than gene trap 2 ( a 96% reduction compared to a 61% reduction in Figures S2C and S2F ) . Because XUtxGT2fl demonstrated incomplete trapping , the 3rd exon was deleted with Cre recombinase to establish XUtxGT2Δ ( containing both the gene trap and deleted 3rd exon , Figure 1A ) . Deletion of the third Utx exon produces a frameshift and introduction of a translational stop codon when Utx is spliced from exon 2 to exon 4 . XUtxGT1 and XUtxΔ are null alleles as UTX protein was eliminated in western blotting of these embryonic lysates ( Figure 1B , 1C ) . Consistent with RT-PCR data , XUtxGT2fl exhibits a reduction but not absence of UTX protein ( Figure 1D ) . Heterozygous Utx female mice were crossed to wild type male mice to produce hemizygous Utx mutant males . At weaning , the hemizygous XUtxGT1 YUty+ , XUtxGT2Δ YUty+ , and XUtxGT2fl YUty+ mice all exhibited reductions of 68% , 83% , and 55% respectively from the expected genotype frequencies based on these crosses , yet expected genotype frequencies were observed at embryonic day E18 . 5 ( Table 1 ) . At E18 . 5 , most of the hemizygous Utx males appeared phenotypically normal; however a small percentage of the fetuses exhibited exencephaly . At birth , the hemizygous Utx males were small and exhibited a failure to thrive phenotype . Those males that survived through this phenocritical phase reached adulthood and were fertile . Hemizygous Utx mutant males were runted compared to wild type littermates and remained smaller than controls throughout their lifespan ( Figure 2A , 2B ) . Backcross of the Utx allele onto a C57BL/6J or 129/SvJ background affected postnatal viability , but hemizygous Utx male embryos were still readily obtained at E18 . 5 ( Table S1 ) . Human UTY lacks demethylase activity based on in vitro assays , so we hypothesized that XUtx− XUtx− homozygous females will phenocopy XUtx− YUty+ hemizygous males in demethylase dependent function ( UTX specific ) , but may demonstrate a more severe phenotype in demethylase independent roles . Homozygous XUtxGT1 XUtxGT1 and XUtxGT2Δ XUtxGT2Δ females were never observed at weaning or embryonic day E18 . 5 ( Table 1 ) , but were observed at expected genotype frequencies at E10 . 5 . However , these embryos were dead and resorbed by E12 . 5 ( Table 1 ) . Notably , at E10 . 5 all homozygous XUtxGT1 XUtxGT1 and XUtxGT2Δ XUtxGT2Δ females were smaller in size and had open neural tubes in the midbrain region ( Figure 3A-ii , iii , vi , vii ) . Variation in severity of the Utx homozygous phenotypes was observed in mutant embryos , ranging from medium sized with typical E10 . 5 features ( Figure 3A-ii , vi ) to much smaller embryos resembling the E9 . 5 timepoint ( Figure 3A-iii , vii ) . The XUtxGT1 and XUtxGT2Δ alleles failed to complement , as trans-heterozygous XUtxGT1 XUtxGT2Δ female embryos resembled individual homozygous alleles ( Figure 3A-viii ) . Hemizygous XUtxGT1 YUty+ male embryos appeared phenotypically normal at E10 . 5 ( Figure 3A-iv ) . Homozygous XUtxGT2fl XUtxGT2fl females exhibited a slight reduction in phenotypic severity; about half of the mutant embryos had open neural tubes and some survival to E12 . 5 ( Table 1 ) . To distinguish between embryonic and extraembryonic contribution of UTX towards the homozygous phenotype , we crossed the Sox2Cre transgene into the Utxfl background . In this cross , paternally inherited Sox2Cre expression will drive Utx deletion specifically in embryonic tissue [40] . No XUtxfl XUtxfl , Sox2Cre female embryos were recovered at E18 . 5 , whereas XUtxfl YUty+ , Sox2Cre male embryos were recovered at expected frequencies ( Table S2 ) . At E10 . 5 , XUtxfl XUtxfl , Sox2Cre embryos produced phenotypes largely identical to Utx homozygotes . In summary , Utx homozygous females demonstrate a significantly more severe embryonic phenotype in comparison to Utx hemizygous males . Mid-gestational lethality is typically associated with defective cardiovascular development . Accordingly , we observed both heart and yolk sac vasculature/hematopoietic phenotypes in Utx homozygotes . Utx homozygous mutant hearts were small and underdeveloped , and more severe embryos exhibited peri-cardial edema ( Figure 3A-ii , iii , vi , vii ) . The yolk sac vasculature of Utx homozygotes was pale with a reduction in the amount of vascular blood ( Figure 3B-ii ) . In more severe examples , homozygous yolk sacs were completely pale with an unremodeled vascular plexus ( Figure 3B-iii ) . Thus , abnormal cardiovascular function may be a source of lethality and developmental delay in Utx homozygous mutant embryos . The most likely explanation for the disparity between Utx hemizygotes and homozygotes is that UTY can compensate for the loss of UTX in embryonic development . We tested Utx and Uty expression in embryonic development to assess any overlap in expression patterns . Utx expression was initially gauged utilizing the B-galactosidase reporter in XUtx+ XUtxGT1 and XUtxGT1 XUtxGT1 whole mount E10 . 5 embryos . Utx was expressed at lower levels throughout the E10 . 5 embryo with a particular enrichment in the neural tube and otic placode ( Figure S3A-ii , iii , iv ) . In situ hybridization for both Utx and Uty demonstrated similar expression patterns characterized by widespread low-level expression with particular enrichment in the neural tube ( Figure S3B-ii , iii , v , vi ) . Our analysis of publicly available RNA-seq data sets [41] , [42] revealed similar low-levels of expression for Utx and Uty . To determine whether UTY can compensate for the loss of UTX , we obtained the Welcome Trust Sanger Institute gene trap line UtyGt ( XS0378 ) Wtsi , designated as YUtyGT ( Figure 4A ) . This line , inserted in intron 4 , traps the Uty transcript in a similar position of the coding sequence as the Utx alleles ( compare to Figure 1A ) . This gene trap line was verified by RT-PCR in ES cells and subsequent mice ( Figures S1 and S2G ) , and it achieved a 99% reduction in Uty expression from XUtx+ YUtyGT mouse tail RNA ( Figure 4B ) . Hemizygous Uty mutant males , XUtx+ YUtyGT , were viable and fertile ( Table 1 ) . However , no compound hemizygous XUtxGT1 YUtyGT and XUtxGT2Δ YUtyGT embryos were recovered at E18 . 5 ( Table 1 ) . At E10 . 5 , expected genotype frequencies of XUtxGT1 YUtyGT and XUtxGT2Δ YUtyGT males were observed , but these embryos phenocopied the developmental delay , neural tube closure , cardiac , and yolk sac defects observed in Utx homozygous embryos ( Figure 4C-iii , iv ) . We performed a more detailed phenotypic assessment of Utx and Uty mutant hearts to scrutinize the extent of phenotypic overlap between XUtx− YUty+ , XUtx− XUtx− , and XUtx− YUty− embryos . Analysis of cardiac development in similar sized E10 . 5 embryos ( Figure 5A-i , ii , iii , iv ) revealed that Utx homozygotes and Utx/Uty compound hemizygotes failed to complete heart looping ( Figure 5A-vi , viii ) , whereas Utx heterozygotes and hemizygotes were phenotypically normal ( Figure 5A-v , vii ) . Additionally , homozygotes and compound hemizygotes had smaller hearts with a lack of constriction between the left and right ventricles . Sectioning of E10 . 5 hearts confirmed that Utx homozygotes and Utx/Uty compound hemizygotes have small hearts with a reduction in ventricular myocardial trabeculation and little or no initiation of interventricular septum formation ( Figure 5B-ii , iv ) . The outer ventricular wall of these embryos is much thinner , and the overall number of cardiomyocytes and myocardial structure is severely deficient ( Figure 5C-ii , iv ) . In summary , while mid-gestational hearts appear normal in XUtx− YUty+ hemizygous males , XUtx− XUtx−homozygous females and XUtx− YUty− compound hemizygous males display identical deficiencies in cardiac development . Therefore , UTY compensates for the loss of UTX in hemizygous Utx mutant males , rescuing mid-gestational cardiac phenotypes . UTX and UTY have redundant function in embryonic development , but it is not known whether mouse UTY is capable of H3K27 demethylation . Two independent publications demonstrated that human UTY has no catalytic activity in H3K27 demethylation in vitro [26] , [29] . It is possible that human UTY ( and not mouse UTY ) has accumulated a specific polymorphism rendering it demethylase deficient . Additionally , in vitro assays remove UTY from its natural cellular context and may lack co-factors required to promote H3K27 demethylation . Therefore , we utilized an intracellular , in vivo demethylation assay , whereby HEK293T cells transiently over-expressing the UTX carboxy-terminus ( encoding the JmjC and surrounding domains essential for proper structure and function ) exhibit a reduction in H3K27me3 immunofluorescence levels [43] . In our assay , wild type and mutant constructs were expressed at similar levels ( Figure S4A ) , and individual cells expressing similar , medium-high expression levels of each construct were selected for analysis ( Figure 6 ) . Expression of Flag-tagged human and mouse UTX demethylated H3K27me3 and H3K27me2 , while a mutation known to disrupt activity ( H1146A ) was unable to demethylate H3K27 ( Figure 6A and 6B , Figure S5A ) . Human UTX expression had no effect on other histone modifications we tested , such as H3K4me2 ( Figure S5B ) . In contrast , neither human nor mouse UTY were capable of demethylating H3K27me3 and H3K27me2 ( Figure 6A and Figure S5A ) . Cells expressing medium-to-high levels of UTY ( N>100 ) never exhibited a reduction in H3K27me3 levels relative to nearby untransfected controls . Our previous structural analysis of human UTX [43] , combined with sequence alignments ( Figure 6C and Figure S6 ) , suggested several amino acid substitutions in human and mouse UTY sequences might make them catalytically inactive . We introduced these mutations into the human UTX C-terminal fragment ( Y1135C , T1141I , SNR1025NKS , G1172D/G1191S , I1267P , I1267V , and H1329P ) , and examined their effects on the in vivo demethylation activity . Of all the mutations tested , only the Y1135C and T1143I mutations completely abolished the ability of UTX to demethylate H3K27 ( Figure 6B , 6D ) . Complete loss of activity was similarly caused by mutations of the corresponding residues in JMJD3 ( Y1377C and T1385I , Figure 6D ) . All qualitative data was also confirmed by immunofluorescence quantification ( Figure S4B ) . Y1135 is conserved throughout all H3K27 demethylases ( Figure S7 ) , and in the crystal structure [43] , it interacts with two of the three methyl groups of the H3K27me3 side chain , as well as N-oxalylglycine ( NOG; an analog of the cofactor alpha-ketoglutarate ) ( Figure 6E ) . The smaller C947 side chain of mouse UTY would not effectively maintain either interaction . T1143 is conserved throughout H3K27 , H3K9 , and H3K36 demethylases ( Figure S8 ) , and also interacts with NOG ( Figure 6E ) . Its replacement with bulky isoleucine not only removes the hydroxyl group for interaction with alpha-ketoglutarate , but also may sterically hinder its binding . These observations are consistent with the fact that no H3K27 demethylation activity has been detected for mouse UTY , and we therefore conclude that the catalytic domain of mouse UTY has crucial amino acid replacements that render the protein incapable of H3K27 demethylation . On the other hand , we failed to identify why human UTY is catalytically inactive . Notably , restoring the 2 crucial mouse UTY polymorphisms ( M-UTY C947Y , I955T ) failed to recover H3K27 demethylase activity ( Figure 6B ) . These data suggest that unidentified structural elements in the UTY C-terminal region are also responsible for the lack of H3K27 demethylase activity . Although human and mouse UTY have lost the ability to demethylate H3K27 , they retain considerable sequence similarity with UTX , suggesting a conserved function . To gain more insight into the overlap in UTX and UTY activities , we performed a biochemical analysis of tagged constructs to determine if UTX and UTY can associate in common protein complexes . Co-transfection of Flag tagged UTX or UTY with HA-UTX followed by immunoprecipitation demonstrates that UTX can form a multimeric complex with itself and UTY ( Figure 7A ) . UTX associates with a H3K4 methyl-transferase complex containing MLL3 , MLL4 , PTIP , ASH2L , RBBP5 , PA-1 , and WDR5 [23] , [44] . To examine incorporation into this complex , we performed immunoprecipitations with Flag tagged UTX and UTY constructs . Both UTX and UTY were capable of associating with RBBP5 ( Figure 7B ) . Thus , UTX and UTY are incorporated into common protein complexes . To identify common gene targets of UTX and UTY mediated regulation we generated E10 . 5 mouse embryonic fibroblast ( MEF ) cell lines containing mutations in Utx and Uty ( alleles XUtxGT2Δ and YUtyGT ) . The gene traps in these MEFs efficiently trapped Utx and Uty transcripts ( Figure 7C ) . These MEFs did not demonstrate differences in levels of global H3K27me3 ( Figure S9A ) . Genome-wide UTX promoter occupancy has been mapped in fibroblasts [30] . Therefore , we screened our mutant MEFs for misregulated genes affected by the loss of both Utx and Uty that had been documented as direct UTX targets . The FNBP1 promoter is bound by UTX [30] . We verified UTX and UTY binding to the Fnbp1 promoter by ChIP ( Figure S9B and S9C ) . Fnbp1 expression was reduced to 68% of WT levels in XUtx− YUty+ MEFs , but was further compromised to 42% in XUtx− XUtx− lines and 48% in XUtx− YUty− MEFs in which all Utx and Uty activity was lost ( Figure 7C ) . Analysis of E12 . 5 MEFs of a secondary allele ( XUtxGT2fl ) also demonstrated diminished Fnbp1 expression in both XUtx− XUtx− and XUtx− YUty− MEFs ( Figure 7D ) . Therefore , Fnbp1 expression is positively regulated by both UTX and UTY . To examine the role of UTX and UTY in Fnbp1 regulation , we performed H3K27me3 ChIP on E12 . 5 XUtx+ YUty+ or XUtx− XUtx− MEFs ( Figure 7E ) . Quantitative PCR for an intergenic region served as a negative control , while HoxB1 served as a positive control for H3K27me3 . Quantitative PCR demonstrated that the Fnbp1 promoter has relatively low levels of H3K27me3 with no additional accumulation in XUtx− XUtx− MEFs ( Figure 7E ) . Alternatively , H3K4me3 significantly accumulated at the Fnbp1 promoter ( Figure 7F ) . Notably , a loss of Fnbp1 H3K4me3 was observed in XUtx− XUtx− MEFs ( Figure 7F ) . Therefore , UTX and UTY appear to function in Fnbp1 activation by regulating promoter H3K4 methylation rather than H3K27 demethylation . It has been documented that UTX can associate with heart transcription factors and with the SWI/SNF chromatin remodeler , BRG1 [39] . It has been hypothesized that UTX association with these factors mediates H3K27 demethylase dependent and demethylase independent induction of the cardiomyocyte specification program . As UTX and UTY have redundant demethylase independent function in embryonic development , we examined whether UTY can also associate with these proteins . Co-transfection of Myc-UTY with Flag-BRG1 followed by immunoprecipitation demonstrated that UTY associates with BRG1 ( Figure 8A ) . Myc-UTY also co-immunoprecipitated with Flag-NKX2–5 , Flag-TBX5 , and Flag-SRF ( Figure 8B and Figure S10A ) . Thus , UTY can form the same protein complexes as UTX with respect to BRG1 and heart transcription factors . To examine function of UTY in directing activation of downstream heart transcription factor targets , we assessed the regulation of one previously characterized target , atrial natriuretic factor ( ANF ) [39] . Co-transfection of NKX2–5 with a ANF promoter-Luciferase reporter construct demonstrated a significant upregulation in expression off the ANF promoter ( Figure 8C ) . The reporter expression was significantly enhanced when NKX2–5 was co-transfected with UTY ( Figure 8C ) . The level of ANF reporter transcriptional enhancement was relatively weaker with UTY as compared to UTX , but this is most likely due to a reduction in the transfection efficiency of full-length UTY relative to UTX ( as demonstrated in Figure 7A and 7B ) . UTY also significantly enhanced the ANF reporter response to TBX5 ( Figure S10B ) . Finally , ANF expression was significantly affected in the hearts of only XUtx− XUtx− and XUtx− YUty− embryos ( with 52% and 57% level of expression respective to XUtx+ YUty+ controls , Figure 8D ) . XUtx− YUty+ hemizygotes only had a moderate loss of ANF expression ( 76% expression level respective to XUtx+ YUty+ ) that was not statistically significant from wild type controls due to the variability in ANF expression . In summary , both UTX and UTY can associate with heart transcription factors to modulate expression of downstream targets .
We have undertaken a rigorous genetic analysis contrasting UTX and UTY function in mouse embryonic development . In alignment with current literature , Utx homozygous females are lethal in mid-gestation with a block in cardiac development [39] . We now demonstrate that Utx hemizygous mutant males are viable at late embryonic timepoints in expected Mendelian frequencies . In fact , approximately 25% are capable of reaching adulthood . Our comprehensive phenotypic analysis of Utx hemizygous males illustrates that these embryos are phenotypically normal at mid-gestation and lack the cardiovascular dysfunction of Utx homozygous females . This stark phenotypic disparity suggests that UTY may compensate for the loss of UTX in the male embryo . Compound hemizygous Utx/Uty mutant male embryos phenocopy the cardiovascular and gross developmental delay of homozygous females , proving that UTX and UTY have redundant function in embryonic development . As we have demonstrated that mouse UTY lacks H3K27 demethylase activity in vivo , the overlap in embryonic UTX and UTY function is due to H3K27 demethylase independent activity . Given the widespread developmental delay and pleiotropy , it is difficult to assess the primary defect and tissue ( s ) responsible for UTX and UTY redundancy . The presence of functional UTY in Utx hemizygous males is not capable of preventing peri-natal runting and lethality , suggesting that UTX and UTY are not completely overlapping in activity . These later phenotypes could be due to H3K27 demethylase dependent activity of UTX . Furthermore , the lack of phenotype in Uty hemizygotes demonstrates the absence of any essential UTY specific function in mouse development . The UTY Jumonji-C domain has maintained high conservation in the absence of catalytic H3K27 demethylase activity . JMJD3 mediated regulation of lymphocyte Th1 response requires an intact Jumonji-C domain , but is also not dependent on H3K27 demethylation [37] . Therefore , this domain may be an essential structural protein component , a protein binding domain , or a domain that may demethylate non-histone substrates . UTX and UTY can associate in a common protein complex and can both interact with RBBP5 of the H3K4 methyl-transferase complex . UTX , UTY , and JMJD3 all associate with H3K4 methyl-transferase complexes from multiple mouse and human cell types [23] , [25] , [44] , [45] . The Fnbp1 promoter is bound by UTX , and gene expression is positively regulated by both UTX and UTY in MEFs . Based on our histone profiling at this locus , UTX and UTY affect the deposition of H3K4 methylation , not H3K27me3 demethylation . Therefore , the common UTX/UTY pathway in embryonic development may involve gene activation rather than removal of gene repression . JMJD3 has been linked more directly to transcriptional activation as the protein complexes with and facilitates factors involved in transcriptional elongation [46] . One cardiac target of UTX regulation , atrial natriuretic factor ( ANF ) , was misregulated in ES cell differentiation [39] . Cell culture experiments suggest that ANF may be a target of both H3K27 demethylase dependent and demethylase independent regulation; however , this study could not distinguish UTX versus UTY function in ES cell differentiation . Both UTX and UTY affect the transcriptional response of an exogenous ANF reporter in the presence of heart specific transcription factors , suggesting that UTX and UTY can operate more directly by aiding in transcriptional activation of this gene rather than altering chromatin structure . Consistently , ANF expression was affected in XUtx− XUtx− and XUtx− YUty− embryonic hearts . UTX and UTY can both associate with the SWI/SNF chromatin remodeler BRG1 , which has been hypothesized to mediate histone demethylase independent gene regulation , but the relevance and mechanism of this interaction is not known . Drosophila UTX associates with BRM ( orthologous to BRG1 ) and CBP ( a H3K27 acetyl-transferase ) , and the coupling of H3K27 demethylation with H3K27 acetylation may be essential for switching from a silent to active state [47] . Female cells are subject to gene silencing of one X-chromosome ( X-inactivation ) to balance gene dosage with males . Theory on establishing X-inactivation for X-Y chromosome homologs hypothesizes that the initial entry step is loss of function or expression of the Y homolog to create dosage imbalance [38] . This prediction also dictates that conservation of X-Y homolog function will maintain gene dosage between sexes , and the female X-homolog will not experience pressure to inactivate . Utx and Uty represent a unique paradox to this untested theory; UTY has lost demethylation activity yet Utx escapes X-inactivation . We now demonstrate that UTX and UTY have retained embryonic redundancy , verifying the presumed correlations between X-inactivation escape and functional dosage balance . Zfx , Sox3 , and Amelx represent unbalanced X-chromosome genes; they have similar hemizygous and homozygous mutant phenotypes indicating that the Y chromosome homologs have lost redundant function [48] , [49] , [50] , [51] , [52] , [53] , [54] . Zfx and Sox3 are inactivated , while the Amelx inactivation status is unknown [55] , [56] . Of all mouse X and Y chromosome homologs , only Utx , Kdm5c , and Eif2s3x are known to escape X-chromosome inactivation [28] , [56] , [57] , [58] . Interestingly , both KDM5C ( SMCX ) and its Y chromosome homolog , KDM5D ( SMCY ) have retained catalytic activity in demethylation of H3K4 di and tri-methyl residues [59] , [60] , [61] , [62] . In contrast to Utx X-chromosome escape driven by demethylation independent redundancy , Kdm5c may escape inactivation due to demethylation dependent redundancy . Our study is the first to demonstrate that an X-Y homologous pair that escapes X inactivation maintains functional conservation , and this escape may stem from an evolutionary benefit to maintain UTY demethylation independent function . H3K27 demethylases are hypothesized to function in early developmental activation of “bivalent” PRC2 targets by coordinating H3K27 demethylation with H3K4 methylation . The H3K27 demethylation dependent phenotype ( UTX specific ) of Utx hemizygotes is not apparent until birth . The UTX H3K27 demethylase activity is dispensable for function in C . elegans [63] . Remarkably , the mammalian embryo , having numerous examples of H3K27me3 repression in early development , can survive to term without UTX histone demethylation . It is possible that there is further redundancy between UTX and JMJD3 . JMJD3 mutant mice are not well characterized , but have been reported to be peri-natal lethal with distinct features in comparison to Utx hemizygotes [32] . Therefore , it is likely that JMJD3 has distinct targets in development . Overall , the earliest H3K27 demethylation dependent phenotypes for all members of this gene family do not manifest until late embryonic development . This timepoint is much later than the converse early embryonic phenotypes from mutations in the H3K27 methyl-transferase complex [7] , . Thus , there appears to be a lack of interplay between H3K27 methylation and demethylation in gene regulation , and the early embryonic removal of H3K27me3 from PRC2 mediated processes ( such as ES cell differentiation , reactivation of the inactive X-chromosome , or establishing autosomal imprinting ) may involve other mechanisms such as histone turnover or chromatin remodeling . H3K27 demethylases may certainly have crucial roles in the specification of progenitor cell populations of organ systems essential in peri-natal or postnatal viability , and genetic model systems will best assess the functional impact that H3K27 demethylation plays in these processes .
HEK293T were maintained in DMEM supplemented with Glutamine , Pen-Strep , and 10%FBS . Flag-Human UTX ( Plasmid #17438 ) and UTY ( Plasmid #17439 ) were obtained through Addgene [29] . The N-terminus of H-UTX and H-UTY were deleted with QuikChange Lightning ( Agilent ) as directed producing H-UTX C-terminus 880–1401 ( Genbank: NP_066963 . 2 ) and H-UTY C-terminus 827–1343 ( Genbank: NP_009056 . 3 , an N-terminal His tag was also incorporated into both constructs ) . Site directed mutagenesis was performed via QuikChange Lightning ( Agilent ) as directed to produce point mutations . The mouse UTX C-terminus ( 880–1401 ) deviated from Human UTX at 2 residues , R1073K and S1263N ( According to the Sanger Vega server , the primary Utx transcript Kdm6a-001 encodes for Genbank: CAM27157 , we also detected this transcript in E14 ES cell RT-PCR ) . These changes were created in the H-UTX C-terminus to generate the M-UTX construct . The Flag-tagged mouse UTY C-terminus ( 692–1212 , Genbank: NP_033510 . 2 ) was subcloned by RT-PCR of E14 ES cell RNA and introduced into the same vector as the other UTX constructs ( PCS2+MT backbone ) . HA tagged H-UTX was obtained through Addgene [24] . Flag tagged mouse JMJD3 was generously provided by Burgold et al . [27] . Flag tagged BRG1 was obtained through Addgene ( Plasmid #19143 ) . Flag tagged NKX2–5 ( Plasmid #32969 ) , TBX5 ( Plasmid #32968 ) , and SRF ( Plasmid #32971 ) were obtained through Addgene and recombined into DEST26 ( Invitrogen ) . Flag tagged NKX2–5 was also generously provided by Benoit Bruneau [64] . Transfection of HEK293T was accomplished with Lipofectamine 2000 as directed ( Invitrogen ) . Lipid complexes were removed 24 hours post-transfection , and analysis was performed after 48 hours total . Fixation , extraction , and immunofluorescence were performed as described [65] . Immunofluorescence antibodies include anti-Flag ( Sigma F3165 , 1∶500 ) , anti-H3K27me3 ( Millipore 07-449 , 1∶500 ) , anti-H3K27me2 ( Millipore 07-452 , 1∶500 ) , and anti-H3K4me2 ( Millipore 07-030 , 1∶500 ) . Cells were imaged with Zeiss axiovision software . Image stacks were deconvolved and z-projected . Quantification of H3K27me3 immunofluorescence was performed on deconvolved z-projected stacks ( with no pixel saturation in images ) using ImageJ software ( NIH ) . The average mean H3K27me3 signal was calculated for untransfected and transfected cells in a given image . For each image , the relative % H3K27me3 was determined , and the average relative % H3K27me3 was calculated for >15 images per construct . For western blotting , nuclear lysates were prepared according to Invitrogen's nuclear extraction protocol . Immunoprecipitations were carried out with 50 µl Flag beads ( Sigma A2220 ) in buffer A as described , using 500 µg ( UTY-UTX , RBBP5 , BRG1 , TBX5 , SRF associations ) or 1 mg ( UTY-NKX2–5 association ) of lysate [44] . Immunoprecipitation reactions were boiled off beads and run with 10% input on an 8% SDS-PAGE gel . Histone extractions were prepared as described [66] . Western blotting was performed as described [67] with anti-Flag ( Cell Signaling 2368 , 1∶4000 ) , anti-RBBP5 ( Bethyl Labs A300-109A , 1∶5000 ) , anti-HA ( Roche 11867423001 , 1∶10000 ) , anti-Myc ( Abcam ab9132 , 1∶5000 ) , anti-H3K27me3 ( Millipore 07-449 , 1∶2000 ) , anti-H3 ( Millipore 06-755 , 1∶5000 ) , and anti-UTX [24] . E10 . 5 and E12 . 5 MEFs were generated by removal of the head and interior organs of respective embryos . The remaining body was passed through a 20G needle 6× and plated in DMEM supplemented with Glutamine , Pen-Strep , and 15%FBS . After 3 passages , RNA was isolated with Trizol , and cleaned with an RNeasy kit ( Quiagen ) . RNA from 3 distinct WT XUtx+ YUty+ MEF lines was compared to 3 XUtxGT2Δ XUtxGT2Δ lines on an Illumina bead array ( University of Tennessee Health Science Center ) . All genes significantly decreased in XUtxGT2Δ XUtxGT2Δ MEFs were cross-referenced to the list of UTX bound promoters in human fibroblasts [30] . These genes were analyzed by qRT-PCR ( Bio-Rad SsoFast EvaGreen , CFX96 real time system ) in all mutant MEF combinations to identify UTY regulated genes . ChIP was performed on these MEFs according to Rahl et al . [68] . MEFs ( 5×106 cells ) were sonicated by a Branson Sonifier at 15% duty cycle ( 0 . 7 s on 0 . 3 s off ) . ChIP was performed with anti-H3K27me3 ( Millipore 07-449 , 10 µl ) , anti-H3K4me3 ( Abcam ab1012 , 5 µl ) , anti-Myc ( Abcam ab9132 , 10 µl ) , anti-UTX ( Santa Cruz H-300 , 50 µl ) , or Rabbit IgG ( Sigma , I5006 ) and qPCR was performed as described above . We received the ANF promoter-Luciferase reporter construct from Benoit Bruneau [69] . This construct was co-transfected in the presence of NKX2–5 or TBX5 with or without UTX or UTY . Luciferase activity was measured using the Promega Dual Luciferase Reporter Assay System on the Promega Glomax Multi Detection System . All readings were normalized to a Renilla Luciferase control that was co-transfected with all samples . Kdm6aGt ( RRA094 ) Byg ( XUtxGT1 ) , Kdm6atm1a ( EUCOMM ) Wtsi ( XUtxGT2fl ) , and UtyGt ( XS0378 ) Wtsi ( YUtyGT ) ES cells were obtained from BayGenomics ( through MMRRC ) , EUCOMM , and SIGTR ( through MMRRC ) respectively . All ES cells were injected into C57BL/6J host blastocysts for chimera generation . Chimeras were crossed to CD1 to assess germline transmission , and were maintained on either a mixed CD1 background or were backcrossed to 129/SvJ or C57BL/6J . Sox2Cre and RosaFlp transgenes were obtained from The Jackson Laboratory [40] . The VasaCre transgene was developed by Gallardo et al . [70] . All mouse experimental procedures were approved by the University of North Carolina Institutional Animal Care and Use Committee . Utx homozygous data was generated either by crosses between Utx hemizygous males and Utx heterozygous females , or by crosses between XUtxGT2fl YUty+ VasaCre males and XUtx+ XUtxGT2Δ heterozygous females . XUtxGT2fl YUty+ VasaCre males were utilized because of an initial difficulty in generating XUtxGT2Δ YUty+ males and due to the efficient and specific activity of VasaCre in the male germline [70] . Utx hemizygous phenotypic data was developed from the previously mentioned homozygous crosses or through crosses between a WT male and heterozygous Utx female . Compound hemizygous Utx/Uty embryos were generated by crossing heterozygous Utx females with hemizygous Uty males . Embryos were PCR genotyped from yolk sac samples for Utx and were sexed by a PCR genotyping scheme to distinguish Utx from Uty . All primer sequences are available upon request . Histology samples , in situ hybridization , and LacZ staining were performed as described [71] . In situ hybridization probes were generated to be identical to previous literature [72] . | Trimethylation at Lysine 27 of histone H3 ( H3K27me3 ) establishes a repressive chromatin state in silencing an array of crucial developmental genes . Polycomb repressive complex 2 ( PRC2 ) catalyzes this precise posttranslational modification and is required in several critical aspects of development including Hox gene repression , gastrulation , X-chromosome inactivation , mono-allelic gene expression and imprinting , stem cell maintenance , and oncogenesis . Removal of H3K27 trimethylation has been proposed to be a mechanistic switch to activate large sets of genes in differentiating cells . Mouse Utx is an X-linked H3K27 demethylase that is essential for embryonic development . We now demonstrate that Uty , the Y-chromosome homolog of Utx , has overlapping redundancy with Utx in embryonic development . Mouse UTY has a polymorphism in the JmjC demethylase domain that renders the protein incapable of H3K27 demethylation . Therefore , the overlapping function of UTX and UTY in embryonic development is due to H3K27 demethylase independent mechanism . Moreover , the presence of UTY allows UTX-deficient mouse embryos to survive until birth . Thus , UTX H3K27 demethylation is not essential for embryonic viability . These intriguing results raise new questions on how H3K27me3 repression is removed in the early embryo . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetic",
"mutation",
"cell",
"differentiation",
"histone",
"modification",
"animal",
"models",
"developmental",
"biology",
"model",
"organisms",
"epigenetics",
"embryology",
"genomic",
"imprinting",
"gene",
"expression",
"biology",
"mouse",
"molecular",
"biology",
"genetics",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | UTX and UTY Demonstrate Histone Demethylase-Independent Function in Mouse Embryonic Development |
When a perturbation is applied in a sensorimotor transformation task , subjects can adapt and maintain performance by either relying on sensory feedback , or , in the absence of such feedback , on information provided by rewards . For example , in a classical rotation task where movement endpoints must be rotated to reach a fixed target , human subjects can successfully adapt their reaching movements solely on the basis of binary rewards , although this proves much more difficult than with visual feedback . Here , we investigate such a reward-driven sensorimotor adaptation process in a minimal computational model of the task . The key assumption of the model is that synaptic plasticity is gated by the reward . We study how the learning dynamics depend on the target size , the movement variability , the rotation angle and the number of targets . We show that when the movement is perturbed for multiple targets , the adaptation process for the different targets can interfere destructively or constructively depending on the similarities between the sensory stimuli ( the targets ) and the overlap in their neuronal representations . Destructive interferences can result in a drastic slowdown of the adaptation . As a result of interference , the time to adapt varies non-linearly with the number of targets . Our analysis shows that these interferences are weaker if the reward varies smoothly with the subject's performance instead of being binary . We demonstrate how shaping the reward or shaping the task can accelerate the adaptation dramatically by reducing the destructive interferences . We argue that experimentally investigating the dynamics of reward-driven sensorimotor adaptation for more than one sensory stimulus can shed light on the underlying learning rules .
Transformations that map sensory inputs to motor commands are referred to as sensorimotor mappings [1] . While sensorimotor mappings are already formed at early stages of development [2] , they are subject to modifications , since the brain , the body and/or the environment are constantly changing . Plasticity in sensorimotor mappings has been extensively studied in situations where subjects receive sensory feedback during the task , allowing them to correct their motor actions and to adapt to the induced perturbation . These include visuomotor rotation [3] , reaching movements under forcefields [4] , adaptation in a smooth pursuit eye movements [5] , prism adaptation [6] , and pitch perturbation in songbirds [7] and in humans [8] . Although these studies involve different sensory modalities and different effectors , they are similar in the sense that they all have sensory goals ( targets ) and a motor gesture is made to reach the target . They consist of three phases namely a standard phase , in which subjects perform the task under regular conditions followed by an adaptation phase , where subjects perform the same task under the perturbed condition and a washout phase during which the perturbation is removed , and the subject readapts toward baseline . Remarkably , in all these three phases , movements display substantial trial to trial variability . Recent theoretical as well as experimental studies suggested that this variability plays a crucial role in sensorimotor learning and adaptation processes [9]–[11] . Another issue concerns the ability of subjects to generalize the adaptation from one context condition to a different context . This has been investigated by testing how subjects perform upon presentation of sensory stimuli that were not present during the adaptation phase [12] , [13] . Generalization is usually good for sensory stimuli that are similar to the one used during adaptation and degrades as the sensory stimuli become different [3] , [14] . Remarkably , subjects can even perform worse than in baseline ( negative generalization ) for sensory stimuli which are very different from those which was presented to the subject during adaptation . This has been observed , for instance , in motor reaching tasks , when the tested stimulus is presented in a direction which is opposite to the adapted direction [4] , [14] . The above mentioned studies implicitly assumed that the neural mechanisms for adaptation are driven by a sensory feedback , which supplies a continuous error signal to the subject . Yet , recent studies show that adaptation is possible even without any sensory feedback , when only a binary reward that informs on a success or a failure of a trial is provided to the subject [15]–[17] . Moreover , recent experimental works suggest that reward based mechanisms also affect the adaptation dynamics in sensorimotor tasks even when a sensory feedback is available [18] , [19] . However , and not surprisingly , adaptation relying solely on rewards at the end of a trial is more difficult than when a sensory feedback on the performance is provided continuously during the task , as adapting with sensory feedback conveys more information regarding errors . For instance , when visual feedback is available in visuomotor rotation tasks , subjects adapt to large perturbation ( e . g . 30 degrees ) in a few dozen trials [3] , [20] , while in the absence of such feedback , but with binary ( success or a failure ) reward feedback , subjects find it notoriously difficult to adapt . Recent studies , nevertheless , have shown that it is possible to adapt to large perturbations relying solely on rewards if the size of the perturbation is slowly increased between rewarded blocks of trials [17] , [21] . The fact that progressively increasing the amount of perturbation makes it possible to adapt , even when the perturbation is large , is reminiscent of the classical shaping strategy [22] . In shaping , the difficulty of the task is increased gradually in order to accelerate learning , or to even make it possible . Although shaping is routinely used in laboratories when training animals to perform complex sensorimotor and cognitive tasks [23]–[25] , it is only in recent years that it started to be explored in a theoretical framework [26]–[28] . What neural mechanisms could be involved in this reward based learning ? Recent experimental evidence [29]–[31] indicates that rewards modulate local synaptic plasticity via global neuromodulatory signals , e . g . dopamine . When combined with the popular idea that synapses are modified according to Hebbian rules , this leads to the hypothesis that reward signals interact with local neuronal activity to modulate synaptic efficacies [32] , [33] . This theoretical paper aims to provide qualitative as well as quantitative insights into the conditions in which sensorimotor adaptation relying solely on rewards can take place . More specifically , we assume that a local learning rule based on the coactivation of pre and postsynaptic neurons is gated by a binary reward signal is the neural basis for modifications of synaptic efficacies [32] , [34] , [35] . We focus here on adaptation to a rotation during reaching movements where subjects are asked to move a cursor on a screen to bring it within a circular target while the cursor trajectory is rotated ( perturbed ) by some angle with respect to the hand trajectory . These perturbation tasks are classically used in behavioral studies of sensorimotor adaptation [3] . We consider a simplified network model of this task where adaptation relies solely on binary rewards [17] . The simplicity of the model allows us to analytically study several aspects of the adaptation dynamics . Combining these results with numerical simulations enables us to investigate the ways in which the learning dynamics depend on the model parameters . The key question is how the dynamics of adaptation are affected when the task involves multiple targets . Four main findings are reported: interferences can occur when adapting to multiple stimuli , interferences can slow down the adaptation dynamics dramatically , this depends on the ( binary , stochastic ) reward , and the slow down can be overcome by using shaping strategies .
We first consider the case where the network has to adapt to a rotation of the cursor when only one target is presented . Figure 2A ( left ) plots the evolution of the error ( see Eq . ( 5 ) ) with the number of trials , hereafter referred to as the learning curve , while the network adapts to an imposed rotation with an angle . On the right panel we plotted for the same parameters the learning curve of the directional error , which takes into account only the direction of the movement . The error is large at the beginning of the process and decreases with the number of trials . Importantly , the dynamics strongly depend on the noise . For a low noise level ( Figure 2A , ) , the error remains large for many trials and learning is slow . When the noise level is higher ( Figure 2B , ) the error declines faster . However , this comes at the cost of increasing the error after learning: the median of this error , called hereafter the final error ( see Materials and Methods ) , is larger when the noise level is larger . Similarly , the probability that the network will perform the task successfully , improves more rapidly with the number of trials for than for , but at very long time it is larger in the latter ( ) than in the former ( ) case . The learning curves plotted in Figure 2A–B were obtained for particular realizations of the noise , . To provide a statistical characterization of these dynamics , we estimated the distributions of the logarithm of the learning duration ( ) over many realizations of the noise ( see Materials and Methods ) . As shown in Figure 2D , this distribution shifts toward longer learning duration as the noise level decreases . Figures 2A and 2C plot the learning curves for and for the same noise level . The learning is substantially faster for but the final error is larger in this case . This is because when the target size is large , a reward might also be delivered for less precise movement , i . e . , for large errors . Figure 2E plots the log learning duration and the final error averaged over realizations vs . the target size: when increasing the target size , the learning duration rapidly decreases , whereas the final error increases . When the noise level or the target size are increased , the dynamics are typically faster because the probability of generating rewarded trials at the beginning of the learning is larger . As this probability increases , the time for the network to generate a rewarded trial decreases , leading to more updates in the connectivity matrix ; hence the probability of the following trials to be rewarded increases further . This argument can be made more quantitative if one considers how the time to get the first reward depends on and . It has a geometrical distribution with a parameter ( see Eq . ( 10 ) ) , which is the probability to get the first reward . Lower values of increase the expectation time to the first reward , and thereby the learning duration . When the noise level is low and the initial error is larger than the target size , the network explores a small region of the two dimensional space and the probability of getting a reward is small . In contrast , for very large noise the target is missed most of the time . The probability therefore varies non-monotonically with the noise level ( Figure 2F ) . The dependency on target size is simpler: increases monotonically with target size , as it is more likely to reach a larger target . What is the learning dynamics when the subject has to perform the task for two targets ? How does learning the task for one of the targets affect learning the other one ? We addressed these questions in numerical simulations , in which two targets were presented at an angular distance , , at consecutive times . Similar results were obtained when the targets were presented in a random order with equal probability . How does the learning duration , i . e . , the time to learn the task for all the presented targets , vary with the number of targets ? We simulated the learning of m targets , whose directions were evenly distributed between and . We took a small target size ( ) , so that up to non-overlapping targets could be considered ( for targets presented on a circle with radius 1 ) . Figure 11A plots the average time to learn the entire task in terms of the total number of target presentations for a fixed noise level and different values of tuning widths . It shows a non-monotonic dependency with the number of targets . This contrasts the monotonically increasing learning duration when targets are learned independently with the same noise level and target size ( dashed line ) .
Models of sensorimotor control and learning frequently assume minimizing a squared error function . This is convenient because of analytical or computational simplicity [13] , [14] . However , it was shown that although these models can be a good approximation they tend to penalize large errors excessively [41] . In contrast , we chose to explore adaptation with a binary reward function , as used in experiments . Our results and predictions stem from the shape of the reward function . Specifically , they do not depend qualitatively on the specific choice of the distance error used , but are based primarily on the fact that the reward function varies sharply with the distance to the target center . The dynamics of the adaptation to more than one target depend on the overlap between the tuning curves of the input neurons . However , the precise shape of the tuning curves is not crucial and the results are unchanged if one replaces the Von Mises function we used here with any other tuning curve function , such as a cosine tuning curve ( see e . g . Eq . 23 ) . As a matter of fact , the results we describe are the outcome of the following: 1 ) the same system is used to learn the task for several targets , leading to interference which depends on the way in which the targets differ physically as well as in their neuronal representation and 2 ) learning the task for one target can deteriorate performance on another target such that the information provided by the reward when attempting to learn the task for it becomes small , thereby delaying the learning . These two properties of the learning process are not specific to the simple model we investigated here . In our model , the latter property stems from the fact that the reward varies sharply with the error . The learning rule we used is part of a general family of gradient-like reinforcement learning rules; i . e . , learning rules that on average form a gradient ascent on the reward function [35]–[37] . In fact , learning with an on-line Gradient Ascent algorithm with a sigmoidal cost function can result in similar effects ( Text S1; Figure S1 ) . It might be claimed that plasticity also occurs when no reward is delivered [42] . Therefore , we also verified that the phenomenology of the model remains qualitatively the same when instead of using a reward function ( unpublished data ) . Note that to avoid a drift of the output vector which occurs when , the synaptic weights must be normalized in this case after each trial . Another extension of our model would be to use a reward prediction error instead of an instantaneous reward; e . g . , by subtracting a running average of the reward from the instantaneous reward . Delayed learning also occurs with this type of learning rule ( results not shown ) . In fact , previous works have argued that this modification does not affect most of the qualitative behavior of the algorithm [32] , [36] . However , it should be noted that in the case of multiple targets , computing the running average of the rewards over all targets is an additional source of interference , as shown recently in [35] . To avoid this , the running average of the reward needs to be monitored for each target separately . We focused on the learning dynamics in a feed-forward network of linear neurons with only two layers . We chose this architecture for the sake of simplicity . However , we verified that similar qualitative behaviors in terms of interference and delayed learning occur in a network model in which an intermediate layer consisting of nonlinear neurons was added , and in which a decoder provides the angle of reach movement instead of a vector ( Text S1 , Figure S2 and unpublished data ) . Note that in the framework of this more complex model , the noise can be unambiguously related to neuronal variability whereas in the simplified two-layer model considered in our paper , the noise is in the decoder . One limitation of our work is that we did not model the trajectory of the movement and/or the muscle activation patterns needed to produce movements [43] . However , we expect that delayed learning and interferences also occur in a more detailed model of movement production , such as the one used in Legenstein et al . [34] . A reward-based component in a sensorimotor task was shown to be involved in adaptation to rotations even when detailed spatial information regarding the error was provided to the subject [18] , [19] . We investigated the ways in which neural possible mechanisms that reinforce successful actions affect adaptation dynamics . This type of reward-based mechanism was also studied in [17] . In this experiment , subjects adapted without visual feedback to a gradually increasing rotation of every 40 trials , up to an rotation . Our modeling results are in line with these experiments ( Figure 4B ) . We thus predict that shaping the reward also accelerates adaptation . Besides demonstrating that adaptation relying on rewards is possible by utilizing a gradual rotation paradigm , the Izawa and Shadmehr [17] results suggested that there is no change in the perceived sensory consequences of the motor commands; i . e . , there should be no change in a “forward model” . Therefore , in [17] adaptation was modeled by an action selection rule . Our model is similar to the latter , as we focused on the reward-based component during adaptation . However , our model differs in that it is value-free , whereas in [17] it involved value-based reinforcement learning . Nevertheless , our model can also account for the experimental results reported in [17] for one target ( see Text S1 , Figure S3 ) . Moreover , it allowed us to investigate the generalization curve and possible interference during adaptation for multiple targets . The key finding of this theoretical work is that if a reward-modulated learning rule underlies adaptation , interferences are likely to be observed when learning multiple targets with a binary reward . It would be valuable to explore whether such effects occur in reward-based sensorimotor adaptation experiments with multiple sensory stimuli . We predict that for a binary reward function , destructive interferences will be observed if the neurons that encode the stimuli have broad tuning curves . These interferences are a dynamical counterpart of the generalization function and might result in a dramatic slowdown because of the abrupt change in the reward from success to failure around target size . We also predict that adding more targets should accelerate adaptation ( Figure 11 ) . From the learning curve of adaptation to one target , the rate and variability in which subjects adapt can be estimated . We predict that at parity of variability , subjects with larger learning rates will tend to display more destructive interferences and therefore slower adaptation to two targets ( see Eq . ( 23 ) ) . By contrast , if the tuning curves are very narrow , destructive interferences are unlikely to be found . However , even in this case , when the stimuli are sufficiently close , constructive interferences should be observed . In this case as well , adding more targets should accelerate the adaptation . Another prediction is that if adaptation is driven by reward modulated plasticity rules similar to the one we used here , smoothing the reward function should reduce interferences . In our model , this stems from the assumption of a reward modulated learning rule and not from the simplifying assumptions we made in constructing the model . Therefore , we suggest that testing this prediction could shed light on the synaptic mechanisms underlying adaptation tasks . Finally , the location of the reward-based mechanism involved in adaptation could be the cortex-basal-ganglia network . As a matter of fact , there is evidence for the involvement of this network in pitch shift adaptation in songbirds . Although the neural correlates for adaptation in songbirds are unknown , when an auditory feedback is available to songbirds ( by using miniature headphones [7] ) , the anterior frontal pathway , which is the avian homologue of the cortex-basal-ganglia network [50] , is essential for adaptation based solely on binary rewards [15] , [16] . Thus , exploring the behavioral and neural differences in auditory feedback versus binary reward adaptations in pitch shift experiments in songbirds may help reveal the neural mechanisms for reward-based adaptation .
We consider a motor reaching task ( see Figure 1A ) in which a subject manually controls the location of a cursor on a screen to bring it within a circular target of radius [16] . The target location is characterized by a two dimensional vector of norm 1 ( we assume that the target is always at distance 1 from the center of the screen ) and direction . In a standard block of trials , the direction of motion of the cursor and the hand are the same . We assume that the subject is able to perform the task perfectly in this case . In a rotation block of trials a perturbation is introduced: the movement of the cursor on the screen is now rotated by an angle with respect to the hand movement . To overcome this perturbation the subject must move his hand in a direction with respect to the target . Here we focus on the case where there is no visual feedback ( the cursor is not on the screen ) : the only information the subject receives about his performance is provided by a reward signal delivered by the experimentalist [17] . We consider a simplified computational model of a network performing this reaching task , see Figure 1B . The input layer of the network encodes the sensory information regarding the direction of the target , . It is composed of directionally tuned neurons labeled by their preferred direction , . For simplicity , we assume that the shape of the tuning curves is the same for all neurons: upon presentation of a target in direction the activity of neuron is . We take: ( 2 ) where characterizes the width of the tuning curve and is the peak response of a neuron . The width of the tuning curves at half of its maximal activity relative to the baseline ( half bandwidth ) in this case is: ( 3 ) The second layer of the network encodes the location of the endpoint of the hand movement . It consists of two output units whose activities , and , represent the two components of the hand position , . Upon presentation of a target in direction : ( 4 ) where is the connectivity matrix between the two layers , denotes the N dimensional vector of the input layer with components , and is a Gaussian noise . The location of the cursor at the end of the movement is related to by a rotation matrix , , of angle . Therefore , the squared distance between the endpoint location of the cursor and the center of the target is: ( 5 ) where . This quantity will be used to measure the error with which the network performs the reaching task . It is also useful to define the noiseless error: ( 6 ) where . This quantity measures the error if the noise is suppressed . Upon presentation of a target in a direction θ at trial t , the network performs the task and a reward R is delivered according to the outcome: ( 7 ) The matrix is then modified according to a reward-modulated learning rule: ( 8 ) where is the learning rate . This learning rule can be derived in a REINFORCE framework [36] . We assume that at the beginning of learning , when there is no rotation , the network is able to perform the reaching task with zero noiseless error for all targets . When all the Fourier components of are non-zero , this constraint fully determines : ( 9 ) where is the first Fourier component of the tuning curves . To get Eq . 10 , one needs to calculate the Fourier expansion of by using the constraint:for each of the possible target directions , . When some of the Fourier coefficients of the tuning curve function are zero , e . g . when using a cosine tuning curves , is determined up to the Fourier coefficient that are irrelevant to the above constraint . This does not affect the learning dynamics . In the numerical simulations described in this paper , the input layer consists of neurons . We normalized the tuning curves ( parameter in Eq . ( 2 ) ) such that remains constant ( ) when changing . This was done to guarantee that the time to learn one target does not depend on the tuning width . | The brain has a robust ability to adapt to external perturbations imposed on acquired sensorimotor transformations . Here , we used a mathematical model to investigate the reward-based component in sensorimotor adaptations . We show that the shape of the delivered reward signal , which in experiments is usually binary to indicate success or failure , affects the adaptation dynamics . We demonstrate how the ability to adapt to perturbations by relying solely on binary rewards depends on motor variability , size of perturbation and the threshold for delivering the reward . When adapting motor responses to multiple sensory stimuli simultaneously , on-line interferences between the motor performance in response to the different stimuli occur as a result of the overlap in the neural representation of the sensory stimuli , as well as the physical distance between them . Adaptation may be extremely slow when perturbations are induced to a few stimuli that are physically different from each other because of destructive interferences . When intermediate stimuli are introduced , the physical distance between neighbor stimuli is reduced , and constructive interferences can emerge , resulting in faster adaptation . Remarkably , adaptation to a widespread sensorimotor perturbation is accelerated by increasing the number of sensory stimuli during training , i . e . learning is faster if one learns more . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience",
"motor",
"systems",
"biology",
"computational",
"neuroscience"
] | 2014 | Interference and Shaping in Sensorimotor Adaptations with Rewards |
Accurate identification of mycetoma causative agent is a priority for treatment . However , current identification tools are far from being satisfactory for both reliable diagnosis and epidemiological investigations . A rapid , simple , and highly efficient molecular based method for identification of agents of black grain eumycetoma is introduced , aiming to improve diagnostic in endemic areas . Rolling Circle Amplification ( RCA ) uses species-specific padlock probes and isothermal DNA amplification . The tests were based on ITS sequences and developed for Falciformispora senegalensis , F . tompkinsii , Madurella fahalii , M . mycetomatis , M . pseudomycetomatis , M . tropicana , Medicopsis romeroi , and Trematosphaeria grisea . With the isothermal RCA assay , 62 isolates were successfully identified with 100% specificity and no cross reactivity or false results . The main advantage of this technique is the low-cost , high specificity , and simplicity . In addition , it is highly reproducible and can be performed within a single day .
Black grain eumycetoma represents the most common fungal mycetoma worldwide . This chronic , erosive infection of subcutaneous tissues particularly affects the lower extremities and leads to severe disability [1] . The disease is considered a major health problem in tropical areas and is prevalent among people of low socio-economic status [2] . Mycetoma presents as a subcutaneous mass with multiple sinuses that discharge pus , serous fluid and grains , i . e . the characteristic compact grains of the causative agent formed inside the lesion [3] . A wide range of microorganisms has been reported to cause mycetoma . For treatment , not only differentiation between ( fungal ) eumycetoma and ( bacterial ) actinomycetoma is important , but also the identity of the causative agent , since species differ in their response to antimicrobial drugs [4] . In endemic countries , clinical diagnosis may be the only diagnostic method . A fully developed mycetoma lesion is easily identified clinically , whereas in early stages with the absence of grains , the infection may be confused with phaeomycosis or soft tissue tumors [1] . In such cases fine needle aspiration cytology or deep surgical biopsy for histological examination are useful [1] , [5] . Some fungal and bacterial grains have a characteristic histological appearance which helps in provisional identification , but recognition of the causative species remains impossible [6] . Isolation of the pathogen from discharged grains or from biopsies allows identification of agents that sporulate , but most of the species lack phenotypic characteristics [3] . Molecular techniques have been introduced to facilitate the identification of nondescript organisms [7] , [8] , [9] , but are of high cost and time-consuming . Thus , there is a need for a fast , simple and reliable method for identification . Rolling circle amplification ( RCA ) is a powerful diagnostic method based on detection of specific nucleic-acid sequences and enzymatic amplification of circularized oligonucleotide probes under isothermal conditions [10] . The probes are linear oligonucleotides that contain two target-complementary sequences at their ends joined by linkers [11] . The ends of the probe hybridize to the complementary target in juxtaposition and then ligate which allows the circularization of the probe [11] . The circular structured molecule then amplifies with DNA polymerase that has strand displacement and progressive DNA synthesis activity resulting in series of repeats of the original circular template [11] , [12] . The technique has been proven to be rapid , specific and low-cost for molecular identification of viruses , bacteria , and fungi [13] , [14] , [15] , [16] . It has been applied for identification of a rare black grain mycetoma species Exophiala jeanselmei [17] . In addition , RCA has been used successfully for identification of white grain mycetoma species Scedosporium boydii [18] . The aim of the present study is to develop RCA-based diagnostics for the most common agents of black-grain eumycetoma .
The study included 62 isolates belonging to eight species causing black grain mycetoma: Madurella mycetomatis ( n = 32 ) , M . fahalii ( n = 1 ) , M . pseudomycetomatis ( n = 3 ) , M . tropicana ( n = 2 ) , Trematosphaeria grisea ( n = 10 ) , Falciformispora senegalensis ( n = 6 ) , F . tompkinsii ( n = 2 ) , and Medicopsis romeroi ( n = 6 ) . Strains were obtained from the reference collections of CBS-KNAW Fungal Biodiversity Centre ( Utrecht , The Netherlands ) and the Mycetoma Research Centre ( MRC , Khartoum , Sudan ) and are listed with metadata in Table 1 . Type strains of all tested species were included . All strains were identified down to species level by sequencing of the rDNA ITS region [19] , [20] . DNA was extracted using cetyltrimethyl ammonium bromide ( CTAB ) method as described by Möller et al . [21] . Amplification of the ITS region was performed using primers V9G and LS266 [22] in a 25 µL reaction mixture containing: 10 ng of template DNA , 0 . 1 mM each dNTP , 0 . 6 U Taq polymerase ( GC Biotech , Alphen aan den Rijn , The Netherlands ) , 1 µL of each primer ( 10 pmol ) and 2 . 5 µL reaction buffer ( 0 . 1 M Tris-HCl , 0 . 5 M KCl , 25 mM MgCl2 , 0 . 1% gelatin , 1% Triton X-100 ) . PCR reactions consisted of a 5 min predenaturation step at 95°C , followed by 30 cycles of 95°C for 30 s , 52°C for 30 s and 72°C for 1 min , with final post elongation step at 72°C for 7 min . PCR products were detected by electrophoresis using 1% agarose gels . Sequences of the ITS region were used to design 8 probes specific for each species used in this study . Two alignments were generated since the analyzed species were known to belong to two different fungal orders [19] , [20] . ITS derived from Madurella ( Sordariales ) were aligned with 200 isolates of Chaetomiaceae including Chaetomium , Thielavia , and Achaetomium . For the remaining species ( Pleosporales ) an alignment was constructed to include representative isolates of the family Trematosphaeriaceae and of coelomycetes in the suborder Pleosporineae . Sequences were aligned using BioNumerics v4 . 61 ( Applied Maths , Sint-Martens-Latem , Belgium ) . Probes were designed with minimum secondary structure and were checked using PrimerSelect ( DNASTAR Lasergene , WI , U . S . A . ) . To insure specificity of the probes , target-specific sites of each padlock probe was submitted to BLAST in NCBI sequence database for homologous sequences . The melting temperature of the 5′ end of the probe binding arm was designed to be>63°C while for the 3′ end binding arm it was designed to be at least 15°C below the annealing temperature . The probes were phosphorylated at the 5′ end and are listed in Table 2 . Probe linkers were taken from Zhou et al . [23] . Padlock probe ligation was performed in a mixture consisting of 1 µl purified ITS amplicons , 2 U pfu DNA ligase ( Epicentre Biotechnologies , Madison , WI , U . S . A . ) , and 0 . 1 µmol/l padlock probe in buffer ( 20 mM Tris-HCl pH 7 . 5 , 10 mM MgCl2 , 20 mM KCl , 0 . 1% Igepal , 0 . 01 mM rATP , 1 mM DTT ) , with a total reaction volume of 10 µl . Ligation conditions were: 5 min denaturation at 94°C , followed by 7 cycles of 94°C for 30 sec , 63°C for 4 min , and final cooling at 10°C . Prior to RCA amplification reaction and in order to reduce the ligation-independent amplification , ligation products were treated by addition of 10 U exonucleases I and 10 U exonucleases III ( New England Biolabs , Hitchin , U . K . ) with a final volume of 20 µl . The mixture was then incubated for 30 min at 37°C , followed by 3 min at 94°C to deactivate the endonuclease enzymes . RCA amplification reaction was performed in a 50 µl mixture containing; 2 µl ligation product , 8 U Bst DNA polymerase ( New England Biolabs ) , 10 pmol of each RCA primer ( Table 2 ) , and 400 µM dNTP mix . The mixture was incubated at 65°C for 60 min and cooled at 10°C . Electrophoresis on a 1% agarose gel was used to visualize RCA products . A positive reaction is indicated by the presence of ladder-like pattern . The reaction was also visualized by adding 1 . 0 µl of a 10-fold diluted SYBR Green I ( Cambrex BioScience , Workingham , U . K . ) to 10 µl of the amplification product . Accumulated double stranded DNA was detected with UV transilluminator ( Vilber Lourmat , Marne-la-Vallée , France ) . The specificity of the 8 RCA probes was tested using strains of black-grain mycetoma causative species listed in table 1 . Analytical sensitivity was determined using 10-fold serial dilution of M . mycetomatis ( CBS 109801 ) and M . fahalii ( CBS 129176 ) DNA and the test was performed as mentioned above . In addition , RCA was performed directly using DNA samples without amplification of the target gene . To evaluate the detection limit from direct DNA samples two-fold serial dilutions of target DNA were tested . The sensitivity of the RCA probes was also determined by 10-fold serial dilution of MYC and MFAH probes tested with amplified ITS of M . mycetomatis and M . fahalii respectively .
RCA was used to identify 62 strains belonging to eight species causing human eumycetoma . Since black grain eumycetoma species are known to be phylogenetically distant , it is easy to find unique sites for their identification . The ribosomal ITS region was sufficient for identification of all species and showed no intraspecific variability within a set of 100 M . mycetomatis strains in our collection . For M . mycetomatis , M . tropicana , M . pseudomycetomatis , and F . senegalensis the ITS1 region was selected for probe design , while for M . fahalii , T . grisea , F . tompkinsii and M . romeroi the ITS 2 region was found to be more suitable . RCA results for the tested strains were easily visualized in 1% agarose gel . Positive reactions demonstrated ladder like patterns while negative reactions resulted in a clear background ( Fig . 1 ) . With SYBR green , positive results showed green fluorescence when exposed to UV light , while negatives did not . When exonucleolysis was performed some inhibition was observed with low RCA positive signals on gel or with fluorescence . Faint non-specific bands were observed when this step was omitted . RCA reactions were performed successfully without digestion with exonucleases , as the non-specific bands did not interfere with RCA results . All M . mycetomatis strains were correctly identified with RCA , irrespective of their geographical origin ( Sudan , India , Mali ) ( Fig . 2 ) . For the other agents , each individual species-specific probes yielded positive results with their corresponding species and with 100% agreement with ITS sequencing ( Fig . 2 , Table 3 ) . No cross reactivity or false positive and negative results were observed . The sensitivity of RCA when using amplified product of the target gene was less than 32×10−3 ng of DNA . A higher concentration of 100 ng is needed when the test is carried out directly from the DNA samples without amplification of the ITS . The probes were very sensitive and a concentration of 6 . 6×10−5 ng was successfully ligated and then amplified with RCA . The turnaround time required for conducting the entire experiment including PCR amplification of target DNA , RCA processing and analysis was found to be 6 hours . DNA sequencing of the ITS region took more than 8 hours to be performed ( Fig . 3 ) .
Mycetoma is a unique tropical disease , endemic in many tropical and subtropical regions that has been recently added to the WHO list of neglected tropical diseases . [24] . It is mainly prevalent in what is known as “mycetoma belt” which includes Mexico , Senegal , Sudan , India and other countries between tropic of cancer [1] . In 2014 , a mycetoma consortium of scientists and physicians published research gaps on mycetoma which need to be addressed in the coming years [2] . One of the research priorities identified was the need to develop a reliable and cost-effective method for species identification to improve diagnosis [2] . Mycetoma agents have been extensively studied in recent years [8] , [9] , [20] . The large phylogenetic distance between a number of these agents provides the possibility to use a moderately variable marker like rDNA ITS for species identity . Ahmed et al . [25] developed PCR-restriction fragment length polymorphism ( RFLP ) for identification of M . mycetomatis targeting the ITS region . However , with the description of the molecular siblings M . fahalii , M . pseudomycetomatis , and M . tropicana [26] the method might be insufficiently accurate . Moreover , there is a need for identification these siblings species; Madurella grisea appeared to be distantly related and was re-named as T . grisea [20] . In the present study we developed a simple , fast and highly specific molecular method for the identification of agents of black grain mycetoma . In this method , the ITS region is easily amplified using one set of primers , which simplifies the use . In a second , isothermal amplification reaction padlock probes are used to identify the species by RCA . The only equipment necessary is a thermocycler for the PCR reaction and a water bath or heating block for the RCA reaction . This relative simplicity enhances possible use in routine laboratories in endemic areas . Due to its robustness , high potential , and reproducibility , RCA is increasingly used as a diagnostic tool in pathogenic fungi , e . g . agents of chromoblastomycosis , dermatophytes , Aspergillus , Candida , and Talaromyces marneffei [16] , [23] , [27] , [28] . The method does not require DNA sequencing and is therefore considered as a rapid and cost-effective . Applications are being expanded to nano- and biotechnology [29] . In the present study eight species-specific probes were designed and used for identification of 62 isolates . For the RCA reaction species probe hybridization to the 3′ and 5′ ends of target DNA and joining of adjacent ends by DNA ligase when both show perfect complementarity . The ligation appears to be highly specific and thus the method can detect single nucleotide polymorphism [30] . The amplification reaction is driven by an isothermal DNA polymerase to amplify the circularized probes with high efficiency and an estimated capacity to synsthesize more than 70 , 000 bp per hour [31] . RCA products can be detected with different methods including gel electrophoresis , radiolabeling , UV absorbance , fluorescence , and single molecule detection [32] . It was known that the positive signals can be detected within 15 min after starting the RCA reaction by real time PCR [23] . In the present study the RCA positive signal was easily visualized using both gel electrophoresis and fluorescent dye . The duration of our RCA protocol was 2 h , but additional time is required for DNA extraction and ITS amplification . Compared to the DNA sequencing the turnaround time for RCA is 2 hours less than sequencing and this even more if there is no in-house sequencer available . Our results with eight padlock probes showed that RCA accurately identified all species with no cross reactivity ( Fig . 1 ) . It may be concluded that RCA is extremely useful for specific identification of agents of mycetoma . Performance and rapid turnaround time features make the RCA suitable for quick and reliable diagnosis , which is an enormous improvement compared to the current phenotypic identification of mostly non-sporulating cultures . Future application of RCA could be the detection of agents DNA directly from clinical samples without requirement of culturing . | Treatment of eumycetoma largely depends on the causative pathogen . Identification of mycetoma agent with phenotypic features is too limited , and physiological and biochemical techniques are laborious , time-consuming and nonspecific , whereas the currently available molecular methods based on DNA sequencing are specific but extremely expensive . We describe rolling circle amplification method for identification of black grain eumycetoma using species-specific padlock probes . Eight probes were designed and successfully used for species identification and the results were easily visualized in 1% agarose gel . RCA provides a simple , reproducible , and cost-effective method for rapid identification of mycetoma agent that can be used in low-resource clinical settings . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences"
] | 2014 | Rapid Identification of Black Grain Eumycetoma Causative Agents Using Rolling Circle Amplification |
Gene regulatory networks are composed of sub-networks that are often shared across biological processes , cell-types , and organisms . Leveraging multiple sources of information , such as publicly available gene expression datasets , could therefore be helpful when learning a network of interest . Integrating data across different studies , however , raises numerous technical concerns . Hence , a common approach in network inference , and broadly in genomics research , is to separately learn models from each dataset and combine the results . Individual models , however , often suffer from under-sampling , poor generalization and limited network recovery . In this study , we explore previous integration strategies , such as batch-correction and model ensembles , and introduce a new multitask learning approach for joint network inference across several datasets . Our method initially estimates the activities of transcription factors , and subsequently , infers the relevant network topology . As regulatory interactions are context-dependent , we estimate model coefficients as a combination of both dataset-specific and conserved components . In addition , adaptive penalties may be used to favor models that include interactions derived from multiple sources of prior knowledge including orthogonal genomics experiments . We evaluate generalization and network recovery using examples from Bacillus subtilis and Saccharomyces cerevisiae , and show that sharing information across models improves network reconstruction . Finally , we demonstrate robustness to both false positives in the prior information and heterogeneity among datasets .
Gene regulatory network inference aims at computationally deriving and ranking regulatory hypotheses on transcription factor-target gene interactions [1–3] . Often , these regulatory models are learned from gene expression measurements across a large number of samples . Strategies to obtain such data range from combining several publicly available datasets to generating large expression datasets from scratch [4–7] . Given decreasing costs of sequencing and the exponential growth in the availability of gene expression data in public databases [8 , 9] , data integration across several studies becomes particularly promising for an increasing number of biological systems . In theory , multi-study analyses provide a better representation of the underlying cellular regulatory network , possibly revealing insights that could not be uncovered from individual studies [6] . In practice , however , biological datasets are highly susceptible to batch effects [10] , which are systematic sources of technical variation due to different reagents , machines , handlers etc . that complicate omics meta-analyses [11 , 12] . Although several methods to remove batch effects from expression data have been developed , they often rely on evenly distributed experimental designs across batches [13 , 14] . Batch-correction methods may deflate relevant biological variability or induce incorrect differences between experimental groups when conditions are unbalanced across batches , which can significantly affect downstream analyses [15] . Therefore these batch effect removal methods are not applicable when integrating public data from multiple sources with widely differing experimental designs . In network inference , an approach often taken to bypass batch effects is to learn models from each dataset separately and combine the resulting networks [16 , 17] . Known as ensemble learning , this idea of synthesizing several weaker models into a stronger aggregate model is commonly used in machine learning to prevent overfitting and build more generalizable prediction models [18] . In several scenarios , ensemble learning avoids introducing additional artifacts and complexity that may be introduced by explicitly modeling batch effects . On the other hand , the relative sample size of each dataset is smaller when using ensemble methods , likely decreasing the ability of an algorithm to detect relevant interactions . As regulatory networks are highly context-dependent [19] , for example , TF binding to several promoters is condition-specific [20] , a drawback for both batch-correction and ensemble methods is that they produce a single network model to explain the data across datasets . Relevant dataset-specific interactions might not be recovered , or just difficult to tell apart using a single model . Although it will not be the primary focus of this paper , most modern network inference algorithms integrate multiple data-types to derive prior or constraints on network structure . These priors/constraints have been shown to dramatically improve network model selection performance when combined with the state variables provided by expression data . In these methods [17 , 21] , priors or constraints on network structure ( derived from multiple sources like known interactions , ATAC-seq , DHS , or ChIP-seq experiments [22–24] ) are used to influence the penalty on adding model components , where edges in the prior are effectively penalized less . Here we describe a method that builds on that work ( and similar work in other fields ) , but in addition we let model inference processes ( each carried out using a separate data-set ) influence each others model penalties , so that edges that agree across inference tasks are more likely to be uncovered [25–31] . Several previous works on this front focused on enforcing similarity across models by penalizing differences on strength and direction of regulatory interactions using a fusion penalty [25 , 27 , 28] . Because the influence of regulators on the expression of targets may vary across datasets , possibly even due to differences in measurement technologies , we look to induce similarity on network structure ( the choice of regulators ) using a group-sparse penalty . Previous methods also applied this type of penalty [26 , 29 , 31] , however , they were not robust to differences in relevant edges across datasets . Here we propose a multitask learning ( MTL ) approach to exploit cross-dataset commonalities while recognizing differences and is able to incorporate prior knowledge on network structure if available [32 , 33] . In this framework , information flow across datasets leads the algorithm to prefer solutions that better generalize across domains , thus reducing chances of overfitting and improving model predictive power [34] . Since biological datasets are often under-sampled , we hypothesize that sharing information across models inferred from multiple datasets using a explicit multitask learning framework will improve accuracy of inferred network models in a variety of common experimental designs/settings . In this paper , we explicitly show that joint inference significantly improves network recovery using examples from two model organisms , Bacillus subtilis and Saccharomyces cerevisiae . We show that models inferred for each dataset using our MTL approach ( which adaptively penalizes conserved and data-set-unique model components separately ) are vastly more accurate than models inferred separately using a single-task learning ( STL ) approach . We also explore commonly used data integration strategies , and show that MTL outperforms both batch-correction and ensemble approaches . In addition , we also demonstrate that our method is robust to noise in the input prior information . Finally , we look at conserved and dataset-specific inferred interactions , and show that our method can leverage cross-dataset commonalities , while being robust to differences .
To improve regulatory network inference from expression data , we developed a framework that leverages training signals across related expression datasets . For each gene , we assume that its regulators may overlap across conditions in related datasets , and thus we could increase our ability to uncover accurate regulatory interactions by inferring them jointly . Our method takes as input multiple expression datasets and priors on network structure , and then outputs regulatory hypotheses associated with a confidence score proportional to our belief that each prediction is true ( Fig 1A ) . As previous studies [17 , 35–37] , our method also includes an intermediate step that estimates transcription factor activities ( TFA ) , and then , models gene expression as a function of those estimates ( Fig 1B ) . In our model , TFA represent a relative quantification of active protein that is inducing or repressing the transcription of its targets in a given sample , and is an attempt to abstract away unmeasured factors that influence TFA in a living cell [37–39] , such as post-translational regulation [40] , protein-protein interactions [41] , and chromatin accessibility [42] . We estimate TFA from partial knowledge of the network topology ( Fig 1C ) [21 , 43–47] and gene expression data as previously proposed ( Fig 1D ) [17] . This is comparable to using a TF’s targets collectively as a reporter for its activity . Next , we learn the dependencies between gene expression and TFA and score predicted interactions . In this step , our method departs from previous work , and we employ multitask learning to learn regulatory models across datasets jointly , as opposed to single-task learning , where network inference is performed for each dataset independently ( Fig 1E ) . As genes are known to be regulated by a small number of TFs [48] , we can assume that these models are sparse , that is , they contain only a few nonzero entries [3] . We thus implement both approaches using sparsity-inducing penalties derived from the lasso [49] . Here the network model is represented as a matrix for each target gene ( where columns are data-sets/cell-types/studies and rows are potential regulators ) with signed entries corresponding to strength and type of regulation . Importantly , our MTL approach decomposes this model coefficients matrix into a dataset-specific component and a conserved component to enable us to penalize dataset-unique and conserved interactions separately for each target gene [32]; this separation captures differences in regulatory networks across datasets ( Fig 2 ) . Specifically , we apply an l1/l∞ penalty to the one component to encourage similarity between network models [50] , and an l1/l1 penalty to the other to accommodate differences [32] . We also incorporate prior knowledge by using adaptive weights when penalizing different coefficients in the l1/l1 penalty [33] . Finally , we perform this step for several bootstraps of the conditions in the expression and activities matrices , and calculate a confidence score for each predicted interaction that represents both the stability across bootstraps and the proportion of variance explained of the target expression dependent on each predictor . Our method is readily available in an open-source package , Inferelator-AMuSR ( Adaptive Multiple Sparse Regression ) , enabling TF activity estimation and multi-source gene regulatory network inference , ultimately facilitating mechanistic interpretations of gene expression data to the Biology community . In addition , this method allows for adaptive penalties to favor interactions with prior knowledge proportional to the user-defined belief that interactions in the prior are true . Finally , our implementation also includes several mechanisms that speed-up computations , making it scalable for the datasets used here , and support for parallel computing across multiple nodes and cores in several computing environments . We validated our approach using two model organisms , a gram-positive bacteria , B . subtilis , and an eukaryote , S . cerevisiae . Availability of validated TF-target regulatory interactions , hereafter referred to as the gold-standard , make both organisms a good choice for exploring inference methods ( 3040 interactions , connecting 153 TFs to 1822 target genes for B . subtilis [17 , 46] , 1198 interactions connecting 91 TFs to 842 targets for S . cerevisiae [51] ) . For B . subtilis , we use two expression datasets . The first one , B . subtilis 1 , was collected for strain PY79 and contains multiple knockouts , competence and sporulation-inducing conditions , and chemical treatments ( 429 samples , 38 experimental designs with multiple time-series experiments ) [17] . The second dataset , B . subtilis 2 , was collected for strain BSB1 and contains several nutritional , and other environmental stresses , as well as competence and sporulation-inducing conditions ( 269 samples , and 104 conditions ) [52] . For S . cerevisiae , we downloaded three expression datasets from the SPELL database [53] . S . cerevisiae 1 is a compendium of steady-state chemostat cultures with several combinations of cultivation parameters ( 170 samples , 55 conditions ) [54] . S . cerevisiae 2 profiles two yeast strains ( BY and RM ) grown with two carbon sources , glucose and ethanol , in different concentrations ( 246 samples , and 109 conditions ) [55] . Finally , S . cerevisiae 3 with expression profiles following several mutations and chemical treatments ( 300 samples ) [56] . Each dataset was collected using a different microarray platform . Cross-platform data aggregation is well known to cause strong batch effects [10] . For each species , we considered the set of genes present across datasets . In our inference framework , prior knowledge on network topology is essential to first estimate transcription factor activities and to then bias model selection towards interactions with prior information during the network inference stage of the algorithm . Therefore , to properly evaluate our method , it is necessary to gather prior interactions independent of the ones in the gold-standard . For B . subtilis , we adopt the previously used strategy of partitioning the initial gold-standard into two disjoint sets , a prior for use in network inference and a gold-standard to evaluate model quality [17] . For S . cerevisiae , on the other hand , we wanted to explore a more realistic scenario , where a gold-standard is often not available . In the absence of such information , we hypothesized that orthogonal high-throughput datasets would provide insight . Because the yeast gold-standard [51] was built as a combination of TF-binding ( ChIP-seq , ChIP-ChIP ) and TF knockout datasets available in the YEASTRACT [47] and the SGD [57] databases , we propose to derive prior knowledge from chromatin accessibility data [22 , 23] and TF binding sites [58] ( as this is a realistic and efficient genomic experimental design for non-model organisms ) . Open regions in the genome can be scanned for transcription factor binding sites , which can provide indirect evidence of regulatory function [59] . We then assigned TFs to the closest downstream gene , and built a prior matrix where entries represent the number of motifs for a particular TF that was associated to a gene [60 , 61] . We obtained a list of regulators from the YeastMine database [62] , which we also used to sign entries in the prior: interactions for regulators described as repressors were marked as negative . Because genome-wide measurements of DNA accessibility can be obtained in a single experiment , using techniques that take advantage of the sensitivity of nucleosome-free DNA to endonuclease digestion ( DNase-seq ) or to Tn5 transposase insertion ( ATAC-seq ) [63] , we expect this approach to be generalizable to several biological systems . Using the above expression datasets and priors , we learn regulatory networks for each organism employing both single-task and our multitask approaches . To provide an intuition for cross-dataset transfer of knowledge , we compare confidence scores attributed to a single gold-standard interaction using either STL or MTL for each organism . For B . subtilis , we look at the interaction between the TF sigB and the gene ydfP ( Fig 3A ) . The relationship between the sigB activity and ydfP expression in the first dataset B . subtilis 1 is weaker than in B . subtilis 2 . This is reflected in the predicted confidence scores , a quarter as strong for B . subtilis 1 than for B . subtilis 2 , when each dataset is used separately to learn networks through STL . On the other hand , when we learn these networks in the MTL framework , information flows from B . subtilis 2 to B . subtilis 1 , and we assign a high confidence score to this interaction in both networks . Similarly , for S . cerevisiae , we look at the interaction between the TF Rap1 and the target gene Rpl12a ( Fig 3B ) . In this particular case , we observe a strong and easier-to-uncover relationship between Rap1 estimated activity and Rpl12a expression for all datasets . Indeed , we assign a nonzero confidence score to this interaction for all datasets using STL , although for S . cerevisiae 2 and 3 these are much smaller than the scores attributed when networks are learned using MTL . In order to evaluate the overall quality of the inferred networks , we use area under precision-recall curves ( AUPR ) [16] , widely used to quantify a classifier’s ability to distinguish two classes and to rank predictions . Networks learned using MTL are significantly more accurate than networks learned using the STL approach . For B . subtilis ( Fig 3D ) , we observe around a 30% gain in AUPR for both datasets , indicating significant complementarity between the datasets . For S . cerevisiae ( Fig 3E ) , we observe a clear increase in performance for networks inferred for every dataset , indicating that our method is very robust to both data heterogeneity and potential false edges derived from chromatin accessibility in the prior . These experiments were also performed using TF expression as covariates , instead of TF activities , and those results are shown at ( S1A and S1B Fig ) . Although we recommend using TFA for the organisms here tested , MTL also improves the performance for each dataset-specific network in this scenario . Next , we asked whether the higher performance of the MTL framework could be achieved by other commonly used data integration strategies , such as batch-correction and ensemble methods . Ensemble methods include several algebraic combinations of predictions from separate classifiers trained within a single-domain ( sum , mean , maximum , minimum [64] ) . To address this question , we evaluated networks inferred using all available data . First , we combined regulatory models inferred for each dataset either through STL or MTL by taking the average rank for each interaction , generating two networks hereafter called STL-C and MTL-C [16] . For each organism , we also merged all datasets into one , and applied ComBat for batch-correction [65] , because of its perceived higher performance [66] . We then learn network models from these larger batch-corrected datasets , STL-BC . Both for B . subtilis ( Fig 4A ) and S . cerevisiae ( Fig 4B ) , the MTL-C networks outperform the STL-C and STL-BC networks , indicating that cross-dataset information sharing during modelling is a better approach to integrate datasets from different domains . Interestingly , for B . subtilis , the STL-BC network has a higher performance than the STL-C network , whereas for yeast we observe the opposite . We speculate that the higher overlap between the conditions in the two B . subtilis datasets improved performance of the batch-correction algorithm here used . For yeast , on the other hand , conditions were very different across datasets , and although much new information is gained by merging datasets into one , it is likely that incorrect relationships between genes were induced as an artifact , possibly confounding the inference . Of note , these approaches emphasize the commonalities across datasets , whereas the motivation to use MTL frameworks is to increase statistical power , while maintaining separate models for each dataset , hopefully improving interpretability . These experiments were also performed using TF expression as covariates , instead of TF activities , and those results are shown at ( S2A and S2B Fig ) . In that case , results hold for yeast , but not for B . subtilis . Because genes are frequently co-regulated , and biological networks are redundant and robust to perturbations , spurious correlations between transcription factors and genes are highly prevalent in gene expression data [67 , 68] . To help discriminate true from false interactions , it is essential to incorporate prior information to bias model selection towards interactions with prior knowledge . Indeed , incorporating prior knowledge has been shown to increase accuracy of inferred models in several studies [3 , 21 , 69] . For example , suppose that two regulators present highly correlated activities , but regulate different sets of genes . A regression-based model would be unable to differentiate between them , and only other sources of information , such as binding evidence nearby a target gene , could help selecting one predictor over the other in a principled way . Thus , we provide an option to integrate prior knowledge to our MTL approach in the model selection step by allowing the user to input a “prior weight” . This weight is used to increase presence of prior interactions to the final model , and should be proportional to the quality of the input prior . Sources of prior information for the two model organisms used in this study are fundamentally different . The B . subtilis prior is high-quality , derived from small-scale experiments , whereas the S . cerevisiae prior is noisier , likely with both high false-positive and false-negative rates , derived from high-throughput chromatin accessibility experiments and TF binding motifs . To understand differences in prior influences for the same organism , we also include the yeast gold-standard as a possible source of prior in this analysis . The number of TFs per target gene in the B . subtilis ( Fig 5A ) and the S . cerevisiae ( Fig 5B ) gold-standards ( GS ) is hardly ever greater than 2 , with median of 1 , whereas for the chromatin accessibility-derived priors ( ATAC ) for S . cerevisiae , the median is 11 ( Fig 5C ) . A large number of regulators per gene likely indicates a high false-positive rate in the yeast ATAC prior . Given the differences in prior quality , we test the sensitivity of our method to the prior weight parameter . We applied increasing prior weights , and measured how the confidence scores attributed to prior interactions was affected ( Fig 5D ) for the three source of priors described above . Interestingly , the confidence scores distributions show dependencies on both the prior quality and the prior weights . When the gold-standard interactions for B . subtilis and S . cerevisiae are used as prior knowledge , they receive significantly higher scores than interactions in the S . cerevisiae chromatin accessibility-derived prior , which is proportional to our belief on the quality of the input prior information . Importantly , even when we set the prior weight value to a very high value , such as 10 , interactions in the ATAC prior are not pushed to very high confidence scores , suggesting that our method is robust to the presence of false interactions in the prior . In order to test this hypothesis , we artificially introduced false edges to both the B . subtilis and the yeast gold-standards . We added 1 randomly chosen “false” interaction for every 5 true edges in the gold-standard . That affects both TFA estimation and model selection ( for prior weights greater than 1 ) . We then ran the inference using the Inferelator-AMuSR method with increasing prior weights , and evaluated both the confidence scores of recovered true and false interactions ( Fig 5C ) as well as the counts of true and false interactions that receive non-zero confidence scores ( Fig 5D ) . For both B . subtilis and yeast , we notice that confidence scores distributions show dependency on whether edges are true or false , indicating that the method is not overfitting the prior for the majority of datasets , even when prior weights used are as high as 10 ( Fig 5C ) . We speculate that the greater completeness of the B . subtilis gold-standard and of the expression datasets make it easier to differentiate true from false prior interactions when compared to yeast . Besides , inferring networks for prokaryotes is regarded as an easier problem [16] . Importantly , we also show the number of non-zero interactions in each of these distributions ( Fig 5D ) . Taken together , these results show that our method is robust to false interactions in the prior , but requires the user to choose an appropriate prior weight for the specific application . As in previous studies [43] , in the presence of a gold-standard , we recommend the user to evaluate performance in leave-out sets of interactions to determine the best prior weight to be used . In the absence of a gold-standard , priors are likely to be of lower confidence , and therefore , smaller prior weights should be used . Because multitask learning approaches are inclined to return models that are more similar to each other , we sought to understand how heterogeneity among datasets affected the inferred networks . Specifically , we quantified the overlap between the networks learned for each dataset for B . subtilis and yeast . That is , the number of edges that are unique or shared across networks inferred for each dataset ( Fig 6 ) . In this analysis , we consider valid only predictions within a 0 . 5 precision cut-off , calculated using only TFs and genes present in the gold-standard . Since the B . subtilis datasets share more conditions than the yeast datasets , we hypothesized that the B . subtilis networks would have a higher overlap than the yeast networks . As expected , we observe that about 40% of the total edges are shared among two B . subtilis networks ( Fig 6A ) , whereas for yeast only about 27% ( Fig 6B ) and 22% ( Fig 6C ) , using gold-standard and chromatin accessibility-derived priors respectively , of the total number of edges is shared by at least two of the three inferred networks . Therefore , our approach for joint inference is robust to cross-dataset influences , preserving relative uniqueness when datasets are more heterogeneous .
In this study , we presented a multitask learning approach for joint inference of gene regulatory networks across multiple expression datasets that improves performance and biological interpretation by factoring network models derived from multiple datasets into conserved and dataset-specific components . Our approach is designed to leverage cross-dataset commonalities while preserving relevant differences . While other multitask methods for network inference penalize for differences in model coefficients across datasets [25–28 , 30] , our method leverages shared underlying topology rather than the influence of TFs on targets . We expect this method to be more robust , because , in living cells , a TF’s influence on a gene’s expression can change in different conditions . In addition , previous methods either deal with dataset-specific interactions [25] , or apply proper sparsity inducing regularization penalties [26–30] . Our approach , on the other hand , addresses both concerns . Finally , we implemented an additional feature to allow for incorporation of prior knowledge on network topology in the model selection step . Using two different model organisms , B . subtilis and S . cerevisiae , we show that joint inference results in accurate network models . We also show that multitask learning leads to more accurate models than other data integration strategies , such as batch-correction and combining fitted models . Generally , the benefits of multitask learning are more obvious when task overlap is high and datasets are slightly under-sampled [34] . Our results support this principle , as the overall performance increase of multitask network inference for B . subtilis is more pronounced than for S . cerevisiae , which datasets sample more heterogeneous conditions . Therefore , to benefit from this approach , defining input datasets that share underlying regulatory mechanisms is essential and user-defined . A key question here , that requires future work , is the partitioning of data into separate datasets . Here we use the boundaries afforded by previous study designs: we use data from two platforms and two strains for B . subtilis ( a fairly natural boundary ) and the separation between studies by different groups ( again using different technologies ) in yeast . We choose these partitions to illustrate robustness to the more common sources of batch effect in meta-analysis . In the future , we expect that multitask methods in this domain will integrate dataset partition estimation ( which data go in which bucket ) with network inference . Such methods would ideally be able to estimate task similarity , taking into account principles of regulatory biology , and apply a weighted approach to information sharing . In addition , a key avenue for future work will be to adapt this method to multi-species studies . Examples of high biological and biomedical interest include joint inferences across model systems and organisms of primary interest ( for example data-sets that include mouse and human data collected for similar cell types in similar conditions ) . These results ( and previous work on many fronts [7 , 25 , 70] ) suggest that this method would perform well in this setting . Nevertheless , because of the increasing practice of data sharing in Biology , we speculate that cross-study inference methods will be largely valuable in the near future , being able to learn more robust and generalizable hypotheses and concepts . Although we present this method as an alternative to batch correction , we should point out that there are many uses to batch correction that fall outside of the scope of network inference , and our results do not lessen the applicability of batch correction methods to these many tasks . There is still great value in properly balancing experimental designs when possible to allow for the estimation of specific gene- and condition-wise batch effects . Experiments where we interact MTL learning with properly balanced designs and quality batch correction are not provided here , but would be superior . Thus , the results here should be strictly interpreted in the context of network inference , pathway inference , and modeling interactions .
For B . subtilis , we downloaded normalized expression datasets from the previously published network study by Arrieta-Ortiz et al [17] . Both datasets are available at GEO , B . subtilis 1 with accession number GSE67023 [17] and B . subtilis 2 with accession number GSE27219 [52] . For yeast , we downloaded expression datasets from the SPELL database , where hundreds of re-processed gene expression data is available for this organism . In particular , we selected three datasets from separate studies based on the number of samples , within-dataset condition diversity , and cross-dataset condition overlap ( such as nutrient-limited stress ) . S . cerevisiae 1 and S . cerevisiae 2 are also available at GEO at accession numbers GSE11452 [54] and GSE9376 [55] . S . cerevisiae 3 does not have a GEO accession number , and was collected in a custom spotted microarray [56] . For network inference , we only kept genes present in all datasets , resulting in 3780 and 4614 genes for B . subtilis and for yeast respectively . In order to join merge , for comparison , we consider each dataset to be a separate batch , since they were generated in different labs as part of separate studies , and applied ComBat for batch-correction using default parameters and no reference to experimental designs [65] . We approach network inference by modeling gene expression as a weighted sum of the activities of transcription factors [17 , 36] . Our goal is to learn these weights from gene expression data as accurately as possible . In this section , we explain our core model of gene regulation , and of transcription factor activities , and state our assumptions . We also describe how we extend our framework to support learning of multiple networks simultaneously , and integration of prior knowledge on network structure . Finally , we explain how we rank predicted interactions which is used to evaluate the ability of these methods to recover the known underlying network . For each predicted interaction we compute a confidence score that represents how well a predictor explains the expression data , and a measure of prediction stability . As previously proposed [17 , 43] , we calculate confidence scores for each interaction by: c k , i = 1 - σ f u l l m o d e l f o r x i 2 σ m o d e l f o r x i w i t h o u t p r e d i c t o r k 2 ( 12 ) where σ2 equals the variance of the residuals for the models , with and without predictor k . The score ck , i is proportional to how much removing regulator k from gene i set of predictors decreases model fit . To measure stability , we perform the inference across multiple bootstraps of the expression data ( we used 20 bootstraps for both B . subtilis and yeast ) , rank-average the interactions across all bootstraps [16 , 43] , and re-scale the ranking between 0 and 1 to output a final ranked list of regulatory hypotheses . | Due to increasing availability of biological data , methods to properly integrate data generated across the globe become essential for extracting reproducible insights into relevant research questions . In this work , we developed a framework to reconstruct gene regulatory networks from expression datasets generated in separate studies—and thus , because of technical variation ( different dates , handlers , laboratories , protocols etc… ) , challenging to integrate . Since regulatory mechanisms are often shared across conditions , we hypothesized that drawing conclusions from various data sources would improve performance of gene regulatory network inference . By transferring knowledge among regulatory models , our method is able to detect weaker patterns that are conserved across datasets , while also being able to detect dataset-unique interactions . We also allow incorporation of prior knowledge on network structure to favor models that are somewhat similar to the prior itself . Using two model organisms , we show that joint network inference outperforms inference from a single dataset . We also demonstrate that our method is robust to false edges in the prior and to low condition overlap across datasets , and that it can outperform current data integration strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"genetic",
"networks",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"pathogens",
"regulatory",
"proteins",
"bacillus",
"microbiology",
"dna-binding",
"proteins",
"fungi",
"prokaryotic",
"models",
"model",
"organisms",
"regulator",
"genes",
"network",
"analysis",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"gene",
"types",
"bacteria",
"bacterial",
"pathogens",
"saccharomyces",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"animal",
"studies",
"proteins",
"medical",
"microbiology",
"gene",
"expression",
"microbial",
"pathogens",
"yeast",
"biochemistry",
"eukaryota",
"gene",
"identification",
"and",
"analysis",
"bacillus",
"subtilis",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"organisms"
] | 2019 | Multi-study inference of regulatory networks for more accurate models of gene regulation |
Peroxisome biogenesis disorders ( PBD ) are a group of multi-system human diseases due to mutations in the PEX genes that are responsible for peroxisome assembly and function . These disorders lead to global defects in peroxisomal function and result in severe brain , liver , bone and kidney disease . In order to study their pathogenesis we undertook a systematic genetic and biochemical study of Drosophila pex16 and pex2 mutants . These mutants are short-lived with defects in locomotion and activity . Moreover these mutants exhibit severe morphologic and functional peroxisomal defects . Using metabolomics we uncovered defects in multiple biochemical pathways including defects outside the canonical specialized lipid pathways performed by peroxisomal enzymes . These included unanticipated changes in metabolites in glycolysis , glycogen metabolism , and the pentose phosphate pathway , carbohydrate metabolic pathways that do not utilize known peroxisomal enzymes . In addition , mutant flies are starvation sensitive and are very sensitive to glucose deprivation exhibiting dramatic shortening of lifespan and hyperactivity on low-sugar food . We use bioinformatic transcriptional profiling to examine gene co-regulation between peroxisomal genes and other metabolic pathways and we observe that the expression of peroxisomal and carbohydrate pathway genes in flies and mouse are tightly correlated . Indeed key steps in carbohydrate metabolism were found to be strongly co-regulated with peroxisomal genes in flies and mice . Moreover mice lacking peroxisomes exhibit defective carbohydrate metabolism at the same key steps in carbohydrate breakdown . Our data indicate an unexpected link between these two metabolic processes and suggest metabolism of carbohydrates could be a new therapeutic target for patients with PBD .
Peroxisomes are ubiquitous organelles present in all eukaryotic cells [1–3] . Peroxisomes perform specific biochemical functions in the cell including fatty acid β-oxidation of very-long-chain fatty acids ( VLCFA ) [4] , α-oxidation of branched chain fatty acids [5 , 6] , plasmalogen biosynthesis [7 , 8] , and also participate in the metabolism of reactive oxygen species [9] and glyoxylate [10 , 11] . Peroxisomes are formed by the action of 14 peroxins encoded by PEX genes , the majority of which are involved in translocation of peroxisomal enzymes into the matrix , with others designating peroxisomal membrane [12–15] . Human diseases due to autosomal recessive loss of function mutations in the PEX genes comprise a group of severe disorders known as peroxisome biogenesis disorders ( PBD ) with involvement of brain , bone , kidney and liver and death within the first year of life [1 , 2 , 16 , 17] . The peroxisome’s well documented role in β-oxidation of VLCFA and synthesis of ether lipids has led to considerable focus on lipid metabolism as the key pathogenic factor in disease pathogenesis in PBD [18] . The accumulation of VLCFA has been proposed as the primary pathway influencing severity and as a therapeutic target [19–21] . A more general alteration of peroxisomal lipids have been proposed as a developmental insult to the brain in PBD [22] . However , while the increases in VLCFA and loss of plasmalogens in peroxisomal metabolism are likely to be a significant part of the pathogenesis of PBD , other metabolic pathways are also likely to play a role . Indeed , patients with pathogenic variants in PEX2 [23 , 24] , PEX10 [25] and PEX16 [26 , 27] that allow survival into childhood or adulthood have been reported with very mild abnormalities in VLCFA metabolism , and plasmalogen biosynthesis . These studies suggest that additional or even distinct peroxisomal functions are involved in PBD pathogenesis . Peroxisomal biology is highly conserved across eukaryotes which has allowed this same genetic machinery to be studied across several model organisms [28] . In mice , studies of a spectrum of enzymatic and biogenesis defects in global and conditional knockouts has allowed insight into the role of peroxisomes in vertebrate tissues [29] . Severe early phenotypes affecting brain , growth , and viability have been observed in Pex2 , Pex5 and Pex13 knock-out mice [30–32] . In addition a Pex1 knock-in for a common missense allele in human PBD produces mice with growth failure , cholestasis and retinopathy [33] . Pex genes have been shown to have tissue specific effects . For example , an oligodendrocyte-specific loss of peroxisomal biogenesis produces much of the axonal loss and demyelination seen in PBD suggesting a cell autonomous role of peroxisomes in oligodendrocytes [34] . Hepatocyte knockouts produce effects on mitochondrial morphology and ER stress [35–37] . Several recent studies have also explored peroxisomal biogenesis in Drosophila demonstrating the evolutionary conservation [38–42] . Studies of Drosophila pex mutants demonstrated a role for VLCFA in interfering with spermatogenesis leading to infertility [42] . In addition , fly pex16 mutants have been shown to have locomotor defects , and shortened lifespan [41] . Collectively , the study of peroxisomes in flies and mice have provided compelling data that the function of peroxisomes in longevity , locomotion and metabolism are conserved from flies to man . A key question that has not been addressed by the previous fly studies is whether the phenotype due to loss of peroxisomes is determined by any pathways in metabolism beyond peroxisomal lipids . Indeed , a comprehensive metabolic profile of peroxisomal biogenesis mutants is lacking . Here we utilize genetics , transcriptional informatics and untargeted metabolomics to show that Drosophila pex mutants exhibit an unanticipated defect in sugar metabolism and are sensitive to reduced dietary sugar . We also find a strong transcriptional co-regulation between peroxisomal genes in the fly and enzymes in glucose metabolism , and show that similar transcriptional signatures are observed in mice .
To perform detailed phenotypic and biochemical analysis and identify new pathways that may be affected when peroxisomal function is lost , we studied two genes required for Drosophila peroxisomal biogenesis . We selected pex16 and pex2 because both are conserved PBD disease gene homologs that act in distinct steps of peroxisomal biogenesis ( Fig 1 ) ensuring that our results will uncover key , cell biological aspects of peroxisomes rather than gene specific ones . While pex16 is involved in early peroxisomal membrane formation , pex2 is a component of an ubiquitination complex that functions in matrix protein import [43] . In order to study loss-of-function mutants for pex2 we obtained an EPg element [44] , a P-element insertion in the fourth coding exon of the pex2 gene ( Fig 1A ) . However , this allele has not been characterized in detail: it is not known if it corresponds to a null allele , nor have the phenotypes been rescued with a transgene . We therefore performed imprecise excision and produced two additional loss-of-function alleles ( pex21 , a 473 bp deletion and pex22 , a 599 bp deletion ) . For pex16 we studied a EPgy2 P-element insertion allele [45] in the 5’UTR of the pex16 gene which behaves as a hypomorphic allele [42] ( Fig 1B ) . We also obtained a pex16 deletion mutant that lacks the coding region of the gene [41] . For each strain we studied the mutant alleles in trans with a genomic deficiency and used genomic rescue constructs to rescue the phenotypes and to ascertain that the observed phenotypes are not due to second site hits ( S1 Fig , see Materials and Methods ) . To investigate the peroxisomal phenotypes we examined the mutant larval salivary glands using the Pex3 antibody [38] and a peroxisomally localized green fluorescent protein ( GFP-SKL ) ( Fig 1A’ ) . The UAS-GFP-SKL construct was driven by the ubiquitous strong driver Act5C-GAL4 . GFP-SKL produces a GFP with PTS1 targeting signal for the peroxisomal biogenesis machinery encoded by the pex genes necessary to import the protein into the peroxisomal matrix [46] . In Drosophila , PTS1 is the only system that allows localization into the peroxisomal matrix as PTS2 proteins are not present[39] . In control and rescued animals we observed microscopic punctae in which GFP-SKL and Pex3 extensively co-localize ( Fig 1A’ and 1A” rescue shown ) . However , pex2 mutant cells exhibited a mostly nuclear distribution of GFP-SKL and severely reduced Pex3 staining ( Fig 1A’ and 1A” ) . Similarly , in pex16 mutants we observe diffuse localization of GFP-SKL and reduced Pex3 signal compared to rescue ( Fig 1B’ and 1B” ) . Our data suggest that loss of either gene causes defective localization of peroxisomal markers and hence are likely to disrupt peroxisomal biogenesis/function . Therefore these genetic tools permit comparison of pex2 and pex16 mutants , two mutants that represent defects in different steps of the peroxisomal biogenesis pathway ( Fig 1C ) . To assess if peroxisomal function was altered in our mutants we examined canonical peroxisomal lipids using the same methods as employed for clinical diagnosis [47–49] . We used Gas chromatograph mass spectrometry ( GC-MS ) to measure very long chain fatty acids in larvae and whole adult flies [49] and observed only mildly increased levels of C30:0 , C28:0 , C26:0 and C24:0 in larval pex2 mutants as well as pex16 mutants ( Fig 2A ) . However , we observed a dramatic increase in these lipids in adult flies compared to larvae in the mutants ( Fig 2A ) . We also measured the levels of lyso-phosphatidylcholines ( LPC ) , C16:0 to C26:0 LPCs , and the 4 phosphatidylethanolamine plasmalogen ( PlsEtn ) species ( 16:0p/20:4 , 16:0p/18:1 , 18:1p/20:4 and 18:0p/20:4 ) in third instar larvae and found increased 24:0 LPC levels and decreased 18:1p/20:4 PlsEtn levels by LC-MSMS [47 , 48 , 50] ( Fig 2C and 2E , S1 Table ) . These assays were conducted along with human samples from controls , and patients with PBD ( Fig 2B and 2D , S1 Table ) . We noted a dramatic difference between the mutant animals compared to control and rescue ( Fig 2C and 2E , S1 Table ) . The 24:0 LPC accumulates quite dramatically concomitant with a severe loss of plasmalogen in both the pex2 and the pex16 mutant larvae , consistent with a defect in VLCFA catabolism and plasmalogen synthesis , defects that are seen in PBD . However , interestingly these biochemical defects are distinct from those observed in patients with PBD as the C26:0 LPC was not increased in the mutant larvae whereas in humans with PBD C26:0 LPC shows the greatest increase when compared to controls ( S1 Table ) . In addition , in PBD levels of all 4 PlsEtn are typically reduced . These data indicate that a fundamental role for peroxisomes in VLCFA catabolism and plasmalogen synthesis is conserved but some of the specific lipids affected differ between flies and humans . We next explored the phenotypes of the pex mutant flies . Drosophila pex2 mutants have been reported to have spermatogenesis defects without appreciable nervous system phenotypes [42] , while pex16 mutants have been noted to have shortened survival and locomotor defects [41] . Given the difference in reported phenotypes , we undertook an extensive phenotypic characterization of both pex2 and pex16 mutants . We noted that the survival of both pex2 ( Fig 3A ) and pex161mutants ( Fig 3B ) is dramatically shortened to approximately 50% of controls . For the pex16EY mutants we observed intermediate survival consistent with this being a hypomorphic allele ( Fig 3B ) . We also observed that young ( 2–3 day ) flies performed poorly in a bang sensitivity assay in which they were subject to mechanical stress and allowed to recover ( Fig 3C , S1 Video ( pex2 control at left , pex22 at center and pex22 Rescue at right; S2 Video pex16 control at left , pex161at center , and pex161 Rescue at right ) . Moreover , the majority of pex2 and pex16 mutant flies are unable to fly at 10 days of age ( Fig 3D , S3 Video- pex22 flight assay compare to S4 Video pex22 Rescue flight assay; S5 Video- pex161 flight assay compare to S6 Video pex161 Rescue flight assay ) . To characterize the activity of the pex2 and pex16 mutant flies we monitored their activity with the Drosophila Activity Monitor ( DAM ) assay [51] . In this assay activity is constantly monitored by an infrared beam for a continuous 3–7 day period in 12hrs light/dark cycles . We noted that 2-3-day-old pex2 and pex16 mutant flies had a dramatic reduction in activity ( Fig 3E ) . In addition , we assessed the function of the Drosophila photoreceptors in the pex mutant animals with electroretinogram ( ERG ) recordings ( S2 Fig ) . The amplitude of the ERG was measured as the size of change in potential occurring during depolarization , after the synaptic “on” transient and before the “off” transient ( S2A Fig ) . There was no significant difference in ERG amplitude in 2-day-old pex2 nor in 2-day-old pex16 mutants ( S2A and S2C Fig , quantification S2B and S2D ) . After 4 weeks of aging the animals in 12 hour light/dark cycle there was a significant reduction in ERG amplitude by approximately one third in pex22 , pex21 and pex161 ( S2E and S2G Fig , quantification S2F and S2H ) . Interestingly , the pex16EY allele did not exhibit a significant functional change . In summary , pex2 and pex16 mutants exhibit similar phenotypes with respect to viability , lifespan , bang sensitivity , flight , photoreceptor function , and locomotor activity . To assess metabolomic changes in two pex mutants we undertook a comprehensive metabolomic and characterization of the pex2 and pex16 mutants using untargeted metabolomics . We tested the adult mutant flies for 347 named analytes in distinct metabolic pathways . The metabolomic profiles across the 347 compounds is distinct as well as overlapping for pex2 and pex16 mutants when compared to control and rescue genotypes ( Fig 4A and 4B , S2 and S3 Tables ) . To determine which pathways were enriched for altered metabolites in peroxisomal biogenesis mutants we performed a Metabolite Set Enrichment Analysis ( MSEA ) [52 , 53] . MSEA is analogous to Gene Set Enrichment Analysis ( GSEA ) in which a set of metabolites is explored for specific biochemical pathways that are enriched for alterations . We selected lists of metabolites that were altered in pex2 mutants ( both alleles ) as well as pex16 null mutants ( pex161 allele ) ( S3 Fig ) . We also selected the subset of metabolites which were 1 ) altered consistently between pex2 ( both alleles ) and pex16 ( pex161 allele ) compared to rescue ( Fig 4C , S4 Table ) . The pathways implicated by MSEA included a broad range of processes . For example , pathways such as RNA transcription were identified owing to abnormal levels of adenosine monophosphate and uridine 5’ monophosphate in the pex mutant flies ( Fig 4C , S4 Table ) . The examination of biochemical pathways verified several known peroxisomal pathways ( Fig 5 ) . For example , Omega oxidation of fatty acids can be affected by loss of β-oxidation in peroxisome[54] ( Fig 5A ) , and synthesis of ether lipids ( plasmalogens ) ( Fig 5B ) were perturbed in the pex2 and pex16 mutants . Strikingly , dicarboxylic fatty acids of different chain lengths were among the most severely ( 5–10 fold increased from mutant to rescue lines ) and consistently deregulated metabolites ( Fig 5A ) . Moreover , defects in purine catabolism were also consistent in the pex2 and pex16 mutants [55] ( Fig 5C ) . Therefore , the untargeted metabolic profiling identified defects in a number of pathways already implicated in peroxisomal biochemistry . Our data also revealed an increase in precursors for phospholipids such as cytidine 5’-diphosphocholine and phosphoethanolamine , while observing decreases in the degradation products of phospholipids such as glycerophosphorylcholine and glycerophosphorylethanolamine in the pex2 and pex16 mutants ( Fig 5D ) . This suggests a defect in synthesis and a reduction in the breakdown of membrane lipids such as phosphatidylcholine and phosphatidylethanolamine . Interestingly , the results of MSEA in the group of metabolites consistently altered in both pex2 and pex16 mutants pointed to a number of analytes in carbohydrate metabolism ( Pentose phosphate pathway , glycolysis , and starch and sucrose metabolism ) . Indeed a number of compounds in these pathways were altered in the pex mutant flies ( Fig 6A ) . Since mitochondrial function could underly changes in carbohydrate metabolism and peroxisomes and mitochondria have a number of functional links[56] , we examined mitochondria in the pex mutant flies ( S4 Fig ) . We performed Transmission electron microscopy ( TEM ) on aged photoreceptors of the hypomorphic pex16EY flies at 2 weeks ( S4A–S4F Fig ) . We observed a statistically significant increase in the number of mitochondria per photoreceptor ( S4B and S4E Fig compared to S4A and S4D ) . In addition we observed electron dense inclusions in the photoreceptors of the pex16EY animals ( S4E Fig white arrows , S4F ) . These data demonstrate an increase in mitochondrial numbers in pex16EY photoreceptors . In addition we examined the function of mitochondrial complexes in purified mitochondria from pex22 compared to y w , and pex22 Rescue ( S4G Fig ) . While some minor differences in individual complexes , there were importantly no dramatic reductions in the ETC complexes in the pex2 mutant flies ( S4G Fig ) . While there was a statistically significant reduction in pex2 mutant flies in complex IV , this was a small , 20% reduction in activity ( S5 Table ) . Taken together we did not observe dramatic differences in mitochondrial function in the pex mutant flies . Based on this finding , we examined the carbohydrate metabolism pathways themselves more carefully in our pex mutant flies and noted a pattern of reduced glycolytic intermediates such as glucose-6 phosphate ( ratio of 0 . 33–0 . 43 in pex2 mutants compared to rescue ( p<0 . 05 ) , 0 . 27–0 . 38 in pex16 mutants compared to rescue ( p<0 . 05 ) ) . We also noted a reduction in glycogen intermediates such as maltose , maltotriose , maltotetraose , maltopentaose and maltohexaose ( Fig 6A ) . Given these changes we sought to test whether the peroxisomal mutants would be sensitive to reduced glucose intake . We tested flies for lifespan and locomotor activity on a low-sugar food providing 21 kcal/100 mL with 0% of calories from sugar , compared to a standard sugar food with 55 kcal/100 mL and 88% of calories from sugar . The standard sugar food was most closely related in composition to standard media but is providing nearly 90% of calories from carbohydrate ( S5A and S5B Fig ) ( see Materials and Methods ) . We noted that the lifespan of both pex mutants was dramatically reduced when the flies are raised on low-sugar food ( Fig 6B and 6C , S5C Fig ) . We also tested the pex2 and pex16 mutants on low-sugar in the DAM assay to score their activity level in these conditions . We observed that control and rescue flies reduce their activity level on low-sugar food . Interestingly , we noted an increase in activity of the mutant flies in the low-sugar food suggesting an altered behavior on low-sugar food ( Fig 6D ) . This increased activity was a somewhat surprising response given the evidence for depletion of glycolytic intermediates and glycogen in the pex mutants . Taken together these results suggest a strong physiological carbohydrate dependence in pex2 and pex16 mutants in vivo which is consistent with the metabolomic analyses . We also grew adult flies on agar media without nutrients to determine their sensitivity to starvation and measured their activity in the DAM assay during starvation ( Fig 7 ) . We noted that both pex2 and pex16 mutants are sensitive to starvation ( Fig 7A and 7B ) . In the DAM assay ( Fig 7C–7F ) the pex2 and pex16 mutants , although still severely impaired , display a doubling of their activity under starvation ( Fig 7C and 7E ) . These data show an increase in activity during starvation for the pex mutants that is more pronounced than in controls . Taken together these data suggest that the pex mutants respond differently than controls to changes in carbohydrate supply . In addition , pex mutant flies are sensitive to reduced glucose in the media and to starvation , and under both conditions these flies have a an increase in their activity level consistent with foraging behavior . The enzymes of glycolysis , glycogen catabolism and pentose phosphate pathway are not present in peroxisomes [57] . We therefore sought to explore the relationship between genes related to peroxisomal function and carbohydrate metabolism . We first assembled a list of peroxisomal genes in the fly [1 , 39 , 58] and examined their expression profile in existing databases to find genes whose expression correlates with the expression of these peroxisomal genes using g:Profiler tools [59 , 60] ( Fig 8; S6 Table ) . This approach revealed five distinct gene clusters containing peroxisomal genes in Drosophila ( Fig 8A ) . These gene clusters were used to create ranked lists from the Drosophila genome of genes whose expression most closely correlated with these gene clusters ( S6 Table ) . Next we examined the specific genes encoding enzymes in the glycolysis and glycogen metabolism pathways . Interestingly , for nearly every step of glycogen synthesis and glycolysis the gene encoding the enzyme is co-regulated with one or more of these peroxisomal transcriptional clusters ( Fig 8B ) . Taken together these results suggest extensive co-regulation of transcription between peroxisomal genes and carbohydrate metabolism genes in Drosophila . Next we sought to explore the conservation of this metabolic control in vertebrates . We undertook an analysis using the g:Profiler starting with a manually curated list of mouse peroxisomal genes which we analyzed informatically in liver transcriptional datasets in mouse ( S5 Fig , S7 Table ) [59] . We observed four distinct clusters of genes that are closely co-regulated . Since we noted a multiple genes implicated in glycolysis , pentose phosphate and glycogen metabolism in the Drosophila peroxisomal cluster , we tested whether carbohydrate metabolism genes were enriched in the peroxisomal gene cluster for murine liver and performed a Gene-set enrichment analysis ( GSEA ) for enzymes in carbohydrate metabolism to test whether there is a statistically significant enrichment in vertebrates for glucose metabolism within the peroxisomal gene clusters . We found a dramatic enrichment for glycolysis ( enrichment score 0 . 60 , P<0 . 0001 ) , and TCA cycle ( enrichment score 0 . 59 , P<0 . 0001 ) there is a dramatic enrichment ( Fig 9A and 9B ) . For genes in gluconeogenesis ( enrichment score 0 . 70 P = 0 . 022 ) , pentose phosphate pathway ( enrichment score 0 . 67 , P = 0 . 016 ) and glucose regulation ( enrichment score 0 . 67 , P = 0 . 039 ) there was a less striking enrichment but still a statistically significant difference . Interestingly , the genes in glycogen metabolism were not significantly enriched ( Synthesis , enrichment score 0 . 72 , P = 0 . 086 , Regulation , enrichment score 0 . 46 , P = 0 . 61 ) possibly reflecting a smaller number of genes . Therefore , in murine liver there is significant co-regulation of peroxisomal genes and genes involved in carbohydrate metabolism . Based on this pattern of similar co-regulation between peroxisomal genes and carbohydrate metabolism we hypothesized that there may be similarities between the carbohydrate metabolism defect we observed in Drosophila pex2 and pex16 mutants and those observed in the Pex5 liver conditional mouse [36] . Pex5 conditional liver knockout leads to altered glycolysis , glycogen production and pentose phosphate pathway . In addition , the Pex5 conditional liver knockout exhibits activation of AMP-activated protein kinase ( AMPK ) pathway and suppression of PPAR-γ and PGC-1α [36] . Of note , Drosophila do not have a PPAR-γ homolog but indeed we observe that the Drosophila gene clusters included several target genes for AMPK and PGC-1α with many AMPK targets appearing in Cluster 1 ( S8 Table ) [61 , 62] . In addition , amongst the top 1000 genes from each gene cluster we selected genes that are 1 ) involved in glucose metabolism and 2 ) represented in the top 1000 of more than one peroxisomal gene cluster in both fly and mouse . This selection identified 10 fly genes , corresponding to 12 mouse homologs in glucose metabolism that are strongly co-regulated with peroxisomal genes in both flies and mice . These genes catalyze 10 steps in glucose metabolism ( Fig 9C and 9D ) , 4 steps in glycogen metabolism and six steps in the TCA cycle . We hypothesized that this co-regulation in these key steps of carbohydrate metabolism might relate to the peroxisome’s dependence on other pathways such as the citric acid cycle to metabolize the end products of peroxisomal catabolism[63] . This would lead to the prediction of altered citric acid cycle metabolites in peroxisomal biogenesis mutants . We therefore undertook additional metabolomic studies in Pex5 mice [32 , 36] . Adult mouse liver was examined in the liver conditional Pex5 mice ( L-Pex5 mice ) [36] . The global Pex5 knockout mice were also examined but we could only test fetal liver because these mice usually die at P1 [32] . We used LC-MS to assess a targeted panel of glucose and TCA cycle metabolites in fetal and adult Pex5 liver tissue and found that for both fetal and adult Pex5 liver , the targeted platform could distinguish the mutants as exhibiting a distinct signature of TCA cycle metabolites ( S7 Fig ) Interestingly , we noted a clear correspondence between the significantly altered metabolites in the adult conditional Pex5 liver and the strongly co-regulated steps in glucose metabolism ( Fig 9D , purple analytes ) . For the fetal liver the only significantly altered metabolites , malate and citrate/isocitrate also corresponded to strongly co-regulated steps ( S7 Fig ) . Taken together these data indicated consistency between the transcriptional evidence for co-regulation of peroxisomes with the TCA cycle and metabolic abnormalities in PBD models . The genes are all enriched in multiple peroxisomal gene clusters in fly and mouse . In conclusion , the most strongly co-regulated steps of carbohydate metabolism with peroxisomal genes between fly and mouse correspond to metabolomic changes seen in liver from the Pex5knockout mice .
Peroxisomes are subcellular organelles tasked with a discrete subset of metabolic reactions principally involving peroxisomal lipids . While a role for lipid metabolism in peroxisomal disorders is well established , carbohydrate metabolism is thought to be a more central energy-producing process utilizing cytosolic and mitochondrial enzymes crucial for energy production and is not generally implicated in PBD . We have uncovered a previously unappreciated metabolic , phenotypic and gene-expression link between peroxisomes and carbohydrate metabolism . We identified co-regulation of peroxisomal genes and carbohydrate metabolic genes along with a carbohydrate dependence phenotype in peroxisomal biogenesis mutants with metabolomic studies in Drosophila . In addition , we link this defect to that observed in mouse liver tissue suggesting these pathways are conserved . Our studies in Drosophila represent a new approach to the studies of Drosophila pex mutants . We studied pex2 and pex16 mutant flies in order to compare different biogenesis defects in the fly . Previous studies did not report similar phenotypes in pex2 and pex16 mutants which was surprising given their strong conservation and similarities in yeast and vertebrates [41 , 42] . However , our study rigorously compared different genetic backgrounds by utilizing two alleles for each gene , studying mutants in trans with genomic deficiencies and creating genomic rescue strains for each mutation . We determined that peroxisomes are similarly functionally and morphologically defective in both pex2 and pex16 mutants . Our functional analysis of the peroxisomal lipids allowed a more comprehensive study of VLCFA metabolites in different stages of development . We also analyzed plasmalogens which had not been characterized previously in Drosophila . We observed a dramatic loss of PlsEtn 18:1/20:4 plasmalogen in pex2 and pex16 mutant flies which confirmed a conserved role for the peroxisome in plasmalogen biosynthesis pathway . We note that pex2 and pex16 mutants display very similar phenotypes including short lifespan , increased bang sensitivity , lack of flight and reduced activity . The consistency of our results could stem from our use of rigorous controls for genetic background , and allowed us to provide downstream metabolomic studies . Consistent with a shared peroxisomal biogenesis phenotype in Drosophila , metabolomic studies of pex2 and pex16 mutants revealed numerous shared metabolic abnormalities that are due to global peroxisomal dysfunction in Drosophila . Many of these were expected to result from peroxisomal dysfunction and had been seen in other organisms from previous targeted methods including fatty alcohols and purine metabolites[1] . However , our metabolomics analysis also revealed new insights indicating that the use of untargeted metabolomics in peroxisomal studies could uncover unsuspected metabolic pathways involved in peroxisomal biochemistry . Specifically we observed accumulations of phospholipid precursors and reduction in phospholipid degradation products . More importantly , we found that Drosophila peroxisomal biogenesis mutants have global reductions in glycolytic intermediates and glycogen with abnormalities in the pentose phosphate pathway . These unanticipated data allowed us to hypothesize that pex2 and pex16 mutants are starvation sensitive and sensitive to glucose deprivation in their diet which we experimentally verified . Of note they exhibit an increase in basal activity under both starvation and low glucose diet , while control flies exhibit starvation-related hyperactivity but seem to reduce their activity in low-sugar . These data demonstrate a particular sensitivity of peroxisomal mutants to glucose deprivation , suggesting that the metabolomics changes we observed in carbohydrate metabolism impact the physiology and behavior of peroxisomal mutant flies . This pathological hyperactivity is to some extent consistent with a starvation hyperactivity that has been observed in flies with altered adipokinetic hormones and octopamine levels [64 , 65] . There are several possible explanations for how peroxisomal biogenesis mutations can perturb carbohydrate metabolism . One is that some enzymatic components of glycolysis or glucose metabolism are in the peroxisome . Indeed , Trypanosomes have the entire glycolytic machinery within a peroxisome-like organelle called the glycosome[66] . While this is not supported by the current inventory of peroxisomal proteins in Drosophila [39] , mechanisms such as read-through at stop codons can lead to peroxisomal localization of some metabolic enzymes in yeast[67] . Another possibility is a secondary effect on metabolism , such as alteration in mitochondria . While we did not observe dramatic mitochondrial phenotypes in the mutant flies ( S5 Fig ) , we cannot rule out this possibility . In addition , we observe many metabolic abnormalities in our flies that are observed in Pex5 conditional liver knock-out mice [36] . Pex5 conditional liver knockout mice exhibit altered glycolysis , glycogen production and pentose phosphate pathway and these livers exhibit altered glucose regulation with AMP-activated protein kinase ( AMPK ) activation and PGC-1α suppression [36] . Indeed , we observed that target genes in the AMPK and PGC-1α pathway are represented in the gene clusters we identified in flies ( S8 Table ) . While we did observe more dramatic changes to the TCA cycle in the mice , and more dramatic changes in glycolysis in the flies , we see that in both fly and mouse liver peroxisomal biogenesis and carbohydrate metabolism interplay in the same transcriptional gene networks . These also allow us to identify additional pathways as the significant co-regulation may suggest that additional pathways beyond AMPK and PGC-1α may be involved in coordinating peroxisomes and carbohydrate metabolism . Recent studies have identified that the activation of the mTORC1 pathway in response to ROS occurs on the peroxisomal membrane , a process dependent on ATM [68 , 69] . Interestingly , some mTOR target genes such as eIF-4E and Cbp80 were also strongly co-regulated with peroxisomes in Drosophila ( S6 Table ) . We find that both flies and mice strongly co-regulate peroxisomal genes and carbohydrate metabolizing genes , and noticed that twelve vertebrate genes are co-regulated with the corresponding ten fly genes . Similarly , the metabolites corresponding to those steps in carbohydrate metabolism are altered in both species when pex/Pex genes are mutated . These data suggest evolutionary conservation of the link between carbohydrate metabolism and peroxisomal biogenesis genes . Our work adds to a growing realization that peroxisomal function is a process that interplays with other metabolic pathways , and suggests that PBD pathogenesis may extend into other metabolic pathways beyond peroxisomal enzymes [29 , 70] . Evolutionarily , ancestral peroxisomes were crucial in allowing ancient eukaryotes to detoxify the byproducts of oxygen and performing β-oxidation [56] . In higher eukaryotes β-oxidation occurs in the mitochondria with the exception of peroxisomal oxidation of VLCFA and some other carboxylates . Peroxisomes in these higher eukaryotes are responsible for steps in oxidative metabolism of lipids but they ultimately depend on the TCA cycle to fully metabolize the lipids [63] . Thus , the co-regulation of genes involved in carbohydrate and fat metabolism may have evolved through this interdependence of peroxisomes and mitochondria and maybe the basis of the strong inter-relationship we observed between peroxisomes and carbohydrate metabolism . Hence , examination of the latter pathways in humans may provide additional mechanistic insights and therapeutic targets for PBD . This dataset will provide valuable entry points for those studies . We have utilized Drosophila to identify a key role for glucose metabolism in pex2 and pex16 mutants . We find that the same key phenotypes observed in pex2 and pex16 , namely their longevity and their locomotor activity , were modifiable by diet and specifically were exquisitely sensitive to starvation stress and reduced glucose media . Interestingly , the Drosophila mutants are somewhat comparable to the findings in Pex5 conditional liver knockout mice which also have reduced glycolytic and glycogen intermediates , and diminished body weight despite increased food intake and carbohydrate dependence [36] . Finally , our work points to the importance of peroxisomal gene regulation in understanding not only peroxisomal biology but also in understanding PBD . We observed closely co-regulated groups of genes in flies and mice . These gene networks were seeded with bona fide peroxisomal genes , but the data suggested close co-regulation between these peroxisomal genes and enzymes in other pathways of metabolism in other cellular organelles . This suggests that the fundamental link in metabolism between peroxisomes and other organelles is regulated . Because peroxisomes lack a TCA cycle but they ultimately rely on mitochondrial TCA cycle for complete oxidation of their metabolites , it seems likely that common transcriptional programs activate peroxisomes and TCA cycle components . Our work suggests that delineating these gene regulatory programs , and how they are altered in PBD is important to our understanding of how defective peroxisomal biogenesis impacts human health .
All flies were maintained at room temperature ( 21°C ) and except where otherwise noted experiments were conducted at room temperature . The pex21 , and pex22 lines were derived from imprecise excision of ( w[1118]; P{w[+mC] = EPg}pex2[HP35039]/TM3 , Sb[1] , see below ) these were then backcrossed 5 generations with y w: FRT80B and studied as: Except where otherwise indicated the 5 strains above each crossed to a genomic deficiency uncovering pex2 locus w1118; Df ( 3L ) BSC376/TM6C , Sb1 cu1 are labeled as pex2 Ctrl , pex22 pex21 pex22 Rescue and pex21 Rescue respectively . The y w: pe161 line [41] was obtained from Kenji Matsuno , derived from: y1 w67c23; P{GSV6}pex16GS14106/TM3 , Sb1 Ser1 . The y w: pex16EY strain was obtained from Bloominton Stock center y[1] w[67c23]; P{w[+mC] y[+mDint2] = EPgy2}Pex16[EY05323] . Except where otherwise indicated the 5 strains above each crossed to a genomic deficiency uncovering the pex16 locus w1118; Df ( 3L ) BSC563/TM6C , cu1 Sb1 are labeled as pex16 Ctrl , pex161 pex16EY pex161 Rescue and pex16EY Rescue respectively . w[1118]; P{w[+mC] = EPg}pex2[HP35039]/TM3 , Sb[1] was crossed to y w; CyO , delta2-3/Egfr and progeny crossed to y w; D/TM6B , Tb . Progeny were screened for loss of w+ marker . 461 independent excision lines were screened by PCR . pex2-1 a 7bp insertion within a 606 bp deletion; and pex2-2 a 11bp insertion within a 473 bp deletion were selected . y[1] w[*]; P{w[+mC] = Act5C-GAL4}25FO1/CyO , y[+] second chromosome Actin GAL4 was recombined with a 2nd chromosome UAS-GFP-SKL transgenic ( courtesy of Hamed Jefar-Nejad ) . Recombinants were scored by GFP expression and balanced over CyO . These strains were crossed into deficiency strains for pex2 and pex16 marker experiments . For pex2 a genomic clone , CH322-53M21 from the CHORI-322_BAC collection was obtained . For pex16 a genomic clone , CH322-115M13 from the CHORI-322_BAC was obtained . These plasmids were amplified and grown , then purified and injected into y[1] w[1118]; PBac{y[+]-attP}VK00037 embryos ( VK00037 contains an attP on 2L at 22A3 ) . Transformants were selected based on the w+ marker and these strains were balanced to generate: These lines were crossed into the mutant strains listed above . Very long chain fatty acid levels were measured as described by GC/MS [49 , 71] . lysophosphatidyl choline ( 26:0 LPC , and 24:0 LPC ) and individual ethanolamine and plasmalogen species by LC-MSMS [47 , 48 , 50] . For salivary gland staining , dissection of third instar larvae was performed and larvae were fixed in 3 . 7% formaldehyde for 20 min at room temperature and washed in PBS containing 0 . 4% Triton X-100 . The primary antibody were used at the following dilution: chicken anti-GFP 1:1000 ( AB13970 , Abcam , Cambridge , MA ) , rabbit anti-pex3 1:500 ( From McNew lab , Rice U ) . Donkey anti Chicken Alexa 488 conjugated ( #703-545-155 , Jackson ImmunoResearch , PA ) and Goat anti-rabbit Cy3 conjugated secondary antibodies ( #111-165-003 , Jackson ImmunoResearch , PA ) were used at 1:250 . DAPI ( D1306 , ThermoFisher ) was used at 300nM . Samples were mounted in Vectashield ( Vector Labs , Burlingame , CA ) . Flies were collected under CO2 between 1 and 24 hours after eclosion . Male and female flies were separated and flies were kept 10 flies per vial at room temperature and the fly food was changed every 3 days . A tally of number of live flies was kept , and the number of live flies was checked every 3 days until the last fly had died . Data was analyzed with a Kaplan-Meier survival curve . Flies were kept without exposure to C02 for at least 48 hours prior to the assay . Flies were vortexed for 10 seconds in a vial , and a video recording was made of each trial . Video recordings were analyzed separately and blinded to genotype for recovery time for each fly . Flies were kept without exposure to C02 for at least 48 hours prior to the assay . Flight assay was performed in 10 day old flies . Flies of the 10 indicated genotypes were lightly tapped into a clear column made of PVC marked at each centimeter . Video recordings of each trial were made and analyzed separately and blinded to genotype for the landing height of each fly . A Drosophila activity monitoring system was used to study activity as described[51] . Briefly , adult flies were placed in small tubes with food , and kept at 25°C in a 12 hour light/dark cycle for 24–48 hours before being placed in the monitor . The monitor was temperature-controlled at 25°C in a 12 hour light dark cycle . Flies were kept in the monitor for 3–7 days for the various experiments described . Flies of the indicated genotypes were aged for either 2 days or 4 weeks after eclosure in a 12 hour light dark cycle . Electroretinograms were performed as described[72] . Briefly , adult flies were glued to glass slides . A recording electrode was placed onto the surface of the eye while another reference electrode was inserted into the cuticle in the posterior portion of the head . The eyes were then exposed to controlled sudden flashes of white light and the response was recorded and analyzed using AXON-pCLAMP8 . TEM of photoreceptor terminal were performed on 2 week aged flies as described[73] . Briefly , fly heads or third instar larva were dissected and fixed at 4°C in 4% paraformaldehyde , 2% glutaraldehyde , 0 . 1 M sodium cacodylate , and 0 . 005% CaCl2 ( PH 7 . 2 ) overnight , post-fixed in 1% OsO4 , dehydrated in ethanol and propylene oxide , and then embedded in Embed-812 resin ( Electron Microscopy Sciences , Hatfield , PA ) . Photoreceptors were then sectioned and stained in 4% uranyl acetate and 2 . 5% lead nitrate . TEM images of PR sections were taken using a JEOL JEM 1010 transmission electron microscope with an AMT XR-16 mid-mount 16 mega-pixel digital camera . Mitochondrial Electron transport chain activity was measured on isolated mitochondria extracted as previously described[74 , 75] Each ETC complex activity was quantified as nmoles/min/mg protein , normalized to citrate synthase activity , and expressed as %control activity . ” ( S5 Table ) . We used an untargeted metabolomics platform through Metabolon . This platform uses ultrahigh performance liquid chromatography/electrospray ionization tandem mass spectrometery [76] . The raw analyte values in this platform are analyzed by performing a z-score calculation compared to mean and standard deviation in other clinical samples . Welch’s two sample t-tests are used to determine significant alterations and multiple comparisons are accounted for with the false discovery rate method . In all experiments except where otherwise indicated , flies were grown on a standard media comprised of Agar , molasses , corn meal , dried yeast , proprionic acid , methyl p-hydroxybenzoate . Food for starvation was comprised of 2% agar only . The conditional food conditions ( Figs 6 and 7 ) consist of agar , cornmeal , yeast ( Sigma ) , dextrose ( Sigma ) , sucrose ( Sigma ) , methyl para-hydroxybenzoate and propionic acid ( Fisher Scientific ) . Recipe for every 100ml has 0 . 6 grams agar , 0 . 5ml methyl para-hydroxybenzoate , and 0 . 75ml propionic acid in common . For other ingredients , low-sugar food has 7 . 5 grams yeast; standard sugar food has 6 grams cornmeal , 1 . 5 grams yeast , 5 grams dextrose , and 2 . 5 grams sucrose . Ingredients including agar , cornmeal , yeast , dextrose , and sucrose were added to a cooking vessel , mixed then brought to a boil then immediately covered , mixed and cooled under 70°C at which time methyl para-hydroxybenzoate and propionic acid were added . The food was then poured into vials for experiments . Drosophila and murine liver gene network analysis was performed using g:Profiler essentially as described [59 , 60] . Mouse liver samples were provided by Myrian Baes . All animal work was conducted according to the guidelines for humane treatment of animals . | Peroxisomes are organelles or component of cells that are involved in body chemistry for a number of specialized fats . Peroxisome biogenesis disorders ( PBD ) are a group of rare diseases in which patients have genetic defects in the synthesis of peroxisomes . These disorders affect multiple organs including the brain and liver . We used fruitfly ( Drosophila melanogaster ) to study the metabolism and genetics of peroxisomal biogenesis to gain insight into the disease process . We generated flies with genetic defects in peroxisomes , carefully characterized these flies finding that they are short-lived and have locomotor problems . We then applied global metabolic profiling in these flies , measuring hundreds of biochemical compounds . The analysis pointed to an unexpected link between peroxisomes and sugar metabolism . Guided by this we found our flies were sensitive to low-sugar diet . We then used gene-expression analysis and targeted biochemical profiling in mouse to confirm that carbohydrate alterations also occur in vertebrates . Our work suggests carbohydrate metabolism may be a crucial process to study in patients with PBD . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"carbohydrate",
"metabolism",
"invertebrates",
"glucose",
"metabolism",
"animals",
"animal",
"models",
"metabolomics",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"mitochondria",
"bioenergetics",
"peroxisomes",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"research",
"and",
"analysis",
"methods",
"metabolic",
"pathways",
"insects",
"arthropoda",
"biochemistry",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"energy-producing",
"organelles",
"metabolism",
"organisms"
] | 2017 | Peroxisomal biogenesis is genetically and biochemically linked to carbohydrate metabolism in Drosophila and mouse |
In comparative genomics one analyzes jointly evolutionarily related species in order to identify conserved and diverged sequences and to infer their function . While such studies enabled the detection of conserved sequences in large genomes , the evolutionary dynamics of regulatory regions as a whole remain poorly understood . Here we present a probabilistic model for the evolution of promoter regions in yeast , combining the effects of regulatory interactions of many different transcription factors . The model expresses explicitly the selection forces acting on transcription factor binding sites in the context of a dynamic evolutionary process . We develop algorithms to compute likelihood and to learn de novo collections of transcription factor binding motifs and their selection parameters from alignments . Using the new techniques , we examine the evolutionary dynamics in Saccharomyces species promoters . Analyses of an evolutionary model constructed using all known transcription factor binding motifs and of a model learned from the data automatically reveal relatively weak selection on most binding sites . Moreover , according to our estimates , strong binding sites are constraining only a fraction of the yeast promoter sequence that is under selection . Our study demonstrates how complex evolutionary dynamics in noncoding regions emerges from formalization of the evolutionary consequences of known regulatory mechanisms .
Genomic regulatory regions harbor complex control schemes that collectively allow the genome to operate in a flexible and dynamic fashion . Such control schemes are encoded into the DNA sequence in a way that is not yet fully understood . Important elements of such regulatory code are short DNA sequences that are bound by transcription factors ( TFs ) . TFs bind regulatory DNA specifically , by recognizing short motifs , and contribute to the assembly of complex switches that govern the transcription of a gene , given various environmental or internal signals . Much of the current understanding of the way in which DNA determines the regulatory program of a gene is based on identification of TF binding sites ( TFBSs ) and their association with TFs of known function . Despite remarkable progress in functional genomics technologies , and in the ability to experimentally profile TF–DNA interactions on a genomic scale [1 , 2] , the understanding of function in regulatory regions remains a major challenge . At the same time , the complete sequencing of evolutionarily close genomes has made the detailed comparative study of regulatory regions possible . Consequently , comparative genomics has emerged as one of the central ways by which regulatory signals are computationally detected and studied . All comparative methods assume ( explicitly or implicitly ) an evolutionary model that distinguishes neutral sequences from functional ones . Most commonly [3–7] , comparative studies focus on conservation , classifying sequences to be functional or nonfunctional by assuming that evolution in functional loci is slower . In yeast , many conserved loci were shown to correspond to TFBSs , allowing detection of novel sites that were not identifiable using single species methods . As more species are sequenced , a desirable challenge is to extend the simple conservation-based studies by adding more structure to the function–evolution relationship in regulatory regions . In coding regions , our understanding of the genetic code makes sophisticated evolutionary predictions possible , e . g . , by identifying cases of positive selection [8] , correlated residues [9] , and more . It is hoped that by acquiring better , more detailed understanding of the function encoded by regulatory loci , one can greatly extend the utility of comparative studies in a similar way . In this study , building on simple assumptions of the mechanisms of transcriptional regulation , we formalize an evolutionary model combining a neutral mutational process with selection on multiple heterogeneous TFBSs . We develop techniques for computing the likelihood of such a model given pairwise alignments and for learning maximum-likelihood model parameters . Using the new techniques , we can express a substantial part of the current functional knowledge on gene regulation in evolutionary terms and evaluate observed patterns of divergence and conservation based on this model . Applying our method to promoter sequences of Saccharomyces yeast species , we validate our approach and exemplify its use . Specifically , we discuss how the selection on binding sites of different TFs vary in intensity , and how some families of similar TFBSs are in fact divided into subgroups that are separated by selection . We compute the fraction of promoter sequence that is under selection due to characterized TFBSs and show that strong TFBSs constitute only a small fraction of the promoter sequences in yeast . The gap between selection due to strong TFBSs and global estimates of the selection on yeast promoters can be used to estimate the relative roles of classical binding sites and of other effects ( low affinity transcriptional interactions , and possibly other factors , e . g . , chromatin organization ) in driving functional transcriptional networks .
We developed an integrated model for the evolution of promoters under the influence of heterogeneous TFBSs ( Methods ) . Briefly , the TF recognition code is a collection of distinct DNA motifs , where each motif ( corresponding to a TF ) is represented by a set of nucleotide k-mers . We assume that each set ( termed target k-mer set ) contains all k-mers recognized by the TF ( see Figure 1A for an illustration ) , and any appearance of a k-mer from the target k-mer set is declared to be a binding site . All k-mers in the same target k-mer set are of the same length , but are otherwise unconstrained . In practice , target k-mer sets are usually variations over a consensus sequence . Our model represents a simplification of a much more complex biological reality , by assuming that binding at each locus is completely determined by the existence of a motif , and is either perfect or nonexisting ( therefore ignoring differences in binding affinity between k-mers of the same target k-mer set ) . These simplifications allow us to develop a model for which computation is practical , but should be carefully evaluated and eventually relaxed in future revisions of the model . To model the evolution of a promoter region , we assume that sequences are evolving neutrally , except for loci affected by selection on TFBSs . Each target k-mer set ( and therefore each TF ) is associated with a selection factor 0 ≤ σ ≤ 1 , which represents the relative fixation probability of a mutation introducing or eliminating a binding site ( note that the selection factor is not equivalent to the classical selection coefficient ) . Smaller σ values represent stronger selective pressure on loci bearing k-mers that belong to a given target set . Our model assumes that each appearance or loss of a TFBS is selected against . The replacement of a k-mer from a target set by another k-mer from the same target set is not selected against , since according to our model all k-mers in the same target set are equivalent . This simple functional model allows us , once equipped with a TF recognition code , to write down a Markov model representing the evolution of an entire promoter sequence . The evolutionary forces outlined in our model affect the rate of mutation at a particular single base if it is in the context of a TFBS . The evolution of one base can therefore depend on several adjacent bases , and the model formalizes this type of epistasis using the simple assumptions of TFBSs described above . Although the epistasis considered by the model is simple and spatially limited ( including only binding site k-mers ) , exact computation of the likelihood of a TF recognition code given a multiple or even pairwise alignment is very difficult and involves exponentiation of a 4l by 4l matrix , l being the sequence length . We developed algorithms for approximate calculation of the likelihood of a model , which provide us with a method for evaluating to what extent our model agrees with the patterns of divergence in the alignment . We score models by comparing their likelihood to that of a null model representing neutral substation rates on independent loci , deriving a log-likelihood ratio ( LLR ) score . Using these tools , we can search for maximum likelihood selection factors for a given recognition code , e . g . , based on available experimental information . We can also learn a recognition code de novo directly from alignments and study the collective evolution of a group of TFBSs in an unbiased fashion ( Methods ) . We note that our framework was not designed as an attempt to develop another TFBS motif finding algorithm , a problem that is already treated extensively in the literature [10] . We focused on the evolutionary dynamics of Saccharomyces gene regulatory regions . The yeast system has the advantage of many well-documented TFBS motifs and clearly identifiable promoters , and was used before in many studies of transcriptional regulation and its evolution [4 , 6 , 11] . We extracted pairwise alignments from multiple alignments of Saccharomyces sensu stricto species ( Methods ) . For example , the resulting alignments for S . cerevisiae–S . mikatae consisted of more than 900 , 000 aligned bases from the upstream regions of 3 , 503 genes , with 74 . 2% identity . Alignments of S . cerevisiae–S . bayanus and of S . cerevisiae–S . paradoxus were also used ( Table S1 ) . We started by constructing an evolutionary model from known TF binding models . We used the compendium of TFBSs composed by MacIsaac et al . based on extensive ChIP-on-chip data and literature review [12] . Out of 124 consensus sequences reported by the authors ( in IUPAC format ) , we chose those 94 that translated to target k-mer sets containing at most 512 k-mers each , and had at least five matches in the aligned S . cerevisiae–S . mikatae promoters . We constructed a model starting from an empty one , and incrementally attempting to add each of the 94 target k-mer sets ( in an arbitrary order ) . For each candidate target k-mer set in turn , we tentatively added it to the model and inferred an optimal selection factor for it . We then tested whether the expanded model with the added target k-mer set had a selection factor smaller than 1 and a higher model likelihood ( note that a target k-mer set with a selection factor equal to 1 would result in a log-likelihood ratio of 0 ) . If this was the case , the target k-mer set was accepted to the model . Otherwise it was rejected and not kept in the model . In total , we accepted 74 target k-mer sets ( 79% ) . Similar results were obtained for the other two yeast species . We next studied possible factors that contribute to acceptance or rejection of literature target k-mer sets in our model . According to our model , every appearance of a motif is considered to be functional and under selection . In reality , not all appearances are necessarily functional , and some may be functionally different than others . We examined the correlation between the number of appearances of k-mers from a target k-mer set in the data and the model acceptance rate . While 87% of the target k-mer sets with 0–149 hits were accepted , only 77% of target k-mer sets with 150–499 hits were accepted , and only 47% of the target k-mer sets with more than 500 hits were accepted . These results suggest that the specificity of some of the literature target k-mer sets may be too low to allow acceptance by our rather stringent model . To try to control for motif specificity in a systematic way , we next examined , for each TF , a model constructed using a limited dataset , containing only pairwise alignments of promoters that were found to be bound by that TF in ChIP-on-chip experiments ( using a p < 0 . 005 cutoff ) [2 , 13] . Since the set of ChIP-bound promoters is different for each TF , we could not construct a complete model in this case , but simply computed the LLR for each TF in a model containing a single k-mer target set . We call the resulting model the ChIP model . Out of 62 TFs with at least five hits in the ChIP bound promoters , 52 target k-mer sets had positive LLR at σ < 1 ( 84% ) , of which six were not accepted in the original model ( Spt2 , Ndd1 , Swi5 , Bas1 , Hap2 and Met31 ) . All of the target k-mer sets that were accepted in the original model but not in the ChIP model ( 27 in total ) had fewer than five hits in the ChIP data and therefore were not considered . In summary , although our model assumptions are simplistic , they are enough to roughly approximate the behavior of a large fraction of the known binding sites in the yeast genome . The cases of known TFBSs whose evolution is not well captured by the model are not resolved by restricting the analysis to experimentally verified TF targets , suggesting that the simple association between motifs and function does not hold for them . We next applied our model learning algorithm to construct a TF recognition code model de novo . By constructing a de novo model we were not hoping to discover new TFBS motifs , but rather to study the evolutionary dynamics of the yeast promoters given the selection on an unbiased set of putative TFBSs . The model was constructed automatically , considering gapless k-mers of width 6–12 as candidate target k-mer seeds ( Methods ) . The learning algorithm produced a model containing 62 target k-mer sets when executed on the S . cerevisiae–S . mikatae alignments ( see Figure 2 and Table S2 ) . The de novo target k-mer sets matched 45 distinct known motifs ( Methods ) . We note that most of the target k-mer sets that we learned are relatively specific , with no or limited redundancy , and that we preferred a larger model over a more stringent one , to allow global properties of the model to be explored . In our modeling framework , it can be assumed that each inferred target k-mer set represents a distinct function and that no two target k-mer sets that represent the same function coexist in the model . The reason for this is that substitutions between k-mers in two equivalent target k-mer sets of the same TF would be predicted by the model to be selected against ( multiplying the probability of neutral substitution by the selection factors of each of the target k-mer sets ) , while , in fact , substitution between redundant target k-mer sets must behave neutrally . This discrepancy should result in lower likelihood for a model that includes two redundant target k-mer sets instead of just their union . Looking at the results , we indeed see only a few cases of seemingly redundant target k-mer sets , each of which can be biologically rationalized as described below . In the first type of model redundancy , one target k-mer set contains substrings of another ( e . g . , GCGATGAGATG and CGATGAG in the PAC motif ) . This can be accounted for by the inaccuracy of the discrete binding assumption . If one target k-mer set represents strong ( more specific ) binding sites and the other represents weaker sites , then having two target k-mer sets with different selection factors improves likelihood . In this case , according to our model the selection on a locus with the more specific version is calculated as if it were part of binding sites for both target k-mer sets ( implying a de facto stronger selection on it ) . In contrast , the selection on a locus with the less-specific version would be affected only by the selection factor of one target k-mer set . This is exemplified in Figure 3A . In the second type of model redundancy , two k-mers from distinct target k-mer sets differ in one position ( e . g . , TTACCCG and TTACCCT in the Reb1 motif , TATTTATA and TATTTACA in the Rlm1 motif ) . In this case , the likelihood of a model in which the two target k-mer sets are combined into one is lower than the likelihood of the redundant model . This suggests that the separation between the two target k-mer sets is selected for , possibly since BSs from each set are functioning differently ( Figure 3B ) . Examples for such separation were argued for heuristically and demonstrated experimentally before [11 , 14] , but now we are equipped with the computational means to quantify such selection . As shown in Figure 3C , substitutions between target k-mer sets that are seemingly redundant can be directly shown to occur at a lower rate than expected using Z-score statistics ( Methods ) , as well as using the LLR of the redundant and combined models . The cases we observed include the previously discussed Reb1 motifs [11 , 15 , 16] and separation among variants of the still cryptic PAC motif . PAC targets are highly enriched in stress response genes [17] , but the mechanisms of PAC based regulation are not well characterized . We discovered two separated PAC-like families ( GCGATGAG and GAGATGAG ) that are significantly separated from each other . Interestingly , both variants of the PAC model tend to co-occur in the same promoters with the RRPE motifs ( co-occurrence Z scores of 15 . 5 and 15 . 6 ) , suggesting that they share a common mechanism rather than representing two distinct factors . An important characteristic of our model is the separation between background substitution rates and the selection factor on target k-mer sets ( σ values ) . Since we analyzed separately pairwise alignments of three different species with S . cerevisiae , and since these species differ significantly in their divergence times from S . cerevisiae , we can compare the σ values of the same TFBS obtained in each pair of species . We can attribute differences in such σ values to changes in selective pressure or to other TF-specific effects ( like divergence of the TF itself ) , rather than to different divergence times between species pairs or other background effects . According to the results ( Figure 4 ) the σ values of the same TFBS are similar across the different species pairs ( Spearman correlation values ranging around 0 . 9 ) , even though some species pairs are four times as distant evolutionarily [4–7 , 18] , suggesting that these values represent a quantity that is by and large independent of background divergence . We observed significant variability in the inferred selection factor of known TF motifs in the literature-based target k-mer sets . Many of the well-known TFs with low degeneracy target k-mer sets ( <8 k-mers ) had small σ values , suggesting specific binding and tight selection . Some examples are Reb1 ( 0 . 18 ) , Rpn4 ( 0 . 16 ) , Ume6 ( 0 . 03 ) , and Leu3 ( 0 . 12 ) . However , for other well-known motifs we derived much higher σ values . These include the CACGTG motifs ( Cbf1 , Pho4 , Tye7 , and Met28 ) ( 0 . 4–0 . 91 ) , Mbp1 ( 0 . 46 ) , Swi6 ( 0 . 54 ) , and Msn2/4 ( 0 . 54 ) . Interestingly , we inferred high σ values ( >0 . 35 ) for these TFs in the ChIP-restricted model , too ( see Table S3 for σ values computed for different ChIP thresholds ) . This suggests that the mild selection factors for these TFBSs are not primarily a side effect of false positives , since it is widely assumed that motifs in promoters that are also ChIP targets are very likely to be bound in vivo . One possible explanation for the reduced selection on some of the target k-mer sets may be that k-mers from these sets tend to appear in multiple copies in each of the promoters they regulate . We therefore examined the percentage of promoters with multiple hits for specific target sets . All the motifs mentioned above as having tight selection ( low σ ) appear exactly once in all of the promoters , while motifs with less tight selection are occasionally repeated in promoter regions ( Swi6 is repeated in 11% of its promoters , Pho4 and Tye7 in 7% , Msn2 and Mbp1 in 5% , and Msn4 in 5% ) . While the number of cases is too limited to reach a clear statistical conclusion on the relation between redundancy and selection , it can be hypothesized that for many TFs , redundancy may be high ( including multiple hits and possibly also low specificity binding sites ) , and that such redundancy can alleviate some of the selective pressure on individual loci . Another possible explanation to low selective pressure on the targets of some critical TFs may be that while some of the physical targets of these TFs are functionally essential and therefore under strong selection , other targets are evolutionarily transient and do not have a major functional role , although they bound specifically in vivo . This hypothesis should be further explored using experimental data on TF binding for additional yeast species . The evolutionary model we described implies three evolutionary regimes on motifs: k-mers can a ) be functional sites ( part of a target k-mer set ) , b ) be one substitution away from becoming a functional site ( boundary k-mers ) , or c ) be at a distance of two substitutions or more from any target k-mer set , and thus behave in a neutral manner ( background k-mers ) ( Figure 5A ) . According to our basic assumptions , only substitutions between target k-mers and boundary k-mers are subject to selection . Consequently , we predict functional sites to be highly conserved , and boundary k-mers to be slightly conserved—not due to functionality but due to possible selection against binding site emergence . As discussed above , in cases where our modeling assumptions are too restrictive , we may be classifying as boundaries certain k-mers that are in fact weak binding sites . In these cases , we expect some selection to act on substitutions between such boundary k-mers and background k-mers . To try to characterize the global effects of selection on boundary k-mers , we compared the degree of conservation of target k-mer set , boundary , and neutral k-mers in the literature-based model . This was done by testing how often motifs from each of these groups appear conserved , compared to what is expected given a neutral model . As shown in Figure 5B , the observed conservation of target k-mers is far above what we expect from a neutral model . A weaker but still significant increase in conservation is observed for boundary k-mers , possibly due to weak selection on binding site appearance , or more likely because of mild selection on weak but functional sites . We next examined the substitutions between target k-mers and boundary k-mers , and between boundary and background k-mers , using again the number of observed substitutions compared to the number expected by a neutral model . As shown in Figure 5C , substitutions between target k-mers and boundary k-mers are occurring much less than expected given the neutral model . We observe a slightly weaker , yet similar pattern for substitutions from boundary to background k-mers . At least some of the boundary k-mers in our model may therefore be functional and under some weak selection , forming together with a target k-mer set a TFBS recognition model that is more complex than our simple assumptions . Based on the literature model , 1 . 77% of the promoter sequence is covered by a TFBS . Using the de novo model , the fraction is 2 . 36% . These models may be extremely incomplete , but even using the entire repertoire of motifs in the MacIsaac study ( without conservation or LLR constraints ) , the fraction is only 3 . 24% . It is therefore reasonable to conclude that only a small fraction of the promoter sequences is under tight selection against losing high-specificity binding sites . Previous global studies on the selection on yeast promoters [19] estimated that about 30% of the sequence in S . cerevisiae is under selection . The gap between these estimates and the scarcity of TFBSs can be explained in several ways . Weak selection may affect low-affinity or weakly functional [20] boundary k-mers , and , in fact , when considering target set k-mers and boundary k-mers together , they cover 27 . 1% of the sequence in the literature-based model and 29 . 2% in the de novo model . Another possible factor contributing to the selection , which is not included in our model , are forces determining chromatin structure [18 , 21] .
In this study we introduced a new probabilistic model for the evolution of promoter regions that takes into account the combined effects of multiple TFs . We developed an algorithm for calculating the likelihood of a model given pairwise alignments of promoters of orthologous genes from two species . Additionally , we developed algorithms for learning maximum likelihood model parameters . We applied our algorithms to Saccharomyces promoter regions , first inferring a model that summarized previously characterized TF specificities in yeast into one principled evolutionary model . We then applied our methods to learn a full model from scratch . We analyzed the patterns of selection on promoter regions as revealed by these models . Specifically , we used our models to study the intensity of selection on TFBSs and to estimate the amount of promoter region under selection due to high specificity TFBSs . Given our results , it is evident that even on very short evolutionary time scales transcriptional regulation in yeast is highly dynamic . Indeed , the selection factors we computed for almost all TFBSs are higher ( less tight ) than what we might expect from functionally essential loci ( averaging around 0 . 5 ) . On average , the calculated selection seems to be weak , even if we restrict the analysis to functionally validated sites ( ChIP targets ) . On the other hand , we observed a significant gap between the amount of selection we can account for using characterized TFBSs and the overall reported selection on yeast promoters . Taken together , it can be hypothesized that much of the functionality of transcriptional networks is encoded in ways other than strong TFBSs , and that due to high levels of redundancy , binding sites are under continuous remodeling [22–25] . Rather than being a deterministic and sparse network , transcriptional programs may be shaped as dense , noisy networks that are continuously changing during evolution . Much of the past research on comparative methods for noncoding regions has focused on the evolutionary dynamics of TFBSs , as they have relatively well-defined features and a clear functional role . In addition to conservation-based methods for identifying TFBSs [4 , 5] , several studies introduced methods for detecting TFBS motifs using phylogeny-based probabilistic models that distinguish between the evolution of TFBSs and of the neutral background [26 , 27] . Other studies associated the evolutionary rate with the physical strength of TF–DNA interactions [11 , 15 , 16] . These studies strongly motivated the development of a general model for the evolution of regulatory regions in the presence of TFBSs . The more general approaches for context-aware molecular evolution were so far limited to modeling of neutral evolutionary processes [28–30] , or tailored to rigidly structured protein coding regions [31] , RNA coding genes [32] , or CpG dinucleotides [33] . The model we develop here is a step toward overcoming the major computational difficulties in handling the evolution of large regions with heterogeneous function ( many binding sites , sparsely and non-uniformly arranged ) . To make the model more realistic , additional effects will have to be considered , including binding sites with variable affinities , chromatin structure , combinatorial regulation , and more . Computationally , the adaptation of our methods for computing likelihood and learning models to general phylogenies will require solution of a difficult ancestral inference problem [34] . Analysis of more than two species will allow better understanding of the different dynamics associated with binding site gain and loss ( which cannot be distinguished based on pairwise alignments ) . We hope that further work on these challenges will open the way to faithful modeling of regulatory evolution in higher eukaryotes .
Define a TF recognition code to be a collection of sets C1 . . . Cm where each Ct is a set of words of length kt . Ct is called the target k-mer set of the t-th TF . Typically , Ct will consist of highly similar words . We define an indicator function βt ( s , i ) whose value is 1 if the i-th position in sequence s falls inside a word from target k-mer set Ct ( i . e . , if the substring s[i−j , . . , i−j+kt-1 ) ] ∈ Ct for some 0 ≤ j < kt ) , and 0 otherwise . Every occurrence of a word from Ct in the promoter sequence s is declared a binding site of the t-th TF . Our model therefore assumes that TFs recognize all loci-bearing words from their target k-mer sets , and no other loci . It also assumes that all words from the same target k-mer set behave identically , and that the target k-mer sets ( and therefore the DNA binding domains of the TFs ) remain constant during the evolutionary period considered . See Figure 1A for an illustration of a TF recognition code . A word of length kt that does not belong to any target k-mer set is called a boundary k-mer for TF t if it is one substitution away from some k-mer in Ct ( see Figure 1B ) . Given two aligned sequences s1 and s2 , we define a model of the evolution of s1 into s2 , based on the TF recognition code model introduced above . We assume that each nucleotide in the sequence is evolving independently with a neutral substitution rate , with the exception that substitutions that change the regulatory role of a nucleotide—either eliminating or introducing a TFBS—are selected against . We model neutral evolution using a standard instantaneous nucleotide substitution rate matrix Qb , defining Qb ( c1 → c2 ) as the neutral rate of substitution from nucleotide c1 to c2 . The effect of selection on TFBSs of the t-th TF is formalized using a selection factor 0 < σt < 1 . We assume that a substitution with neutral rate p has a reduced rate σt p whenever it adds or removes a binding site for the TF t . The instantaneous rate of mutation at the i-th position of a sequence s is therefore defined by: where s′ is equal to s in all positions but i . To compute the probability of evolving from an entire sequence s to a sequence s′ , we have to combine the effects of multiple TFBSs and of the neutral process , taking into account the epistasis between nucleotides at nearby positions that code for the same TFBS ( Figure 1C ) . The interactions between loci make the common approach of decomposing the sequence into independently evolving loci impossible , since , for example , a mutation in one position that falls within a TFBS may abolish binding and completely alleviate the selective pressure on all other positions . To enable the computation in practice , we will rely on a parsimonious assumption that we outline below . We will show elsewhere how to derive this approximation from an unrestricted Markov model . We also note that our model is defined as symmetric and reversible , so the generalization to phylogenetic trees is direct . Given two aligned sequences s1 , s2 , we define the set S ( s1 , s2 ) as the collection of sequences ŝ such that for all positions i either ŝ[i]= s1[i] or ŝ[i]= s2[i] . Our simplifying approximation is that in the evolutionary trajectory between s1 and s2 , only sequences in S ( s1 , s2 ) have occurred . Those sequences are called parsimonious with respect to s1 , s2 . Given a TF recognition code , we say that two positions i , j are epistatic if there exists a state s′ ∈ S ( s1 , s2 ) and a TF t such that s′[i] and s′[j] are part of the same appearance of a k-mer from Ct , or a boundary k-mer for TF t . An epistatic block is defined to be a maximum interval in the alignment in which every two adjacent positions are epistatic . The simplest epistatic block is a single neutral nucleotide , which does not interact with any TF in the extant sequences or in any parsimonious trajectory between them . The next basic case is that of an interval including exactly one TFBS ( compare Figure 1C ) . In general , when there are several sites overlapping each other , the epistatic blocks define the smallest possible units for which we can compute the model likelihood independently . It can be shown that under the parsimonious assumption , the probability of s1 evolving into s2 equals the product of the probabilities of evolution in each of the epistatic blocks . Working inside an epistatic block , we still have to compute the probability of evolving from s1 to s2 in time t using only sequences in S ( s1 , s2 ) . This can be done by constructing a continuous time Markov model on all the parsimonious states of the block and an additional state designated OUT which absorbs all probability of transitions to nonparsimonious states . The total probability can then be computed using exponentiation of the model's rate matrix [35] . In practice , we further approximate the matrix exponential using a time-quantized Markov model as follows: Define a time step dt as t/L where L is larger than the number of point mutations between s1 and s2 . The background mutation probabilities Pb for the time step dt are computed by exponentiation of the background rate matrix Qb . We define model states ut , s for times t = 0 … L−1 and sequences s ∈ S ( s1 , s2 ) and add the special state OUT . The transition probabilities Pr ( ut , s → ut+1 , s′ ) are defined as We complete the transition probability from each state to 1 by adding an appropriate transition to the state OUT . Using this model , we can approximate the probability in one epistatic block by standard dynamic programming in the discrete Markov model . In summary , to compute the likelihood of a model ( target k-mer sets and selection factors ) given a set of pairwise alignments , we work in phases ( see Figure 1D ) . First we partition the alignment into epistatic blocks by searching for target k-mer set or boundary k-mers in the aligned sequences and their parsimonious combinations . We then compute the log-likelihood of each block using the discrete Markov model , and sum the contributions . Note that in a typical scenario , a substantial fraction of the sequence is neutral with respect to the model , which translates to epistatic blocks of length one . When computing the log-likelihood ratio of some model versus the null model , we can ignore all of these single-nucleotide blocks . Note that to compute log-likelihood ratios , we apply the time-quantized parsimonious approximation to both the null and the target models , thereby avoiding biases introduced by the approximation . To learn a maximum likelihood model given a set of pairwise alignments , we devised a multiphase greedy algorithm . Formally , given a set of pairwise alignments and assuming a background neutral substitution model Qb ( which we compute directly from the alignments ) , we wish to find target k-mer sets C1 . . . Cm and their selection factors σ1 . . . σm such that the model likelihood is optimal . The algorithm optimizes the selection factors given fixed target k-mer sets , and repeatedly attempts to add additional target k-mer sets and to refine existing ones . The key point in the implementation of the algorithm is in careful weighting of candidate motifs for extending the recognition code , since considering all motifs at each step of the algorithm is infeasible . A detailed description of the algorithm is available in Text S1 and Figure S1 . Additional details and data on simulation experiments is available in [36] . Additional methods are in described in Text S1 . | Cells use sophisticated regulation to transform static genomic information into flexible function . We are still far from understanding how such regulation evolves . Short DNA sequences that physically bind transcription factors in promoter areas near target genes play an important role in gene regulation and are directly subject to mutation and selection . In this work , we develop a methodology for studying the evolution of promoter sequences under the effect of multiple regulatory interactions . We present a model that describes the evolutionary process at each genomic locus , taking into account a random flux of mutations that occur in it and the effects of transcription factor binding sites gain or loss . Our model accounts for dependencies ( or epistasis ) between adjacent loci that contribute to the same regulatory interactions: mutation in one such locus immediately changes the effect of mutations in the other . Using our model , we characterize the evolution of promoters in yeast , showing that many regulatory interactions that were discovered experimentally or computationally are evolutionarily unstable . The dynamic nature of transcriptional interactions may be explained if the regulatory phenotype is achieved through multiple interactions at different levels of specificity , and if only relatively few of these interactions are essential for themselves . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"transcriptional",
"regulation",
"computational",
"biology",
"evolutionary",
"biology",
"evolution",
"yeast"
] | 2008 | Evolution and Selection in Yeast Promoters: Analyzing the Combined Effect of Diverse Transcription Factor Binding Sites |
Currently there are few studies characterising the nature and aetiology of human schistosome-related inflammatory processes . The aim of this study was to determine the relationship between Chitinase 3-like 1 ( CHI3L1 ) , also known as YKL-40 , a molecule associated with inflammatory processes , and schistosome infection , morbidity and systemic cytokine levels . Serological levels of CHI3L1 and a panel of cytokines ( IFN-y , IL-4/5/6/9/10/13 and 17 ) were measured in two Zimbabwean populations resident in a high and low schistosome infection area . CHI3L1 levels were related to schistosome infection , haematuria status and cytokine levels after allowing for confounding variables . The effect of antihelminthic treatment with praziquantel on CHI3L1 levels was determined in 246 participants 6 weeks post-treatment . CHI3L1 levels increased with age in both areas but were significantly higher in the high infection areas compared to the low infection area . CHI3L1 levels were also higher in infected compared to uninfected individuals with this difference being significant in the youngest age group . Curative antihelminthic treatment resulted in a significant decrease in CHI3L1 levels . Of the cytokines , only IL-10 and IL-17 had a significant association with CHI3L1 levels , and this association was negative . Serum CHI3L1 levels differ between infected and uninfected people before and after antihelminthic treatment . The greatest difference occurs in the youngest age group , in keeping with the period when schistosome-related pathological processes are initiated . Following from previous studies in non-infectious diseases showing that CHI3L1 is a biomarker for the inflammatory process , this study suggests that the potential for CHI3L1 as a biomarker for schistosome-related pathology should be explored further .
Urogenital schistosomiasis , caused by the blood fluke Schistosoma haematobium , is one of the world's most prevalent human parasitic diseases . Morbidity includes anaemia , malnutrition , impaired memory cognition and physical growth [1] , and immune-mediated pathology in the urogenital tract , the kidneys and ureters [2] . Immunopathology commences with the laying of eggs in the blood vessels at the site of infection where they cause damage to the bladder walls and genital tissues in the form of lesions in blood vessels and tissues leading to the common symptom of haematuria . Antigens secreted by these eggs induce a Th2 response , involving secretion of interleukin-4 ( IL-4 ) , IL-5 and IL-13 [3] , [4] and the eventual development of granulomas [5] , [6] . Resolution of the granuloma involves deposition of collagen and extra-cellular matrix components , which are the source of fibrosis [5] , [7] , [8] . In young children infected with S . haematobium egg deposition is associated with more acute inflammatory symptoms , visible haematuria , anaemia and a high egg output in what is known as the active stage of disease [7] . This progresses with age to an inactive stage characterised by a drop in urine egg counts as fewer eggs are excreted but more become trapped and calcified in tissues . It is during the inactive stage of disease that signs of extensive and sometimes irreversible fibrotic pathology can be detected [7] , while in those still acquiring active infection most of the schistosomiasis associated morbidity is reversible upon antihelminthic treatment [9] . Schistosome control programmes aim to reduce morbidity by reducing infection intensity and their success is monitored through egg counts [10] , [11] . However , egg counts as a measure of disease burden can be misleading as the relationship between morbidity and infection intensity can be non linear , for example there can be a discrepancy between an individual with a high worm burden displaying less pathology than an individual with a low worm burden [10] , [12] . Haematuria , has been used in numerous studies as both a marker for infection and onset of pathology [10] , it is particularly effective in detecting infection in younger age groups . However , haematuria can have a low sensitivity rate [13] . Ultrasonography provides a more reliable method for assessing pathology in schistosomiasis infection [10] but operational logistics , such as the requirement for trained personnel and a consistent power source , make it an unsuitable tool for large scale or rural field studies [14] . In addition , despite its effectiveness in detecting late stage disease , ultrasound can fail to detect the earlier pathological changes associated with infection mediated inflammation [15] . Identification of biomarkers that can detect early pathological changes related to schistosomiasis infection , and changes associated with treatment , would provide an invaluable tool in monitoring schistosome control programmes [11] . Chitinase-like proteins are a characteristic feature of multiple helminth infection models [16] , [17] . Chitinase 3-like 1 ( CHI3L1 ) , also known as human cartilage glycoprotein 39 ( HCgp-39 ) , and YKL-40 , is a human chitinase-like chitin-binding lectin with no chitinase enzymatic activity [16] , [17] . It is expressed by numerous cell types and has been associated with collagen and extra-cellular matrix formation [18] as well as with a wide range of diseases characterised by inflammation and tissue remodelling including asthma , arthritis , numerous cancers and liver fibrosis [19] , [20] , [21] . Recently CHI3L1 has also been linked to schistosome related hepatic fibrosis with S . japonicum infection [22] . In order to evaluate the potential of CHI3L1 as a pathology marker in urogenital schistosomiasis , and to investigate the nature of pathological changes associated with infection , the association of serum CHI3L1 levels with schistosome infection status , prevalence and history of exposure to infection was determined in individuals from three villages in rural Zimbabwe with differing levels of S . haematobium endemicity . Comparing villages of different endemicity allows a comparative approach separating the effects of history of schistosome infection from current infection levels , both factors that can potentially influence CHI3L1 levels [23] . The effect of curative antihelminthic treatment on CHI3L1 was determined in a post-treatment follow-up focussed on the village with highest endemicity . In addition we assessed the protein's relationship with a panel of systemic cytokines encompassing the range of Th1 , Th2 , Th17 and T-regulatory responses associated with schistosome infection and pathology ( see [24] , [25] for review ) , thus allowing CHI3L1 levels to be related to the immune environment in the host relative to their immune phenotype .
The study was conducted in the Mashonaland East Province of Zimbabwe where S . haematobium is endemic . Ethical approval was received from the University of Zimbabwe's Ethics Review Board ( UZERB ) and the Medical Research Council of Zimbabwe ( MRCZ ) , and permission to conduct the study was obtained from the Provincial Medical Director . The study design , aims and procedures were explained in the local language , Shona , prior to enrolment . Written consent was obtained from participants or their guardians prior to taking part in the study . The study was conducted within schools , recruiting school children and community members from three villages and details are described elsewhere [26] . Of the three villages Magaya ( 17°37′21″S 31°54′36″E ) and Chipinda ( 17°41′46″S 31°56′03″E ) are considered high infection areas ( HIA ) for schistosome infection , while Chitate ( 17°39′45″S 31°53′21″E ) is a low infection area ( LIA ) as defined by the World Health Organization ( WHO ) [11] . The area has a low prevalence of soil transmitted helminths [27] , and the residents are subsistence farmers with frequent contact with infected water for purposes of bathing , washing and collecting water . Plasmodium falciparum is the predominant species of malaria in Zimbabwe , and in this region malaria transmission is largely unstable in nature , classified as low and sporadic [28] , [29] , giving a low prevalence of co-infection between schistosome infection and Plasmodium infection in the study population . From each participant a stool and urine specimen was collected on three consecutive days and examined microscopically . Urine samples were examined for S . haematobium infection and stool samples were examined for S . mansoni and soil transmitted helminths using standard techniques [30] , [31] . Urine samples from children were examined for visible haematuria and designated positive when haematuria was observed or negative when no haematuria was observed . Macrohaematuria was confirmed on a different urine sample by dipstick ( Uripath , Plasmatec , UK ) . 5 ml venous blood was collected and processed using routine methods to collect serum which was frozen at −20°C prior to freighting to Edinburgh for serological assays . Following parasitology sample collection , participants were offered treatment with the antihelminthic drug praziquantel at the recommended dose of 40 mg/kg of body weight and treatment efficacy was checked 6 weeks later during follow-up parasitology and serological surveys . Treatment of the three communities was staggered by 6 weeks for the purpose of an untreated control group except in the case of heavily infected people as defined by the WHO [11] . In order to be included in this study participants had to meet the following criteria: 1 ) be lifelong residents of the study area to allow age to be used as a proxy for history of exposure to schistosome infection , 2 ) have provided at least two urine and two stool samples on consecutive days for parasite detection , 3 ) have provided a blood sample for serological assays , 4 ) not have previously received antihelminthic treatment , 5 ) be negative for co-infections ( malaria , intestinal helminths , S . mansoni and HIV ) . From an initial number of 2000 individuals , a total of 859 individuals 2–86 years old , 642 from the HIA and 217 from the LIA , met these criteria . The post-treatment study focused on the HIA where there was a higher prevalence of infection . To be included in the post-treatment study individuals had to meet the criteria above and be confirmed egg negative 6 weeks post-treatment or remain untreated for the control group . 246 participants ( 110 treated and 136 untreated ) fulfilled these criteria and formed the follow-up cohort . Serum levels of CHI3L1 were quantified using an ELISA kit from R&D Systems ( Cat DY2599 , Minneapolis , USA ) , following the manufacturer's instructions . Systemic levels of IL-10 , IL-17 , IFN-γ , IL-13 , IL-4 , IL-5 and IL-6 were measured by ELISA as previously described [32] , [33] . All serological assays were performed in duplicate with a positive control on each plate . Statistical analyses were conducted using the statistical package SPSS version 14 . 0 . To describe the age profiles of schistosome infection and CHI3L1 protein , an analysis of variance ( ANOVA ) was performed using infection intensity ( eggs/10 ml urine ) and CHI3L1 levels ( ng/ml ) log transformed ( log10 ( x+1 ) ) . The independent variables were age categorised into 6 groups as shown in Table 1 . Following these analyses , age was re-categorised to reflect the changing dynamics of rising ( 2–10 years ) , peaking ( 11–20 years ) and declining ( 21+years ) infection . Using these age groups the effect of schistosome infection ( infection intensity ( log transformed ( log10 ( x+1 ) ) or infection status ( infected vs . uninfected ) ) and age group on CHI3L1 protein was determined by ANOVA after allowing for the effects of sex and village of residence through sequential sums of squares . The interaction between infection and age groups was also tested . To determine if CHI3L1 levels were higher in children with visible haematuria , the positive cases were matched with age , sex and infection intensity matched controls and the data analysed by t-test . The effect of antihelminthic treatment on CHI3L1 levels was analysed using a repeated measures ANOVA followed by posthoc analysis within each age group . In order to relate systemic cytokine to CHI3L1 levels the number of cytokine variables was reduced using factor analysis . Individual cytokine responses ( square-root transformed ) with no rotation were used to extract uncorrelated principal components ( PC ) accounting for a significant proportion of variance in the data ( eigenvalues greater than 1 ) and with a factor loading of >0 . 5 or <0 . 5 for one or more of the original cytokine variables . The PCs arising from this procedure were then used in a partial correlation ( controlling for age group , sex and residential area ) relating the PCs to CHI3L1 levels with the 2-tailed p value from the correlation being reported . P values in all statistical tests were considered significant if p≤0 . 05 .
After accounting for sex schistosome infection intensity and prevalence were significantly higher in the HIA ( mean infection intensity = 27 eggs/10 ml urine , prevalence = 40% ) compared to the LIA ( prevalence = 11 . 1%; mean infection intensity = 2 . 7 eggs/10 ml urine ) ( F1 , 858 = 55 . 158 , p<0 . 001 ) . The schistosome-age profiles differed in the two residential areas resulting in a significant age group*residential area effect ( F5 , 858 = 2 . 446 , p = 0 . 033 ) giving rise to the ‘peak shift’ [34]; whereby the infection peak occurred at a higher intensity and at an earlier age in the HIA compared to the LIA ( as shown in Figure 1A ) . Overall , after accounting for sex , CHI3L1 levels were also significantly higher in the HIA ( mean , 59 . 9 ng/ml; SEM , 2 . 03 ) compared to the LIA ( mean , 49 . 1; SEM 5 . 15 ) , ( F1 , 858 = 59 . 722 , p<0 . 001 ) . In both villages CHI3L1 levels rose significantly with age group as shown in Figure 1B and Table 2 , but no peak shift occurred in CHI3L1 levels . CHI3L1 levels were higher in infected vs . uninfected people , after allowing for the confounding effects of age and residential area , with this difference being significant in the youngest age group ( Table 2 , Figure 2 ) . Interestingly , infection intensity , when analysed in the same model , did not significantly affect CHI3L1 levels ( F1 , 858 = 2 . 016 , p = 0 . 156 ) . The pattern was maintained , though was not significant , in the older age groups . Comparison of CHI3L1 levels in 34 children aged 6–16 years showed a higher level of CHI3L1 in the children with visible haematuria compared to the infection matched children who were not presenting with haematuria ( t = −1 . 1662; df = 32; p = 0 . 053 ) as shown in Figure 3 . Neither the mean age nor the mean infection intensity was significantly different between these 2 groups ( t = 0 . 067; df = 32; p = 0 . 947 for age , and t = −0 . 595; df = 32; p = 0 . 556 for S . haematobium infection intensity ) . To address the question of whether CHI3L1 is associated with infection , 246 participants from the HIA were followed up for 6 weeks post-treatment . All treated individuals included in this study were negative for infection 6 weeks post-treatment , as determined by egg counts . In treated people , mean CHI3L1 levels declined significantly from 63 . 7 ng/ml ( SEM , 3 . 06 ) pre-treatment to 48 . 2 ng/ml ( SEM , 4 . 75 ) post-treatment ( p<0 . 001 ) . In contrast , there was no significant change in mean CHI3L1 levels in the untreated group ( 47 . 3 ng/ml; SEM = 2 . 18 at baseline , and 52 . 3 ng/ml; SEM = 3 . 52 6 weeks later , p = 0 . 855 ) . Partitioning the cohort by age group showed that the post-treatment decline was significant in the youngest age groups ( 2–10 year olds: F53 = 23 . 942 , p<0 . 001; 11–20 year olds F51 , = 4 . 161 , p = 0 . 047 ) , while there was no significant change in the oldest age group ( F6 = 3 . 25 , p = 0 . 146 ) ( Figure 4 ) . The data reduction procedure conducted on the 8 cytokines returned 4 main PCs summarised in Table 3 . PC1 , accounting for the largest variation in cytokine levels , was composed of Th2-type cytokines ( IL-4 , IL-5 and IL-9 ) . PC2 was composed of T-regulatory ( IL-10 ) and Th17 ( IL-17A ) type cytokines . PC3 was composed of IL-13 , and PC4 was a pro-inflammatory grouping composed of the Th1 cytokine IFN-γ and the innate cytokine IL-6 . Correlating the extracted PCs from the factor analysis with CHI3L1 while controlling for sex , age group , infection area , and infection status revealed a negative association between PC2 and CHI3L1 levels ( r = −0 . 184 , p = 0 . 037 ) . All other PCs showed no significant correlation with CHI3L1 levels PC1 ( r = −0 . 006 , p = 0 . 934 ) , PC3 ( r = 0 . 028 , p = 0 . 755 ) , and PC4 ( r = −0 . 112 , p = 0 . 206 ) .
Infection with schistosome parasites causes a host inflammatory response responsible for much of the disease-associated pathology [3] , [4] , [5] . As CHI3L1 has been associated with numerous other inflammatory and type-2 diseases we investigated the relationship between CHI3L1 levels and schistosome infection in a population naturally exposed to S . haematobium [19] , [20] , [21] . Our study presents the first comparative analysis of CHI3L1 levels in human populations who are of the same ethnicity and socioeconomic status but differing helminth endemicity . Similar to reports from non-schistosome endemic areas , CHI3L1 showed a significant increase in levels with increasing age [35] , a pattern reported to be due to an increasing level of background inflammation associated with ageing and the onset of age-related diseases [35] , [36] . In this study we demonstrated that overall levels of CHI3L1 are higher in the HIA compared to the LIA across all age groups . As the populations are similar in ethnicity and in exposure to other common infections such as Plasmodium , the only differing characteristic being exposure to schistosome infection [37] , this result suggests that the difference in schistosome infection may contribute to differences in CHI3L1 levels and is supported by the observation that schistosome infected people had higher levels of CHI3L1 than uninfected people , independent of residential area . Furthermore CHI3L1 levels were reduced following elimination of adult worms from the body via curative antihelminthic treatment [38] , determined via egg count and the disappearance of visible haematuria . Taken together these observations suggest that CHI3L1 levels are related to current levels of infection . The significant interaction between age and infection status with the difference between schistosome-infected and uninfected people being the most apparent in the youngest age group is consistent with a dynamic relationship between infection status and CHI3L1 levels as people age . Other factors confounding the relationship between CHI3L1 and infection status pre- and post- antihelminthic treatment , such as the onset of age-related diseases [35] requires further investigation . A previous study of schistosome related pathology has suggested that it is infection status rather than infection intensity that is related to pathology [12] . Our results are consistent with this observation in that it was infection status rather than intensity that was associated with CHI3L1 levels . Accordingly , the change in CHI3L1 on clearing infection was not related to pre-treatment infection intensity ( data not shown ) . To determine if CHI3L1 levels differ in people with visible signs of pathology , we compared levels of the protein in children who had visible haematuria to age and infection intensity matched controls confirmed negative for haematuria . We showed that within infected children , CHI3L1 levels were higher in haematuria positive compared to haematuria negative individuals , suggesting that elevated CHI3L1 levels are associated with infection and morbidity in S . haematobium infection . Further studies expanding on the numbers of individuals assessed for haematuria and CHI3L1 levels will be required to confirm this association . The inflammatory immune response associated with schistosomiasis infection is modulated by adult schistosome worms as well as by the egg antigens [39] . Treatment with PZQ kills the adult schistosome worms and removes their recently laid eggs [40] , eliminating the source of inflammation . The decrease in CHI3L1 levels in the treated group may therefore be related to a reduction in inflammatory processes associated with schistosome infection and egg laying , and support previous studies in liver disease that suggests CHI3L1 levels are a bio-indicator of early phases of fibrogenesis [21] . However , while CHI3L1 is associated with inflammation , as yet , its precise function remains unknown . In mice , for instance , inflammatory signals induced by LPS and IFN-γ can both stimulate and inhibit chitinase like protein ( CLP ) expression in a context dependent manner [41] . As the function of these CLPs is elucidated , it may be possible to make a more educated assessment as to why CHI3L1 is downregulated on clearance of schistosome adult worms . To determine if elevated CHI3L1 levels are a marker of schistosome-related immunopathology , further studies with larger numbers of people with defined levels of pathology ( assessed using current standard markers of schistosome- related immunopathology such as ultrasonography ) are necessary . Furthermore , in order to resolve the mechanism behind the observed elevation as well as the relationship between CHI3L1 levels and infection status in the different age groups mechanistic studies on both human and mouse models will be required . One possible explanation is that this could be due to differences in the stage of the pathological processes that the hosts were experiencing , with the younger individuals being at a stage of disease where lesion repair and granuloma formation is actively occurring [7] , [9] . Given the lack of significant difference in CHI3L1 levels in infected vs . uninfected individuals in the older age groups , it is unsurprising that treating these individuals does not result in a significant reduction in their CHI3L1 levels . Older individuals have had a lifetime of exposure to infection , and may be experiencing more advanced pathology associated with inactive disease including bladder calcification and a drop in egg excretion [7] , [9] , as well as a higher background level of CHI3L1 associated with age [35] . There were no macrohaematuria cases observed post-treatment and this is consistent with other studies reporting a reversal/clearance of pathology [42] . These previous studies also reported that the groups experiencing the greatest reduction in morbidity following PZQ treatment are the younger age groups where infection levels are rising but who are not yet suffering from irreversible schistosome pathology [42] . In determining the systemic cytokine environment in the participants relative to CHI3L1 , we showed that levels of the measured Th1 and Th2 cytokines were not associated with CHI3L1 levels , perhaps due to the cytokines acting locally at sites of infection [2] , but there was an inverse association between CHI3L1 and levels of IL-10 and IL-17A . IL-10 can dampen a damaging pro-inflammatory response [43] and has previously been associated with schistosome infection levels [44] where it can limit pathology during infection as is suggested in the case of bladder pathology [45] . IL-17A is primarily regarded as a pro-inflammatory cytokine and has been found to be associated with numerous inflammatory diseases [46] . Despite being reported to have a role in the schistosome mediated pathological process in murine studies [47] , a previous study of people resident in the same region showed less systemic IL-17A levels in S . haematobium infected people compared to uninfected individuals [32] , and here we describe a negative association with CHI3L1 levels and systemic IL-17A . The relevance and implications of our cytokine findings for CHI3L1 levels in schistosome infection and for schistosome-related pathology remains to be determined in mechanistic studies . Our study investigated CHI3L1 in a single infection setting; it will be interesting to see if the same relationship with infection status pre- and post- treatment is also observed in S . mansoni infection where there is no existing marker for early morbidity [10] . Indeed , in order to fully evaluate CHI3L1 as a potential pathology marker the study must be extended to other settings , including co-infections and differing ethnic populations , and the results compared to existing and candidate pathological markers . Overall , our study has demonstrated that CHI3L1 levels are associated with schistosome infection status regardless of cumulative history of exposure to infection . Moreover , CHI3L1 is higher in people showing clinical signs of urogenital schistosomiasis ( haematuria ) compared to infected people presenting no haematuria . Antihelminthic treatment results in a significant reduction of CHI3L1 levels coinciding with the removal of adult schistosome worms . Both before and after curative treatment the difference in CHI3L1 between infected and uninfected people is greatest in the youngest age group where infection levels are rising rapidly , indicating not only a potential mechanism for schistosomiasis associated pathological processes , but also of an early onset of these pathological processes . The associations between CHI3L1 and schistosome infection warrants further investigation to determine the utility of CHI3L1 as a diagnostic marker for schistosome related morbidity as well as a tool for evaluating the impact of S . haematobium control programmes . | Over 100 million people , mainly living in sub-Saharan Africa , are infected with the blood fluke Schistosoma haematobium . Morbidity includes anaemia , malnutrition , impaired cognition and physical growth as well as pathology of the urogenital tract . Urogenital morbidity is initiated by eggs laid at the site of infection causing mechanical damage and leading to haematuria , or blood in the urine . Eggs also secrete antigens , triggering a protective granuloma to form around the eggs which can lead to fibrosis and organ failure . Currently there are few non invasive markers of schistosome-related pathology hampering diagnosis and evaluation of control programmes . Chitinase 3-like 1 ( CHI3L1 ) has been associated with several inflammatory disorders involving tissue remodelling . In this study we assessed the levels of CHI3L1 in a group of individuals from a schistosomiasis endemic area in Zimbabwe . We showed that CHI3L1 is elevated in schistosomiasis infected individuals over uninfected individuals , with the strongest association seen in the youngest age group who have a shorter history of schistosome infection . Treating individuals with a drug that eliminates the adult worm leads to a decrease in CHI3L1 in infected individuals . These results suggest that the potential for CHI3L1 as a biomarker for schistosomerelated pathology should be explored further in different human populations . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"immunopathology",
"medicine",
"infectious",
"diseases",
"schistosomiasis",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"infectious",
"disease",
"epidemiology",
"neglected",
"tropical",
"diseases",
"immunology",
"biology",
"parasitic",
"diseases",
"immune",
"response",
"helminth",
"infection"
] | 2012 | Chitinase 3-Like 1 Protein Levels Are Elevated in Schistosoma haematobium Infected Children |
As malignant transformation requires synchronization of growth-driving signaling ( S ) and metabolic ( M ) pathways , defining cancer-specific S-M interconnected networks ( SMINs ) could lead to better understanding of oncogenic processes . In a systems-biology approach , we developed a mathematical model for SMINs in mutated EGF receptor ( EGFRvIII ) compared to wild-type EGF receptor ( EGFRwt ) expressing glioblastoma multiforme ( GBM ) . Starting with experimentally validated human protein-protein interactome data for S-M pathways , and incorporating proteomic data for EGFRvIII and EGFRwt GBM cells and patient transcriptomic data , we designed a dynamic model for EGFR-driven GBM-specific information flow . Key nodes and paths identified by in silico perturbation were validated experimentally when inhibition of signaling pathway proteins altered expression of metabolic proteins as predicted by the model . This demonstrated capacity of the model to identify unknown connections between signaling and metabolic pathways , explain the robustness of oncogenic SMINs , predict drug escape , and assist identification of drug targets and the development of combination therapies .
Diseases like cancer involve a large range of components that interact via complex and highly dynamic networks [1–3] , and are interconnected with biochemical pathways [4–7] . These multipath interconnections may allow cancer and other diseases to take alternate routes and bypass the effects of therapeutic interventions . Traditional approaches to biological studies which focus on single molecules or pathways may not be able to capture and understand these complex networks of molecular interactions . To predict alternative or escape routes around blockades and to develop effective therapies [8] , sophisticated mathematical and computational models are required [9–10] . Transforming traditional drug discovery approaches toward smarter therapeutic strategies , the field of systems biology is emerging [1 , 9 , 11–20] . Systems approach generally involve large-scale data collections , most often from high-throughput transcriptome or proteome analyses , incorporation of the data into mathematical models to deduce systems properties , model building and finally computational and/or experimental validation of model-derived hypotheses . Systems biology approaches may predict combination therapies for cancers driven by different oncogenic signaling and metabolic pathways . Signaling and metabolic networks were studied using separate model systems [15 , 21–26] . Mathematical models for signaling pathways had been based on logic models [27–30] , kinetic models [31–33] , decision tree [34] , and other differential equation-based models [35] . Computational models of molecular signaling [36–41] have the potential to improve drug discovery and development [32 , 42–44] . Analyses of knockdown experiments [45] using mass spectrometry [46] and transcriptomics [47–49] are progressively refined and tuned towards specific physiological situations . While these studies have helped considerably to extend our understanding of tumor biology , they are still restricted to signaling pathways and do not integrate the metabolic pathways , which in some initial studies have been subjected to separate systems biology analysis . Predicting the effects of multiple targeted drugs [8 , 50] with modeling the information flow from new molecular interactions within pathways is challenging [51–55] . Here we report the development , test and validation of an integrated model for signaling and metabolic pathways in cancer using glioblastoma multiforme ( GBM ) as an example [47 , 56–59] . GBM is the most prevalent and most aggressive brain tumor . In the majority of cases , tumor development is dependent on signaling via the epidermal growth factor receptor ( EGFR ) and requires EGF in lower-grade forms or is EGF-independent in the more aggressive forms . In most cases , the expression of EGFR is up-regulated , often related to the amplification of the EGFR gene . More than fifty per cent of EGFR-amplified GBM cases have in-frame deletions of exons 2–7 that code for the extracellular ligand-binding domain ( EGFRvIII mutation ) resulting in EGF-independent constitutive signaling and more aggressive tumor growth , higher invasiveness , increased resistance to treatment , and poor prognosis [60–62] . In this study , we choose the cell line U87MG expressing the EGF-dependent EGFRwt and its derivative U87MGvIII expressing the EGF-independent EGFRvIII-mutant , as models of low and high-grade GBM , respectively . Except for the EGFR mutation , the two cell lines have the same genotype but differ in growth behavior suggesting different metabolic requirements , and are experimental examples for the regulation of the interconnection of signaling and metabolic pathways , which is considered among the basic characteristics of cancer [63–64] . We implemented a probabilistic approach based on the Hidden Markov Model ( HMM ) utilizing the information of experimentally established protein-protein interactions ( PPIs ) [65–66] to extract novel paths and interconnections between signaling pathway proteins ( S ) and metabolic pathway proteins ( M ) . To cope with the limitations of PPI identification , for example high error rates of the detection methods [67] , incomplete data sets , ignorance of the physiological conditions in the cell or tissue compartments [68] , technical problems and study biases [69] , we collected information from curated sources ( http://string-db . org ) with high experimental score cut-off , which reduces false positive rates , and used transcriptome data from clinical samples to build more reliable and GBM context-specific PPI networks . To bridge the gap between transcriptome data and cell-biological processes , we incorporated proteome data from the GBM cell lines complemented with transcriptome data to solve the missing data problem caused by the failure of proteomics to capture all proteins in the cells due to sensitivity and reproducibility issues . Including experimental data , our dynamic model can make use of multiple weighted network properties to add biologically relevance and can extract novel paths of information propagation in networks . The results of the model were tested in rigorous in-silico perturbation experiments and experimentally validated in cell culture systems . Fig 1 depicts the overall strategy implemented with this study .
As a starting model , we constructed an integrated network where signaling ( S ) and metabolic ( M ) pathway proteins were connected through protein-protein interactors ( PPIs ) . The signaling pathways were the apoptosis , Akt , EGFR , hedgehog ( Hh ) , JAK-STAT , JNK , MAPK , mTOR , NF-kappa B ( NF-κB ) , Notch , p53 , Ras , TGF-β , and Wnt pathways . On the metabolic site , there were 81 pathways grouped into the six categories carbohydrate , lipid , amino acid , nucleotide , energy and xenobiotic metabolism ( Fig 2 ) . To build an initial human protein-protein interactome network ( HPPIN ) , the total of all human protein-protein interactions was extracted from the protein-interaction database STRING [70] for experimentally validated interactions including physical and functional associations . Then to structure the signaling-metabolic interaction network ( SMIN ) , cross-connections between the fourteen signal transduction and six groups of metabolic pathway proteins selected from pathway databases ( see Materials and methods ) were constructed based on protein-protein interacting proteins ( node ) and cross-connecting links/paths . All possible connections between any given signaling pathway protein ( S ) and metabolic pathway protein ( M ) via protein-protein interconnectors ( PPIs ) were included ( Fig 2A ) . To derive a simplified but informative network , the interactions were restricted to the second level of protein interactors , i . e . up to interactors of interactors . This led to four different types of cross-connected paths where signaling pathway proteins were either directly connected with metabolic pathway proteins or through one , two or three PPIs , respectively ( Fig 2A ) . These paths connecting all above-mentioned signaling ( S ) and metabolic ( M ) pathway proteins were then converted into networks based on the protein-protein interaction status between the involved proteins ( Fig 2B ) . The resulting signaling-metabolic interaction network ( SMIN ) is shown in the left panel of Fig 3A with nodes in orange and edges ( connection between two nodes ) in blue , within the total interactome containing HPPIN network with grey nodes and edges . As a result of afore-said restriction criteria , Fig 3A shows the reduction of the network size ( number of nodes/proteins and their interactions/edges ) of SMIN ( 11 , 059 interactions formed by 2 , 785 proteins ) from HPPIN ( 16 , 828 interactions formed by 5 , 703 proteins ) . As an example of network links , the detailed connections and interactions between one signaling pathway protein , CSNK2A1 , and one metabolic pathway protein , NDUFA13 , through different PPIs in the SMIN are shown on the right panel of Fig 3A . These two selected examples are indicated by asterisks . The SMIN was found to contain 158 direct ( S-M ) linking paths , 4 , 036 with one interactor ( S-P-M ) , 91 , 847 with two interactors ( S-P-P-M ) and 2 , 110 , 205 with three interactors ( S-P-P-P-M ) . These paths were formed between 158 , 2 , 967 , 22 , 307 , 69 , 032 S-M pathway protein pairs , respectively ( Table 1 ) . Comparisons with respective random networks proved that our selected HPPIN and SMIN are non-random scale-free networks ( S1 Fig and S1 Table ) . To render the above condition-independent SMIN GBM-specific , quantitative comparative proteome analysis of the low- and high-grade GBM cell lines U87MG and U87MGvIII were performed and the data incorporated in the model to derive an enriched GBM-specific network . The use of cell lines helped to reduce the noise associated with the usual small size and heterogeneous cellular compositions of clinical samples . Proteomes was chosen as primary expression data to focus the models on the protein level , which is more closely related to the biological processes to be modeled than transcriptome data . Transcriptome date from clinical samples ( see below ) were subsequently added to reduce the missing data problem of proteomics and link the model to the clinical level . To decrease the complexity of the proteins in the cell extracts and minimize signal suppression by overabundant peptides , the proteins were first separated by SDS-PAGE . Then the gels were cut into one mm slices followed by separate in-gel digestion of the proteins with trypsin . The resulting fragments were extracted from the gel slices as individual samples , separated by reverse-phase nano-HPLC and analyzed on-line by ESI-Q-TOF mass spectrometry ( S2 Fig ) . Quantification was done label-free by calculating the exponentially modified protein abundance index ( emPAI ) to avoid the drawback of mere signal intensity-based measurements . Only proteins with at least two tryptic fragments were identified by MS/MS with high confidence were considered , which further reduced the noise of protein identification although resulting in lower numbers of hits . The analyses were done with three independent replicates per cell line and the resulting data processed in two different ways . First , each of the three datasets for U87MG was compared to each of the three datasets for U87MGvIII resulting in nine pair-wise comparisons . Second , the average of three datasets of one U87MG was compared with the average of the three datasets for U87MGvIII . A total of 907 unique proteins were identified , 771 from U87MG and 664 from U87MGvIII . Five hundred twenty-eight proteins were expressed in both cell lines , 243 only in U87MG ( EGFRwt ) and 136 only in U87MGvIII ( EGFRvIII ) . These distinctively expressed proteins were considered as overexpressed in the respective cell lines in comparison to the other . Compared quantitatively , 458 of the 528 common proteins were expressed at similar levels , 70 proteins were either up ( 45 ) or down ( 25 ) regulated in U87MGvIII compared to U87MG ( Fig 4A ) . Together , nearly half of the identified proteins ( 449 ) were differentially expressed , 268 down-regulated and 181 up-regulated in U87MGvIII versus U87MG ( Fig 4B ) . Over-representation analysis ( ORA ) based enrichment for cellular pathways , biological processes and molecular functions were performed using cell line specific and commonly overexpressed proteins . Fig 4C and 4D show the top 20 most enriched pathways for proteins exclusively overexpressed in U87MGvIII ( EGFRvIII ) and U87MG ( EGFRwt ) , respectively . Similarly , Fig 4E and 4F provide the top 20 enriched pathways for commonly over and under expressed proteins , respectively . The most significantly enriched pathways were found to be proteasome ( p-value = 0 . 97763E-07 ) in U87MGvIII ( EGFRvIII ) and TCA cycle ( p-value = 1 . 48369E-06 ) in U87MG ( EGFRwt ) . However , Fructose and mannose metabolism ( p-value = 6 . 7086E-04 ) and pentose phosphate pathways ( p-value = 3 . 1503E-05 ) were found to be most significantly enriched for proteins commonly overexpressed and underexpressed , respectively . Gene ontology ( GO ) based biological process and molecular function over-representation analysis was also performed using cell line specific overexpressed proteins . Most significantly enriched biological processes and molecular function were found to be Tricarboxylic acid metabolic process ( p-value = 2 . 7304E-04 ) and Threonine-type peptidase activity ( p-value = 0 . 001053394 ) , respectively in U87MGvIII ( EGFRvIII ) . In U87MG ( EGFRwt ) , TCA metabolic process ( p-value = 7 . 76842E-08 ) and pre-mRNA binding ( p-value = 0 . 007857451 ) were found to be the most significantly enriched biological process and molecular functions , respectively ( S3 Fig ) . To uncover the signature of the reprogramming of global cellular processes by the EGF-independent constitutively active EGFRvIII in GBM , we mapped the data from the comparative proteomics of U87MGvIII ( EGFRvIII ) versus U87MG ( EGFRwt ) onto the above-described SMIN . All S-M interconnecting paths with at least one differentially expressed protein were extracted from the SMIN to generate a GBM-specific network ( Fig 3B , left panel: highlighted with yellow nodes and green edges within the orange nodes and blue edges of the SMIN ) with the assumption that those paths will have higher probability to be differentially active in mutant GBM condition . As an illustration of proteome data mediated extraction of a GBM-specific network , the interconnections between the aforementioned signaling pathway protein , CSNK2A1 and the metabolic pathway protein , NDUFA13 are highlighted after extraction from SMIN ( Fig 3B right panel in comparison to Fig 3A right panel ) . Table 1 provides the details of the paths/pairs present at different stages of network development . To make this network further enriched with potentially disease-relevant paths/pairs , weights specifying disease-related biological properties and expression states of the proteins were assigned to each node ( protein/gene ) and edge ( interaction ) of the GBM-specific network . The following three categories of proteins ( nodes ) were given additional weights . First , proteins cross-talking between different signaling pathways ( signaling cross-talk , SC ) , second , rate-limiting enzymes ( RLE ) for their roles in regulating metabolic rates and pathways , third , EGFR mutation-specific differentially expressed proteins ( dEXP ) for their GBM-specific impact . The GBM-specific network included 446 dEXP , 349 SC , and 267 RLE proteins . Of these , 11 SC and 17 RLE proteins were up or down regulated suggesting their involvement in signaling-metabolic cross-connection in EGFR-mutated condition ( S4A Fig ) . For systems-level interpretation and understanding the network property , local signaling entropy ( Si ) was introduced . Previous studies [71–73] showed that Si can be used as a measure of uncertainty in signaling information flow over a network and to identify important signaling pathways and genes/proteins in cancer . Effect-on-node ( effs ) of every protein ( node ) in the network provided significance of a protein based on SC , RLE and dEXP in its local network . To identify probable paths of information flow from a signaling to metabolic pathways , network entropy ( Si ) and effect-on-node ( effs ) properties were incorporated as node weights into the logic of the Hidden Markov Model ( HMM ) . The edge weight of every two interacting nodes ( gene/protein ) were defined based on the principle of mass action ( assuming that the probability of interaction of two genes in a given sample is proportional to the product of their expression values in the study samples ) as probability of interaction ( pij , where i and j are the two nodes ) in GBM condition . To assign the expression value of each node present in the GBM specific network , the average expression value of each gene was calculated from the normalized transcriptome data from 239 GBM patients . Incorporating these transcriptome data as edge weights linked the network with biological information from GBM patients . It helped to assign an extra weight other than previously mentioned node weight for all connections made by two nodes based on their expression in clinical GBM patients . Furthermore , it helped to add another level of constraint on over-prediction of information flow for nodes , which got an extra weight based on SC , RLE , and dEXP but are not expressed at higher levels in GBM patients . This helped to incorporate the contribution of those nodes to the disease , which were identified by neither of the three before-mentioned node weights nor by proteomics . Moreover , the much broader coverage of gene expression by genome-wide transcriptomics compared to proteomics helped to overcome some of the missing-data-problem of proteomic datasets . An HMM-based simple mathematical formalism was used to understand context-specific information propagation from signaling to metabolic pathways in the human biological network . Node weights and edge weights were used to define the two major parameters of the Markov model , emission ( Ef ) and transition ( Tj ) probabilities , respectively . Two model systems were implemented to apply HMM logic , Model 1 for SM pair identification ( Fig 5A ) and Model 2 for S-M linking path identification ( Fig 5B ) . Model 1 emphasized source ( Signaling Pathway Protein , S ) and destination ( Metabolic Pathway Protein , M ) pairs i . e . SM pairs having higher chances of information flow for each type of connections ( Figs 2A and 5A ) . Model 2 was applied to find S-M linking paths between those selected pairs from Model 1 . Selection of SM pairs ( Model 1 ) and S-M linking paths ( Model 2 ) was based on Pathscore ( see Methods for more details ) a mathematical function of emission probability ( Ef ) and transition probability ( Tj ) . For Model 1 positional emission probability was calculated considering the similar number of proteins because Model 1 was applied after grouping interconnecting links having the similar number of proteins forming the connection ( Fig 5Ai–5Aiv ) . The calculated path scores of linking paths from the individual models were converted to statistical Z scores to identify the paths deviating from the mean . Based on the Z score under the individual models , the signaling-metabolic linking paths were classified as highly significant with Z score ≥3 ( more stringent ) or less significant with Z score ≥1 ( less stringent ) in EGFR-mutated GBM . The signaling and metabolic pathway proteins from the two ends of linking paths containing significant Z sores of each of the models were defined as the significant SM pairs . Multiple identifications of the same S-M pair from different models i . e . the formation of different types of significant linking paths involving different PPIs or different numbers of PPIs were nullified by considering them as a single . With that we identified 1 , 2 , 114 and 758 SM pairs meeting the more stringent cut-off ( Z score ≥ 3 ) and 1 , 8 , 334 and 1 , 961 pairs with the less stringent cut-off ( Z score ≥ 1 ) for the S-M , S-P-M , S-P-P-M and S-P-P-P-M linking types , respectively ( Table 1 , Model-1 ) . In total 801 Z ≥ 3 and 2 , 055 Z ≥ 1 signaling-metabolic cross-connected SM pairs were identified between the 14 signaling and 6 groups of metabolic pathways as potentially important in EGFR-mutated GBM . These SM pairs were categorized according to the proteomic expression states of the source ( S ) and destination ( M ) proteins as UP-DOWN , UP-UP , DOWN-UP , and DOWN-DOWN . Including unidentified proteins in proteomics analysis ( NA ) of the cell lines , the couplet categories UP-NA , DOWN-NA , NA-UP and NA-UP , and unchanged expression states ( NC ) in the cell lines with SM categories UP-NC , DOWN-NC , NC-UP and NC-DOWN , and unidentified and unchanged SM pairs NA-NA and NC-NC were added ( S4B Fig ) . A number of cross-connections between signaling and metabolic pathways were identified with significant cutoff levels where either one or both pathway proteins ( S and/or M ) were not identified ( NA ) and/or unchanged ( NC ) by mass spectrometry indicating that the integrated network model can identify connections also where intermediate interactors are more important than SC , RLE or dEXP . The identification of pathway cross-connections is thus not dependent on proteomic identification of all constituent members but can be based on signaling crosstalk proteins and their expression status in GBM patients . It is important to include unchanged ( NC ) proteins in the model building as they might represent nodes in the paths that include other proteins that are differentially expressed . They might also play a role in the crosstalk between signaling paths and pathways or they might become important when known primary paths are blocked , e . g . by therapeutic intervention . The model could thus help to identify potential therapeutic targets for alternative therapies in cases of treatment failures or to design combination therapies that target primary together with potential escape pathways . Mapping the pathway information of proteins in SM pairs showed which signaling pathways made a higher number of connections with which type of metabolic pathways in EGFR-mutated GBM . Six hundred one significant pairs with Z ≥ 1 were cross-connecting the MAP kinase pathway with all six groups of metabolic pathways ( S2 Table ) . Similarly , the Ras , EGFR , AKT and p53 pathways were significantly connected to metabolic pathways through 570 , 543 , 549 and 179 SM pairs , respectively ( S2 Table ) . As crosstalk between availabe signaling pathways is common whereas it is less common in between metabolic pathways ( S5A Fig ) , some identified SM pairs and the respective proteins/genes may be shared . Analyzing the shared components in the five most connected signaling pathways revealed that MAPK pathway had the highest number of unique significant SM pairs ( 256 ) and genes/proteins ( 63 ) involved , followed by 194 , 129 , 116 and 95 unique significant SM pairs ( S5B Fig ) and 20 , 30 , 36 , 19 genes/proteins ( S5C Fig left ) for the EGFR , AKT , p53 and Ras pathways respectively . Twelve cross-connected pathway protein pairs were common to all five pathways and 141 pairs shared by the Ras , EGFR , AKT and MAPK pathways indicating high connectivity between them , and their cross-connection with metabolic pathways suggesting important roles of the respective proteins in EGFR-mutated GBM ( S5B Fig for pairs , C for genes left ) . In turn , the amino acid , carbohydrate , and nucleotide metabolic pathway groups were connected to all fourteen signaling pathways through 327 , 289 and 326 cross-connected SM pairs ( S2 Table , S5B Fig right ) and 296 , 260 and 268 genes in significant S-M paths , respectively ( S5C Fig right ) . This indicates that altered cellular signaling related to the EGFR mutation and its constitutive activity affects most strongly these three metabolic pathway groups . As metabolic pathway enzymes interact via their substrates and products , there are few possibilities for interconnection between metabolic pathways except for the end steps , which is confirmed by the analysis of shared proteins . Twenty-three shared pairs were identified between the amino acid and the carbohydrate metabolic pathway , which relates to the low number of amino acids metabolites feeding into the tri-carboxylic acid cycle ( S5B Fig right ) . Multiple linking paths of different types ( S-M , S-P-M , S-P-P-M , and S-P-P-P-M ) or of the same type but through different PPIs were possible between SM pairs . Not all of these linking paths could be equally significant in EGFR-mutated GBM . To find the significant linking paths between the above-identified significant SM pairs , all possible paths between a single SM pair were considered under a single model ( Model 2 , Fig 5B ) . Since in Model 2 proteins forming interconnections between SM pairs vary , positional emission probabilities were calculated for these proteins . As an example , in Fig 5B the second position contained two proteins and third position one protein . Paths with path scores ≥80% of the highest path score for each SM pair were selected as significant from Model 2 . Accordingly , all the significant paths were identified from high ( more stringent Z ≥ 3 ) and less ( less stringent Z ≥ 1 ) significantly specified SM pairs to identify the total of significant linking paths in the network . These analyses showed that 2 , 21 , 228 , 625 and 876 significant paths were present with more stringent cut-off ( Z ≥ 3 ) and 5 , 84 , 600 , 1 , 564 and 2 , 253 paths with the less stringent cut-off ( Z ≥ 1 ) for the four pathway types ( Table 1 ) . By these pathway-based analyses under less stringent condition ( Z ≥ 1 ) , 652 significant linking-paths were identified between 570 cross-connected SM pairs of the Ras pathway with all six groups of metabolic pathways ( S2 Table ) . In addition , 668 , 629 and 569 significant linking-paths were identified between 601 , 549 and 543 cross-connected S-M pairs between the MAPK , AKT and EGFR pathways , respectively , and all 6 groups of metabolic pathways . Together , in EGFR-mutated GBM , these four signaling pathways were involved in the highest number of SM cross-connections with metabolic pathways: 368 , 344 and 298 cross-connecting paths were found between the fourteen signaling pathways and 327 , 326 and 289 cross-connected SM pairs involving the amino acid , nucleotide and carbohydrate metabolism , respectively ( S2 Table ) . Based on the identified significant ( less stringent Z ≥ 1 ) SM pairs and the significant linking paths ( path score ≥80% of the highest path score of each SM pair ) , we converted ( as in Fig 2B ) the significant paths into a network ( Fig 3C left panel: highlighted with blue nodes and red edges within yellow nodes and green edges of the GBM network ) . This filtered network is more specific for EGFR-mutated GBM conditions . The filtration further eliminated non-significant or unimportant SM pairs and PPIs , which is shown , as an example , for the interconnections between signaling pathway protein CSNK2A1 and metabolic pathway protein NDUFA13 ( Fig 3C right panel compared to Fig 3B right panel ) . The GBM-specific network based on significant SM pairs and linking paths was restructured to implement the biological consequences as network properties i . e . color-coded proteomic expression states ( up , down , no change and not identified ) , size of node symbols proportional to the numbers of connections passing through it , colors of the edges as connection formed between more ( Z ≥ 3 ) or less ( Z ≥ 1 ) stringently defined SM pairs , width of the edge as the probability of interaction or product of the average expression values of two interacting genes in GBM patients from the transcriptome data ( Fig 6A ) . The resulting network showed the important signaling pathways and their interconnections with metabolic pathways in EGFR-mutated GBM with the significance of every protein ( size of the node ) and their interactions with interacting partners ( width of the edge ) . Around the network , representative paths between 14 signaling to metabolic pathways are shown as examples ( Fig 6A , S6 Fig as more details of RAS pathway ) . To explore the importance of signal-crosstalk proteins in signaling to metabolic pathway interconnections in EGFR-mutated GBM , the sub-network dependent on the top fifteen crosstalk protein-based interconnecting paths were extracted from the GBM-specific significant network ( Fig 6B ) . These sub-networks showed which signaling pathways are mostly cross-talking and how they are connected with metabolic pathways . This information was used to extract candidate genes/proteins/paths of EGFR-mutated GBM for further analysis . In silico perturbation analysis was performed for identification of paths that significantly change upon removal ( e . g . by mutation or down-regulation ) of a node ( protein ) . To test the importance of the nodes/proteins in the final weighted network , each of the 654 nodes present in the Z ≥ 1 network ( Table 1 ) was removed individually from the human interactome ( HPPIN ) and the node and edge weights were recalculated for the resulting networks and paths by recalculating Model 1 and Model 2 ( Fig 2 ) . Accordingly , new significant SM pairs ( Z ≥ 3 or 1 ) were identified on the basis of Model 1 and significant paths ( path score ≥80% of the highest path score ) between them on the basis of Model 2 . We mapped the pathway details of the SM pairs and calculated the average path scores before and after perturbation for the 14 signaling pathways to all 6 groups of metabolic pathways and vice-versa . The difference of values ( before vs . after perturbation ) for the 654 proteins for all pathways were converted to Z-scores and plotted for each perturbed node for each signaling pathway ( Fig 7A ) . The nodes for which the Z-scores deviated from the mean as -2 ≥ Z ≥ 2 were selected as effective for the respective signaling pathway to all metabolic pathway interconnections in EGFR-mutated GBM ( Fig 7B ) . S3 Table lists the numbers of significant and effective proteins identified for the individual signaling pathways connected to all metabolic pathways and from all signaling pathways to the individual metabolic pathways . As a measure of its effect on signaling-metabolic interconnection , each perturbed node was ranked according to the difference between baseline and perturbed condition . This means that highly ranked proteins have an important role in the connections of the respective signaling pathway to all metabolic pathways or vice-versa . The NOTCH pathway is shown as an example for the level of reduction in the network size when the GBM network is transformed to the significant GBM network to identify significant interconnecting paths and proteins ( Fig 8 ) . Fig 8A shows the interconnections of NOTCH pathway proteins with all metabolic pathway proteins present in the signaling-metabolic interaction network ( SMIN ) and Fig 8B shows only those NOTCH pathway proteins with interconnections to metabolic pathway proteins passing through effective nodes , i . e . nodes identified by the perturbation experiments ( Fig 7B ) . Fig 8C presents the interconnections of the NOTCH pathway to all metabolic pathways in the GBM-specific network filtered on the basis of the weightage parameters for the nodes and edges , and Fig 8D shows the interconnections between the significant proteins only . Fig 8E shows the interconnections of the NOTCH pathway to all metabolic pathways in the GBM-specific significant network and Fig 8F the same only for significant nodes . The comparison of Fig 8B , 8D and 8F on basis of the effective node identified by the perturbation study ( red colored ) indicate the levels of filtration from the starting SMIN to the final GBM-specific significant network . We found 457 , 111 paths involving 1941 genes/proteins and 8047 interactions out of a total of 2 , 206 , 246 paths in the SMIN connecting the NOTCH signaling pathway to all metabolic pathways , of which 10% were found to have a significant ( Z-score cut-off -2 ≥ Z ≥ 2 ) perturbation impact ( PI ) ( Fig 8B ) . Comparable reductions of approximate 75% ( Fig 8C ) and 50% ( Fig 8D ) for both nodes and interaction were found when going to the GBM-specific condition . After Pathscore based filtration ( Z-score ≥ 1 ) , 146 paths involving 125 nodes and 166 interactions were identified in the NOTCH pathway ( Fig 8E ) of which 51 paths involving 59 nodes and 65 interactions formed by nodes with significant PIs . The result of perturbation study proved the importance of the respective nodes in the final network for information flow from signaling to metabolic pathways . To validate these findings of interconnections between signaling pathway alterations and metabolic rearrangement , some of SM connections were selected from the final network for experimental validation ( Fig 9 ) . The selection was based on the effect of the removal of the respective node in the in silico perturbation experiments , the predicted effects on the metabolic pathways and the availability of specific small-molecule inhibitors , excluding transcription factors as their impact is obvious . The selected targets were calmodulin ( CALM2 ) , casein kinase II subunit alpha ( CSNK2A1 ) , 1-phosphatidylinositol 4 , 5-bisphosphate phosphodiesterase gamma-1 ( PLCG1 ) , the tyrosine-protein kinase ABL1 and B-cell lymphoma 2 ( BCL2 ) . The signaling-metabolic interconnecting paths including the selected proteins were extracted from the final GBM-specific significant network using the most stringent cutoff and the connected metabolic proteins linked to the above-mentioned signaling pathway proteins were identified ( S1 File ) . As the translation of oncogenic signaling to metabolic adjustment is via the expression of metabolic enzymes , the changes of expression of the metabolic pathway proteins upon inhibition of the selected signaling proteins with small molecule inhibitors and cell viability were analyzed . The inhibitors were CGS-9343B for calmodulin ( Fig 9A ) , Emodin for CSNK2A1 ( Fig 9B ) , U73122 for PLCG1 ( Fig 9C ) , Dasatinib for ABL1 ( Fig 9D ) , and ABT199 for BCL2 ( Fig 9E ) . U87MG ( EGFRwt ) and U87MGvIII ( EGFRvIII ) cells were incubated with the inhibitors and their effects on the expression of the predicted interconnected metabolic pathway proteins analyzed by quantitative RT-PCR , and the viability of the cells tested in MTT assays . The blockade of the signaling pathway proteins had significant impacts on the expression of the metabolic pathway proteins ( Fig 9 ) . In the case of the calmodulin inhibitor , the expression levels were reduced in both cell lines but for prostaglandin-endoperoxide synthase 2 ( PTGS2 ) and O-linked β-N-acetylglucosamine transferase ( OGT ) much more pronounced in the EGFR-mutant cell line than in the wild type . The strongest effect was seen for phosphoglycerate kinase 1 ( PGK1 ) with around 400-fold reduction of expression in both cell lines . The effects of the CSNK2A1 inhibitor are more differentiated . While the OGT and PGK1 expression is inhibited uniformly in both cell lines , 800-fold in case of PGK1 , the inhibition of glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) is much stronger in U87MGvIII than in the wild type , the expression of ribonucleoside-diphosphatereductase 2 ( RRM2 ) is decreased in the wild-type and enhanced in the mutant , and that of NADH dehydrogenase [ubiquinone] 1 alpha sub-complex subunit 13 ( NDUFA13 ) is affected inversely . Inhibition of phospholipase C gamma 1 ( PLCG1 ) results in enhanced expression of GAPDH and reduced expression of NDUFA13 . The ABL1 inhibitor reduced the expression of pyruvate dehydrogenase kinase subunit 1 ( PDK1 ) in both cell lines but much more in the mutant cells . In contrast , the expression of pyruvate dehydrogenase ( lipoamide ) beta ( PDHB ) is more reduced in the wild type , and that of pyruvate dehydrogenase subunit alpha 1 ( PDHA1 ) is slightly enhanced in the wild-type and reduced in the mutant cells . Finally , the BCL-2 inhibitor reduces the expression of 6-phosphofructo-2-kinase/fructose-2 , 6-biphosphatase 3 ( PFKFB3 ) , GAPDH and RRM2 2-fold in both cell lines but enhances the expression of PGK1 in the wild-type 4-fold while reducing it in the mutant cells 10-fold . All signaling pathway inhibitors have pronounced negative effects on the viability of both GBM tumor cell lines ( Fig 9 ) but U87MGvIII is significantly more sensitive to the inhibition of casein kinase than the wild-type whereas the wild-type is more sensitive to the inhibition of PLCG1 , ABL1 and BCL2 . Half-maximal reduction of viability is seen for the PLCG1 and the ABL1 inhibitors in the single-digit , for the others in the double-digit micromolar range . Further , the effect of the signaling molecule inhibitors on cell migration and invasion was tested . The results showed the similar negative effects of the inhibitors with a higher sensitivity of U87MGvIII , which reconfirms the potency of the signaling-metabolic interconnected network model ( S7 Fig ) .
In all cellular condition , signaling , gene expression , and metabolic pathways must be coordinated to maintain cellular integrity and functions [74] . In this study , we propose an integrative systems biology model of information flow between signaling and metabolic pathways that use an interconnected scaffold based on the complete set of human interactome ( HPPIN ) . The model successfully detects probable unique connections of genes/proteins involved in the interconnection of different signaling and metabolic pathways . The identified components of the network ( i . e . genes/proteins as nodes and connections through edges ) represent known physical and/or functional associations between proteins/genes . Depending on the oncogenic signaling , the interconnectors involved modulating the metabolic pathways change . We have successfully applied the model to identify the interconnections altered in the constitutive signaling of the mutated EGFR in glioblastoma multiforme ( GBM ) compared to EGF-dependent and wild-type EGFR ( EGFRwt ) signaling . So far , the development of integrated models including all three or any two of signaling pathways , gene regulation and metabolic pathway in any cellular context is still very restricted . Difficulties of integration arise from the mode of information flow in these three processes significantly involving activation/inactivation , inhibition/induction and substrate/product respectively , as well as different extents and time frames or kinetics [74] . Attempts to integrate metabolism and gene regulatory networks were based on regulatory flux balance analysis ( rFBA ) [75 , 76] , steady-state regulatory FBA ( SR-FBA ) [77] , probabilistic regulation of metabolism ( PROM ) [78] , integrative omics-metabolic analysis ( IOMA ) [79] and ordinary differential equation ( ODE ) [80] . There is a straightforward interconnection between signaling and gene regulation through transcription factor ( TF ) but the little informative interface and the dynamics of cellular localization of TF affect the integrative modeling . A few studies addressed the issues using the integrative logical , influence graph , Boolean and thermodynamics models [81–86] . Attempts to integrate signaling and metabolism is rarer because information in the two process areas flows in different ways via activation/inactivation and substrate/product interactions respectively . The few reports that address this issue are based on ODE , combined dynamic ODE and Boolean models [33 , 80 , 87] . Here we have taken a novel approach and integrated signaling and metabolic pathways by i ) creating a weighted PPI network depending on the path of information flow from signaling pathway molecule to metabolic enzymes , and ii ) applying the probabilistic framework of Markov processes to calculate information flow scores . We have used two major functions of the Markov model . First , to estimate the probability of a protein to be present at a particular point in the network to transmit information , transition functions ( node weights ) were implemented . Second , to estimate the connection strength between two proteins , emission functions ( edge weights ) were introduced . The two functions utilize different biological parameters to explore the integrated network and to calculate S-M Pathscore . In contrast , to other models for integration of signaling and metabolism pathways , the comparative proteomic expression of the signaling molecule , metabolic enzymes and interactors are effectively incorporated to construct a weighted PPI network ( one criterion of node weights ) and to filter out context-dependent ( mutated vs . wild-type EGFR GBM ) S-M interconnections . The second attempt toward integration of the two pathway types to make modeling information flow more accurate and render the model context-dependent based on patient-related disease-specific parameter , trancriptomic expression value from GBM patients were implemented for each node and , by using the principle of mass action , edge weight were defined . The implementation of these two layers of information distinguishes the model from others by integrating weighted network properties like network entropy ( Si ) and effect-on-node ( effs ) . We believe that these have the potential to capture context-dependent properties of PPI networks in terms of information flow [88] . Genes/proteins having a high impact on the path formation are important in their local network as well as the global context . Unlike considering conventional network centralities ( hubs , bottlenecks etc . ) of nodes , biological state and expression profiles are used . Integrating proteome and transcriptome data in the development of the model helps to mitigate problems in translating high-throughput omics into systems biology models . While proteome data more closely related to the biomedical and pathological systems properties of interest , they are less complete and more prone to variation and experimental errors then transcriptome data . The use of proteome data from cell lines helps to some extent to overcome the experimental problems , however , cell lines are selected cell culture-adapted models and may be quite remote from in vivo situations . Transcriptome data from clinical specimens are getting increasingly available and may add a close-to-clinic dimension to the systems biology model , and help to overcome the missing-data problem of proteomics . Furthermore , the two omics data sets were used at different steps of the network establishment and for different biological relevance . The proteome data of the EGFRwt and EGFRvIII-mutated cell lines were used first to filter out the GBM-specific network from signaling-metabolic interaction network ( SMIN ) . On that GBM-specific network , we have incorporated the human GBM patient transcriptomic data as edge weights to make it more relevant to human GBM condition . The resulting model is probabilistic , based on the assumption that the likelihood of signal flow relates to the expression levels of the nodes/proteins in the different paths/pathways . It is not computing the actual signaling status of the paths and their nodes . Incorporation of phosphoproteome data in this integrated network could provide additional information about the activation/deactivation state of the nodes thus adding elements of the actual information flow . This could help to add some rationale weights for nodes that remained unchanged ( NC ) or were unidentified ( NA ) in the proteomic study as well as in GBM patient transcriptome data . Accordingly , it could add to solving the missing-data-problem of proteomics . Although phosphoproteome data could provide relevant information about signaling flow it could be less advantageous in the model of interconnection between the oncogenic signaling and metabolic pathways that are required to sustain the oncogenic processes . It could make the model more complex as expression and phosphorylation of proteins are not correlated directly; phosphorylation states are highly context- and time-dependent and requires to take into account the ratio of phosphorylated and non-phosphorylated signaling proteins at any given position within the cell and the spatial arrangements of active/inactive signaling pathway molecules . To test the network properties , the dynamic behavior and the robustness of the model , we performed in silico perturbation experiments . The in silico perturbation experiments were done with the final GBM-specific model with all interactome , proteome and transcriptome data incorporated in this order to render a general interactome map GBM-specific , and calculate node/edge weights and path score . In silico perturbation was done by silencing the signaling molecules one by one , and then redeveloping the model and recalculating node/edge weights and path scores to determine the relevance of the respective signaling molecule and the affected signaling path , to identify possible escape paths bypassing the blockade , and to identify targets for experimental validation of the model . Based on the perturbation analysis we postulated that proteins with higher impact as loss-of-function or gain-of-function or Pathscore difference between signaling-metabolic pathways and vice-versa have a higher ranking and chances to transmit information between the studied pair of pathways . The perturbation studies also captured the weighted topological changes in the dynamic network and the distinct sets of interactors that transmit information between proteins . Implementation of path score differences before and after perturbation adds additional accuracy to the model to identify the most important paths of information flow among all probable paths . Applying this model from 437 , 449 probable GBM-specific S-M paths , we identified 2 , 253 ( Z≥1 ) ( 0 . 52% ) and 876 ( Z≥3 ) ( 0 . 20% ) significant paths of information flow in mutated EGFR-dependent compared to wtEGFR EGF-driven GBM . To experimentally test and validate the interconnections predicted by the model , inhibition of signaling pathway proteins ( S ) present in the SMIN with small molecule inhibitor was performed , which resulted in alteration of metabolic pathway protein expression ( M ) . We found that the performance of the model in predicting the dynamics of the large-scale signaling networks is comparable to state-of-art methods for extracting context-dependent information flow [89] . One node knockout at a time in the in silico perturbation experiments indicated , first , the importance of that protein in the signaling-metabolic interconnected paths in establishing low or high grade GBM condition , and second , alternative signaling routes that can become relevant as escape responses to therapy . It is important to note that in biological networks including signal transduction , PPI is highly dynamic in nature and may undergo continuous changes [90] . These context-specific network dynamics need to be predicted by systems biology models . Our probabilistic integrated network model captures network dynamics using the topology of PPI networks based on dynamic data under different circumstances . The probabilistic network-based dynamic model uses experimental data and deals with the uncertainty of systems to predict drug targets and understand the effect of therapeutics . In our study , nodes forming paths significant to information flow in a particular biological context is the key to network inference . Nodes present in information flow path with high network weights and connected with higher edge weight can be considered for drug intervention and combinatorial therapy . Our model efficiently identified key nodes which are important in S-M information flow with confirmation of some known and prediction of the potential novel drug targets [89] that can be used alone or in combination to inhibit mutated EGFR-mediated GBM . This model is now a useful tool and will be used to develop strategies and experiments to study causal relationships . Among the questions to be addressed will be to understand the molecular biology of high-grade GBM versus low-grade GBM , and mechanisms of therapy resistance . However , in its current form , this model is not engineered to measure signal flow in and across signaling pathway ( s ) . It is also not applicable to other regulatory mechanisms such as transcriptional , post-translational and miRNA based modulations of biomolecular interactions . Nonetheless , the model should be robust enough to integrate and utilize large-scale pan-omics data including genomics , transcriptomics , proteomics , metabolomics and post-translational modification information to accurately model the cellular context under a given scenario . These issues are important for assessing systemic changes and we are addressing them in follow-up studies . The strategy for developing the model can also be applied to other cancers and to non-oncological conditions . Interestingly , GAPDH and PGK1 , commonly reported reference [91] and housekeeping genes [92] , were identified ( Fig 7B ) as significant ( ~40% of 20 networks in the study ) by our model and previously reported studies [47] have shown their association with cancer [93–95] . In this study , also significant expression change in two cell lines for GAPDH and PGK1 was found after inhibition of CSNK2A1 and BCL2 , respectively . To our knowledge , this has not been reported before . A study [96] with a mice model has shown increased expression level of bcl-2 under the control of pgk-1 , which suggests a possible relation to our findings . Not surprisingly , we have also found FOXO , a key player in cell fate decision [97] as candidate target . In addition , the HSP90 family gene ( HSP90AA1 ) and CREBBP were identified as important agents in forming interconnections between multiple signaling/metabolic pathways . Inhibitors of HSP90 ( 17-AAG ) and CREBBP ( ICG-001 ) have recently shown effects in glioblastoma and other cancer models/cell lines [98–103] . They have entered clinical trials for several cancer types [104 , Clinical trial numbers:NCT01606579 , NCT01764477 , NCT02413853] . This confirms that our model can identify potential therapeutic targets before performing actual drug tests . The presented network model can be used to explore novel therapeutic strategies against cancer including combination therapies . This work also represents a step towards finding alternative routes/pathways in cancer or other diseases and thereby predicting the potential path of therapy evasion that can be included in the development of new therapies that aim to prevent therapy resistance in cancer and other diseases .
The human grade-IV glioblastoma cell line U87MG expressing wild-type EGFR ( U87MG ) was purchased from ATCC ( USA ) . The U87MG-derived genetically engineered U87MGvIII cell line with exons 2–7 deleted from the EGFR gene ( EGFRvIII ) cell was a kind gift from Professors Webster K . Cavenee and Frank B . Furnari , Department of Medicine and Cancer Center , University of California at San Diego , La Jolla , USA . The cells were grown in Iscove’s Modified Dulbecco’s Medium ( IMDM; Gibco , Thermo Fisher Scientific Inc . , Schwerte , Germany ) supplemented with 10% heat-inactivated fetal calf serum ( FCS; Biochrome , Berlin , Germany ) and 1% penicillin/streptomycin solution ( Gibco , Thermo Fisher Scientific Inc . , Schwerte , Germany ) in a humidified atmosphere with 8% CO2 at 37°C . GBM cells ( 1 × 104 ) were plated in 96 well plates and treated with inhibitors of Calmodulin ( CGS-9343B , Sigma ) , Casein kinase II subunit alpha ( Emodin , Sigma ) , 1-phosphatidylinositol 4 , 5-bisphosphate phosphodiesterase gamma-1 ( U73122 , Sigma ) , Tyrosine-protein kinase ABL1 ( Dasatinib , Santacruz ) and B-cell lymphoma 2 ( ABT199 , Santacruz ) for 24hr and subsequently incubated with 3- ( 4 , 5-dimethylthiazol-2-yl ) -2 , 5-diphenyl tetrazolium bromide ( MTT , 100 μg/ml DMSO ) in fresh culture medium for 3 hr . Optical density ( OD ) was taken at 550 nm with an ELISA reader ( Thermo ) as described elsewhere [105] . U87MG and U87MGvIII were plated separately in 6-well plates with>90% confluence . Scratch-wounds were made with a micropipette tip , washed thrice to remove the floating cells and treated separately with ABL1 , PLCG1 , BCL2 , CALM2 and CSNK IIA inhibitors at their IC50 dose ( 5μM , 3μM , 25μM , 45μM and 40μM , respectively ) in medium and incubating them for the indicated time periods . Images were taken at 0 hrs and 8 hrs . The wound width was measured for untreated and treated groups from at least five different fields of three separate experiments , and percentage wound healing was calculated from the width at 8 hrs versus the initial width at 0 hrs , all using ImageJ software . U87MG and U87MGvIII cells ( 5x104 ) were seeded separately to the upper part of matrigel-coated invasion chamber in serum-free medium ( 200 μl ) and the lower chamber was filled with medium ( 600 μl ) containing respective inhibitors of ABL1 and PLCG1 at their IC50 doses . The cells on the lower surface of the insert were stained with the crystal violet after 24 hrs and the numbers of invaded cells were counted by inverted light microscopy from at least three different fields of three different experiments . U87MG and U87MGvIII cells were treated with the indicated inhibitors at half of the IC50 dose for 24hr . Total mRNA was isolated by RNeasy Mini Kit ( QIAGEN ) as per manufacturer’s instructions . RNA ( 1 . 0 μg ) was reverse-transcribed to cDNA with the Reverse Transcriptase kit ( Promega ) . For RT-PCR , the cDNA was amplified with specific primers for the mRNA of the indicated metabolic proteins in a Perkin-Elmer DNA thermal cycler . Real-time PCR analysis was performed by mixing cDNA with 2× SYBR green master mix using Roche Applied Science light cycler 480 . 0 instruments with the software version 1 . 5 . 0 . Relative quantification of each target gene was normalized to two housekeeping genes ( 18s rRNA and β-actin ) and expressed as a fold change compared with untreated control using the comparative cycle threshold ( CT ) method [6] . For proteome analysis , total proteins were extracted with SDS-PAGE sample buffer from U87MG and U87MGvIII cells grown to 80% confluency . EGFR expression status of both cell lines was confirmed by western blot analysis . Proteins ( 100 μg ) of each cell line were separated by SDS-PAGE ( 10% acrylamide/0 . 8% bisacrylamide ) . After staining with Coomassie blue , the lanes for both cell lysate were sliced into one-mm thick gel pieces . The slices were transferred into 96 well round bottom polypropylene plate and destained with 100 μl , 20 mM ammonium carbonate/acetonitrile ( 60%/40% v/v ) , dehydrated by adding twice acetonitrile ( 50 μl ) and dried under vacuum in a speed vac . The dried gel slides were rehydrated in 20 μl trypsin solutions ( 10 ng/μl ) in 30 μl NH4HCO3 ( 20 mM ) for digestion by incubation for 16–18 hr at 37°C . The supernatants were collected in 1 . 5 ml reaction tubes and the remaining tryptic fragments extracted from the gel slices first with 50 μl 50% acetonitrile with 0 . 1% TFA for 15 min , and then with 5% acetonitrile with 0 . 1% TFA . The extracts from each gel slice were combined and dried in a speed vac . The peptides were re-dissolved in 12 μl 2% acetonitrile with 0 . 05% TFA . For MS/MS , the peptides of the tryptic digests ( 10 μl ) were loaded via a pre-column at a flow rate 20 μl/min ( 2% acetonitrile , 0 . 05%TFA ) onto an Acclaim PepMap C18 nano-HPLC column ( 75μm inner diameter×15 cm length; Thermo Fisher Scientific Inc . , Schwerte , Germany ) using an Ultimate 3000 nano-HPLC system ( Dionex , Darmstadt , Germany ) . The peptides were eluted with a gradient of 5–60% solvent B ( 0 . 1% formic acid in 95% acetonitrile/5% water ) in solvent A ( 0 . 1% formic acid in 1% acetonitrile/water ) over 60 min , followed by 60–90% solvent B over 5 min , and 90% solvent B for 5 min at a flow rate of 220 nl/min . The nano HPLC system was directly coupled to a Micro-TOF-Q I mass spectrometer ( BrukerDaltonics , Bremen , Germany ) . Mass spectra were acquired in the m/z range 50–2500 at an acquisition rate of 1 . 3 per sec . MS/MS spectra were acquired in the data-dependent mode with the fragmentation of the 5 most intensive peaks ( absolute threshold 3000 ) with argon as collision gas , and fragment masses in the range of 400–1400 m/z . The mass spectrometry was run with dynamic exclusion of a time interval established and tested beforehand to avoid or minimize signal suppression by over-abundant peptides . A dynamic exclusion of 1 min was set to avoid repeated fragmentation of the most abundant peptides . The MS and MS/MS spectra were processed with Data Analysis 3 . 4 and Biotools 3 . 1 ( BrukerDaltonics ) . MS/MS searches for peptide identification were done via Biotools on a local Mascot ( version 2 . 2 ) server with a precursor mass tolerance of 50 ppm and a fragment mass tolerance of 0 . 2 Da . Trypsin was specified as an enzyme , and one allowed missed cleavage and oxidation of methionine as variable modification . Data searches were done in SwissProt databank for human proteins . Proteins were identified with at least two peptides with a mascot score of higher than 26 . The false discovery rate with these search and filter parameters was below 5% as confirmed with a decoy database . Exponentially modified protein abundance index ( emPAI ) was employed for protein quantification using the equation emPAI=10NobservedNobservable-1 ( i ) where Nobserved is the number of experimentally observed and Nobservable the calculated number of observable peptides for each protein [106] . Comparative proteome analysis was done with proteins identified by at least two peptides based on the emPAI scores for differentially expressed proteins . The proteome analysis was done three times for each cell line as an independent biological replicates . Comparative proteome analysis was done for each proteome dataset ( batch ) of U87MG with each batch of U87MGvIII using the in-house PROTEOMESTAT-12 software . For this , the emPAI values of all proteins were normalized as emPAInorm=emPAIprotein[∑emPAIprotein]batch×100 ( ii ) where emPAIprotein was the emPAI value of a protein in a batch . Multiple identifications of a protein within a proteome dataset were excluded , only keeping the highest emPAI value for subsequent analysis . Intra-batch analysis was performed to determine the total unique proteins as well as the overlap between the three batches of each cell line . Total unique proteins identified from U87MG and U87MGvIII cells were compared to identify the common and the differentially expressed proteins . Proteins identified only for one of the cell lines were considered up-regulated in that cell line . Commonly expressed proteins were further analyzed in two different ways to identify differentially expressed proteins . In one approach , the emPAIave values for each protein were compared between the two cell lines ( iii ) . emPAIave=∑batch=13emPAIproteinNumberofbatcheswithidentifiedprotein ( iii ) where emPAIprotein was the emPAI value of a protein in a proteome dataset . The percentages of up or down-regulation of a protein in U87MGvIII were calculated as % ExpU87MGvIII using the following the equation %ExpU87MGvIII= ( emPAIaveU87MGvIII-emPAIaveU87MG ) emPAIaveU87MG×100 ( iv ) where emPAIave U87MGvIII and emPAIaveU87MG were the average emPAI value of a protein among the three proteome datasets for U87MGvIII or U87MG . Proteins that were ≥ 50% up- or down-regulated with p values ≤ 0 . 2 were considered differentially expressed in the U87MGvIII cell . In a second approach , emPAInorm of a protein of each batch of one of the cell lines were compared separately with the emPAInorm for the same protein in each of the three batches of the other cell line . Then these comparative data were normalized by the LOESS method . Up- or down-regulation of a protein in U87MGvIII were calculated using the Eq ( iv ) . From this second approach ≥ 50% up- or down-regulated proteins with an SE ≤ 35% were taken as differentially expressed in U87MGvIII . Both the results were combined and duplicates were excluded to identify the differentially expressed proteins . A signaling-metabolic cross-connected network was created using 14 signaling and 81 metabolic pathways and protein-protein interactors of the pathway proteins . Signaling pathway databases were created by integrating information from on-line resources including KEGG [107] , Reactome [108] Signallink [109] , NetPath [110] and Biocarta ( http://www . biocarta . com ) for extensive coverage of the steps involved in the pathways . The 14 signaling pathways were included based on their association with cancer in general according to KEGG pathway database . Metabolic pathway datasets were collected from KEGG . Human protein interaction data were taken from the STRING interactome database using only protein-protein interactors [70] established by high-throughput analyses with the high confidence score ( ≥ 0 . 7 ) to generate a protein-protein interaction network ( PPIN ) . PPI were restricted up to the second level interactors , i . e . interactors of interactors , which brings about S ( PPI ) 3M , meaning S-PPI1-PPI2-PPI3-M paths . PPI1 is the first level interactor of signaling molecule and PPI3 is the first level interactor of metabolic enzyme and PPI2 the second level interactor of both . Thereby , there are three PPIs between the signaling and the metabolic pathway proteins . The resulting network consisted of 5 , 703 proteins with 16 , 828 experimentally confirmed interactions . NetworkX ( https://networkx . github . io/ ) , a network analysis framework in a Python language , was used to calculate the degrees , Estrata Index and betweenness centrality of nodes ( genes/proteins ) . Gene expression data used for edge weight determination in the weighted-network were collected from five gene expression data sets for GBM patients taken from GEO ( Gene Expression Omnibus ) [111] GSE4290 ( 77 samples ) , GSE53733 ( 70 samples ) , GSE50161 ( 34 samples ) , GSE36245 ( 46 samples ) and GSE15824 ( 12 samples ) ( http://www . ncbi . nlm . nih . gov/geo/ ) . All had been acquired with the Affymetrix GPL570 platform [http://www . ncbi . nlm . nih . gov/geo/query/acc . cgi ? acc=GPL570] . The raw data of the total of 239 samples were normalized across all arrays within a set using the RMA quintile normalization procedure of the Gene Pattern Expression File Creator module [112] . Averages of normalized raw expression values were used for the study . To construct cross-connecting linking paths ( CCLPs ) between signaling pathway proteins ( S ) and metabolic pathways proteins ( M ) , common interacting proteins from the HPPIN were used . All possible unique connections between S and M with S-M ( direct S-M interaction ) , and S-P-M , S-P-P-M , and S-P-P-P-M where P is a common interacting protein were established for common interacting proteins up to the second level in HPPIN ( Fig 3A ) . Detailed descriptions of all possible connecting links are provided in the Fig 3B and Table 1 . Differentially expressed ( i . e . up-regulated or down-regulated ) proteins in EGFRvIII versus EGFRwt expressing cells obtained from proteome analyses were mapped onto the cross-connected links [Table 1] . To assess the relevance of signaling to metabolic pathway interconnections for cancer pathogenesis , connections with at least one up-regulated or down-regulated protein ( U87MGvIII versus U87MG ) were selected to create functional cross-connecting sub-network ( CCsN ) with , in all , 2 , 320 protein nodes and 8 , 914 interactions . This CCsN was used for network topology analysis . The local entropy of a protein in CCsN was calculated on the basis of probabilities of an interaction of that protein with its interactors determined by using the principle of mass action . The calculation of the interaction probabilities is based on the assumption that two proteins known to interact will have a higher probability of interaction when they are highly expressed . Thereby , the interaction probability ( Wij ) of two proteins in a network is proportional to the product of expression values ( E of the corresponding genes ( Ei and Ej ) [72]: Wij∝EiXEj ( v ) with the expression value of CCsN genes calculated as an average expression of a gene in the data set of 239 GBM patients . Based on Wij , a stochastic matrix of normalized interaction probabilities ( Pij ) in the network was created for signaling entropy calculation . Probability of interaction between node i and node j was calculated as Pij=Wij∑k∈NiWik ( vi ) where Ni is interactors of node I with ΣPij = 1 . The local entropy of node i was calculated as Si=−1logki∑j∈N ( i ) pijlogpij ( vii ) where ki is the degree of node i in the CCsN . To incorporate the impact of the interactors of a particular protein in the cross-connected network , the node-weight of every node i was specified based on the categories signaling cross-talk protein ( SC ) , rate limiting enzyme ( RLE ) and up- or down-regulation in U87MGvIII versus U87MG ( dEXP ) : Wi={1;if ( dEXP=13 , RLE=13 , SC=13 ) , 0;else} ( viii ) Effect of interactors on a node s in CCsN was defined as effect-on-node ( effs ) depending on the node weight of the interactors up to the second level: effs=∑jk ( ∑inWidegreei+Wjnj ) ( ix ) where k is the degree of node s , nj is the degree of protein j , wj and wi are the node weights of protein j and i . Rate-limiting enzymes ( RLE ) are the enzymes in metabolic pathways whose kinetics determines the overall kinetics of the entire pathway . It is usually the enzyme with the slowest enzyme kinetics . The identification of the RLE is done based on the Michaelis–Menten equation , and the established Michaelis constants ( KM ) and Vmax of all enzymatic steps in metabolic pathways are listed in the respective biophysics and enzyme kinetic textbooks and databases . The CCLP for information flow from signaling pathway protein ( S ) to metabolic pathway protein ( M ) was defined to be one of the four types shown in Fig 5A , and path scores were calculated on the basis of node and edge weights of the proteins involved in a path . To select important S-M pairs , imaginary penultimate signaling and metabolic proteins are considered as starting and ending state , and path score was calculated based on a hidden Markov model ( HMM ) with a forward algorithm . Emission probability ( Ej ) , i . e . the positional probability of a protein at the particular position in that state was calculated as Ej=Sj+Effj∑ik ( Si+Effi ) ( x ) where k is the number of proteins in that state , Si and Sj are normalized local entropy of proteins i and j , Effi is effect-on-node for protein i . Within the S-M pairs of a path ( Fig 5B ) information flow is again scored by considering all types of paths formed between single S-M pairs and calculated as Pathscore=I∏j=1nEjTj ( xi ) where n is number states in a path , I initial probability ( in our case it is equal to one ) , Ej is emission probability at state j , Tj is transmission probability at state j ( in our case it is the probability of interaction Pij ) . Pathscore is converted into Z-score as Zscore=X-μσ ( xii ) where X is raw Pathscore , μ is mean , σ is the standard deviation . A cut-off of ≥1 is applied to select significant S-M pairs and their cross-connecting links . To understand the significance of the path scores in comparison to randomly generated paths , we have performed permutations of node weights ( Sij ) and edge weights ( Pij ) and generated 20 random paths with the same number of proteins involved in the identified paths . Averages of 20 path scores are compared with the corresponding original path score . S8 Fig shows significant difference between the original and the randomly placed weighted path scores . In silico perturbation analysis was done by removing nodes/proteins from the network using an in-house programme for nodes/proteins present in the final weighted sub-network . All nodes in this sub-network were removed individually from the SMIN , and the impact was studied by performing the same weighted network analysis with the sub-networks generated after node removal as the starting point . Perturbation score ( Ps ) was calculated in two steps . First , we calculated significant pairs using Model 1 shown in Fig 5A and significant paths using Model 2 shown in Fig 5B using SIN-Ni network where Ni is the interaction of perturbed node N . Second , path scores after perturbation ( Pathscore' ) for significant paths identified from step 1 were calculated . Perturbation score ( Ps ) was defined as Ps=Pathscore−Pathscore , To identify nodes with significant impact after perturbation we converted perturbation scores ( Ps ) into Z-score for all signaling to metabolic pathways and vice-versa . In the perturbation analysis , node and edge weights were re-calculated for networks and paths generated after removing a node ( protein ) from the initial interactome . New weights in Model 1 and Model 2 were used to score the new paths and to compare the scores before and after perturbation thus to identify significant nodes in the network forming connections between signaling and metabolic pathway proteins . KEGG pathway based over-representation analysis ( ORA ) was performed using 136 up-regulated and 243 down-regulated proteins ( EGFRvIII vs EGFRwt ) and 45 commonly up-regulated and 25 down-regulated proteins using ‘protein coding gene set’ as the reference gene set in WebGestalt [113] web tool . Additionally , Gene ontology ( GO ) based biological process and molecular function over-representation analysis was performed for the same genes . Top 20 significant ( p-values <= 0 . 01 and <= 0 . 05 ) categories were ranked based on the false detection rate ( FDR ) calculated using Benjamini and Hochberg procedure . | Complex and highly dynamic interconnected networks allow cancer to take different routes and circumvent chemotherapy . Therefore , understanding these context-specific networks and their dynamics of molecular interactions driven by different oncogenic signaling and metabolic pathways is very much needed to predict drug targets and the effect of therapeutics . We incorporated high-throughput transcriptome and proteome data into mathematical models to deduce properties of cancer cells through systems biology approach . Here we report the development , testing and validation of an integrated systems biology model of information flow between signaling and metabolic pathways to understand the regulation of the interconnection between them in cancer . Our model efficiently identified unique connections and key nodes important in signaling-metabolic information flow . We predicted some potential novel targets before performing actual drug tests . We have successfully applied this model to identify the interconnections altered in the constitutive signaling of the mutated EGFR by comparing EGF-dependent and wild-type EGFR signaling in glioblastoma multiforme . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"protein",
"interaction",
"networks",
"signaling",
"networks",
"protein",
"expression",
"network",
"analysis",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"oncogenic",
"signaling",
"proteins",
"metabolic",
"pathways",
"ras",
"signaling",
"proteomics",
"metabolism",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"proteomes",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling"
] | 2019 | Connecting signaling and metabolic pathways in EGF receptor-mediated oncogenesis of glioblastoma |
Viral infection triggers induction of type I interferons ( IFNs ) , which are critical mediators of innate antiviral immune response . Mediator of IRF3 activation ( MITA , also called STING ) is an adapter essential for virus-triggered IFN induction pathways . How post-translational modifications regulate the activity of MITA is not fully elucidated . In expression screens , we identified RING finger protein 26 ( RNF26 ) , an E3 ubiquitin ligase , could mediate polyubiquitination of MITA . Interestingly , RNF26 promoted K11-linked polyubiquitination of MITA at lysine 150 , a residue also targeted by RNF5 for K48-linked polyubiquitination . Further experiments indicated that RNF26 protected MITA from RNF5-mediated K48-linked polyubiquitination and degradation that was required for quick and efficient type I IFN and proinflammatory cytokine induction after viral infection . On the other hand , RNF26 was required to limit excessive type I IFN response but not proinflammatory cytokine induction by promoting autophagic degradation of IRF3 . Consistently , knockdown of RNF26 inhibited the expression of IFNB1 gene in various cells at the early phase and promoted it at the late phase of viral infection , respectively . Furthermore , knockdown of RNF26 inhibited viral replication , indicating that RNF26 antagonizes cellular antiviral response . Our findings thus suggest that RNF26 temporally regulates innate antiviral response by two distinct mechanisms .
Host pattern-recognition receptors ( PRRs ) detect nucleic acid from invading viruses or necrotic cells and trigger a series of signaling events that lead to the induction of type I interferons ( IFNs ) , which plays a central role in autoimmune diseases as well as protective immune responses against viruses , respectively [1] , [2] . Much progress has been made to characterize viral nucleic acid-triggered signaling pathways that result in transcriptional activation of type I IFN genes . A family of DExD/H box RNA helicases consisting of retinoic acid inducible gene I ( RIG-I ) , melanoma differentiation-associated gene 5 ( MDA5 ) , and LGP2 are RNA sensors and recruit the adaptors VISA ( also called MAVS , IPS-1 and Cardif ) [3]–[6] and MITA ( also known as STING , MPYS and ERIS ) [7]–[10]to activate the transcription factors NF-κB and interferon regulatory factor ( IRF ) 3 or IRF7 , leading to transcriptional induction of the genes encoding type I IFNs and other antiviral effectors [2] . A number of DNA sensors have been identified , including DAI , IFI16 , DDX41 and MRE11 which may function in a ligand- and/or cell-type-specific manner [11]–[14] . In addition , cyclic GMP-AMP synthase ( cGAS ) has been recently characterized as a viral DNA sensor in almost all types of cells [15] . These sensors depend exclusively on the adaptor MITA to activate NF-κB and IRF3/7 and lead to subsequent induction of type I IFNs [16] . MITA is localized to the endoplasmic reticulum ( ER ) , mitochondria-associated membrane and mitochondria [7] , [8] , [10] , [17] . Upon viral infection , MITA translocates to intracellular membrane-containing compartments to form punctate aggregates and acts as a scaffold protein to facilitate the phosphorylation of IRF3 and STAT6 by the kinases TBK1 and IKKε [18]–[21] . Recent studies also suggest that MITA is a direct sensor that recognizes cyclic dinucleotides such as c-di-AMP , c-di-GMP and cGAMP generated from self and viral DNA infection [22]–[24] . It has been demonstrated that MITA undergoes various post-translational modifications and such modifications are key to the activity and stability of MITA [7] , [21] , [25]–[27] . MITA is phosphorylated at Ser358 by TBK1 which is critical for phosphorylation and activation of IRF3 [7] , while UNC-51-like kinase ( ULK1 ) phosphorylates MITA at Ser366 which impairs MITA-IRF3 interaction and subsequent activation of IRF3 [21] . In addition , RNF5 catalyzes K48-linked polyubiquitination of MITA and targets MITA for degradation [25] , whereas TRIM56 and TRIM32 promote K63-linked polyubiquitination of MITA and positively regulates virus-triggered type I IFN induction [26] , [27] . Whether and how additional proteins mediate other types of modifications of MITA is unknown . In the present study , we identified an E3 ubiquitin ligase , RING finger protein 26 ( RNF26 ) that targeted MITA for K11-linked polyubiquitination upon viral infection . MITA with K11-linked polyubiquitin chains remained as reservoir of MITA and was protected from RNF5-mediated K48-linked polyubiquitination and degradation . However , overexpression of RNF26 indirectly induced degradation of IRF3 through an autophagy pathway . As a result , knockdown of RNF26 promoted degradation of MITA after viral infection and prevented degradation of IRF3 . Virus-triggered phosphorylation of IRF3 and induction of IFN-β were inhibited at the early time points and potentiated at late time points in the absence of RNF26 , respectively . Our findings suggest that RNF26 temporally regulates virus-triggered induction of type I IFNs by two distinct mechanisms .
Because post-translational modifications of MITA are critical for mediating viral nucleic acid-triggered type I IFN induction , we assumed there are additional proteins that interact with and target MITA for various modifications and thereby regulate innate antiviral signaling . We attempted to unambiguously identify E3 ubiquitin ligases that regulate MITA ubiquitination and function [27] . This effort led to the identification of RNF26 which could promote polyubiquitination of MITA [27] . Analysis of the NCBI EST profile database indicates that RNF26 is ubiquitously expressed in most examined cells and tissues . In overexpression experiments , RNF26 dose-dependently promoted polyubiquitination of MITA ( Figure 1A ) . In contrast , the enzymatic inactive mutants RNF26 ( C395S ) , RNF26 ( C399S ) or RNF26 ( C401S ) failed to mediate polyubiquitination of MITA ( Figure 1B ) . Results from in vitro ubiquitination assays demonstrated that RNF26 catalyzed polyubiquitination of MITA in vitro , which depended on the enzymatic activity of RNF26 ( Figure 1C–D ) . These data suggest that RNF26 is an E3 ubiquitin ligase targeting MITA for polyubiquitination . Since RNF26 caused polyubiquitination of MITA , we examined whether RNF26 interacted with MITA . Transient transfection and coimmunoprecipitation experiments indicated that RNF26 was associated with MITA in 293 cells ( Figure 2A ) . In untransfected THP-1 cells , endogenous RNF26 constitutively interacted with MITA . This interaction was enhanced at 6 hours and decreased at 12–24 hours after SeV or HSV-1 infection ( Figure 2B ) . It has been reported that MITA is localized to the ER , mitochondria-associated membrane and mitochondria [7] , [8] , [10] , [17] . The subcellular localization of RNF26 was examined . Fluorescent confocal microscopy and cellular fractionation analysis suggested that RNF26 was mainly localized to the ER and a minor fraction of RNF26 was found to be colocalized with the mitochondria marker ( Figure 2C and D ) . In contrast , RNF26 was not colocalized with the Golgi marker ( Figure 2C ) . RNF26 was colocalized with MITA mostly at the ER and formed punctate dots with MITA after SeV or HSV-1 infection ( Figure 2E ) . In this context , MITA was reported to translocate to microsomes to form punctate aggregates after viral infection or transfection of poly ( dA:dT ) or interferon stimulatory DNA ( ISD ) [17] , [18] , [20] , [21] . These data suggest that RNF26 is physically associated with MITA . Previously , we have demonstrated that the N-terminal transmembrane domains of MITA are critical for its subcellular localization and function [7] . Interestingly , sequence analysis indicated that RNF26 contained five transmembrane domains at the N-terminus and a RING domain at the C-terminus ( Figure S1A ) . Domain mapping experiments indicated that the association of RNF26 and MITA depended on their respective transmembrane domains ( Figure S1A and B ) . These data collectively suggest that RNF26 is physically associated with MITA and the interaction is dependent on their transmembrane domains . To map the residue ( s ) of MITA that are targeted by RNF26 , we examined RNF26-mediated polyubiquitination of MITA mutants in which all the lysine residues of MITA were individually substituted by arginine . As shown in Figure 3A , mutation of K150 to arginine impaired its polyubiquitination by RNF26 . In addition , RNF26 could not induce polyubiquitination of MITA ( K150R ) in in vitro ubiquitination assays ( Figure 3B ) . These data suggest that RNF26 targets K150 of MITA for polyubiquitination . Having demonstrated that RNF26 is a MITA-interacting E3 ubiquitin ligase targeting MITA for polyubiquitination , the types of polyubiquitin chains conjugated to MITA by RNF26 were next examined . Ubiquitin mutants were constructed in which all but one lysine residues were simultaneously mutated to arginines ( K-O ) or all seven lysine residues were individually mutated to arginine ( K-R ) . These mutants were examined for their abilities to be conjugated to MITA by RNF26 . As shown in Figure 4A , the ubiquitin mutant retaining only lysine 11 ( Ub-K11O ) but not other lysine residues could be conjugated to MITA . Conversely , mutation of lysine 11 to arginine ( Ub-K11R ) markedly reduced its ability to be conjugated to MITA by RNF26 ( Figure 4B ) . These data indicate that RNF26 induces K11-linked polyubiquitination of MITA . Since an antibody against K11-linkage of polyubiquitin chains was not available to us , THP-1 cells stably transfected with Flag-Ub-K11O ( THP-1-Flag-Ub-K11O ) together with a control or RNF26-RNAi plasmids ( Figure 4C and D ) were established to examine viral infection-induced K11-linked polyubiquitination of MITA and the role of RNF26 in this ubiquitination . As shown in Figure 4E , MITA was modified with K11-linked polyubiquitin chains at 6–12 hours after SeV or HSV-1 infection , and such polyubiquitination was substantially impaired by knockdown of RNF26 . Similar results were obtained in 293 cells stably transfected with HA-Ub-K11O plasmids ( 293-HA-Ub-K11O ) together with a control or RNF26-RNAi plasmids ( Figure 4F ) . In similar experiments , K11-linked polyubiquitination of VISA after viral infection was not affected by RNF26 knockdown ( Figure S2 ) . These data suggest that RNF26 mediates K11-linked polyubiquitination of MITA upon viral infection . Because multiple E3s have been reported to target polyubiquitin chains of distinct linkages ( K48-linked or K63-linked ) to K150 of MITA [25]–[27] , we speculated that RNF26-mediated K11-linked polyubiquitination at K150 might compete with K48- or K63-linked polyubiquitination at the same lysine residue . Thus , virus-induced K48- or K63-linked polyubiquitination of MITA in THP-1-RNF26-RNAi and control cells was examined . As shown in Figure 5A , SeV or HSV-1 infection triggered K48- and K63-linked polyubiquitination of MITA . Knockdown of RNF26 potentiated K48- but not K63-linked polyubiquitination of MITA after viral infection ( Figure 5A ) . RNF5 has been shown to induce K48-linked polyubiquitination of MITA at K150 [25] . Interestingly , we found that knockdown of RNF5 greatly enhanced SeV-induced K11-linked polyubiquitination of MITA in 293-HA-Ub-K11O cells ( Figure 5B ) . In addition , RNF26-mediated K11-linked polyubiquitination of MITA was diminished by co-expression of RNF5 , and RNF5-mediated K48-linked polyubiquitination and degradation of MITA was partially inhibited by co-expression of RNF26 ( Figure 5C ) . Consistent with these observations , SeV- or HSV-1-induced degradation of MITA was accelerated by knockdown of RNF26 ( Figure 5D ) . These data suggest that RNF26 catalyzes K11-linked polyubiquitination of MITA which protects MITA from RNF5-mediated K48-linked polyubiquitination . Since RNF26-mediated K11-linked polyubiquitination of MITA protected its degradation after viral infection , whether RNF26 regulates virus-triggered induction of type I IFNs was determined . Luciferase reporter assays suggested that knockdown of RNF26 inhibited SeV-induced activation of NF-κB , ISRE and IFN-β promoter but had no marked effects on TNFα- or IL-1β-induced activation of NF-κB ( Figure 6A and S3A ) . The expression of IFNB1 gene was impaired at 6 hours after SeV infection by knockdown of RNF26 in 293 cells ( Figure S3B ) . However , we unexpectedly observed that SeV-induced expression of IFNB1 was potentiated in RNF26 knockdown cells compared to control cells at 12–24 hours after SeV infection ( Figure S3B ) . In a mouse macrophage cell line Raw264 . 7 cells , SeV- or HSV-1-induced expression of Ifnb1 gene was also inhibited and potentiated at the early and late time points by knockdown of murine Rnf26 , respectively ( Figure S3C and D ) . It should be noted that the degrees of inhibition or potentiation of expression of IFNB1 or Ifnb1 genes were correlated with the knockdown efficiencies of the RNF26-RNAi or Rnf26-RNAi plasmids , indicating that RNF26 is involved in regulating RNA and DNA virus-triggered induction of type I IFNs . To further confirm this notion , RNAi-transduced stable THP-1 cell lines were established and the expression of IFNB1 in these cells was examined after stimulation with SeV or HSV-1 . As shown in Figure 6B and C , the mRNA and protein levels of IFN-β were impaired at 6 hours and potentiated at 12–24 hours in THP-1-RNF26-RNAi compared to control cells after viral infection , respectively . Similar results were obtained with various RNA ( VSV and EMCV ) or DNA ( ECTV ) viruses ( Figure 6D ) as well as virus-induced expression of CCL5 ( Figure S3E ) . Interestingly , SeV- or HSV-1-induced expression of the proinflammatory cytokines TNFα and IL-6 was inhibited by RNF26 knockdown at all examined time points after viral infection ( Figure 6E , 6F , S4A and B ) . In similar experiments , RNF26 knockdown did not affect IFN-β-induced expression of ISG15 or ISG56 genes ( Figure S4C ) . Consistent with the gene induction experiments , we found that although SeV- or HSV-1-induced phosphorylation of TBK1 and IκBα was inhibited in THP-1-RNF26-RNAi cells , phosphorylation of IRF3 was inhibited at the early time points and increased at the late time points in THP-1-RNF26-RNAi stable cells compared to that in control cells after viral infection ( Figure 6G ) . Thus , we conclude that RNF26 temporally regulates virus-triggered induction of type I IFNs by two distinct mechanisms . When examining virus-triggered activation of IRF3 in RNF26 knockdown and control cells , we observed that the level of IRF3 protein was raised in THP-1-RNF26-RNAi compared to control cells ( Figure 6G and 7A ) , and the mRNA levels of IRF3 were comparable in these cells ( Figure 7A ) , indicating that RNF26 regulates IRF3 at the protein level . In support of this notion , we found that overexpression of RNF26 but not RNF26 ( C395S ) promoted degradation of IRF3 and inhibited SeV-induced activation of the IFN-β promoter ( Figure 7B and C ) . However , we failed to observe an interaction between IRF3 and RNF26 ( Figure 2A ) or polyubiquitination of IRF3 by RNF26 ( Figure 7D ) . Interestingly , RNF26-mediated degradation of IRF3 was blocked by the autophagy inhibitor 3-methyladenine ( 3-MA ) but not the lysosome inhibitor ammonium chloride ( NH4Cl ) or the proteasome inhibitor MG132 ( Figure 7E ) . To further determine whether autophagic degradation system is responsible for RNF26-mediated degradation of IRF3 , we determined the effect of knockdown of ATG12 , an important component of the autophagic degradation system [28] , on RNF26-mediated IRF3 degradation . The results indicated that RNF26-mediated IRF3 degradation was markedly inhibited in ATG12 knockdown cells ( Figure 7F ) . Thus , RNF26 might temporally regulate virus-triggered type I IFN induction through regulating K11-linked polyubiquitination of MITA at the early phase and autophagy-dependent degradation of IRF3 at late phase , respectively . Since RNF26 regulates virus-induced expression of IFN-β and other downstream genes , we examined its roles in cellular antiviral response . We found that knockdown of RNF26 inhibited VSV and HSV-1 replication in plaque assays ( Figure S5 ) , suggesting that RNF26 functions as a negative regulator in cellular antiviral response .
MITA plays critical roles in virus-triggered type I IFN induction and innate antiviral immune response [7] , [8] , [10] , [15] , [17] , [22]–[24] . Various studies have shown that post-translational modifications of MITA are essential for its function [7] , [21] , [25]–[27] . In this study , we demonstrated that RNF26 but not the enzymatic inactive mutants induced polyubiquitination of MITA . RNF26 mediated K11-linked polyubiquitination of MITA and modulated expression of type I IFN triggered by viral infection . RNF26 was localized mainly at the ER and constitutively interacted with MITA through their respective transmembrane domains . Viral infection potentiated this association and induced RNF26 to form punctate dots with MITA . Our studies suggest that RNF26 is a MITA-interacting E3 ubiquitin ligase which targets MITA for K11-linked polyubiquitination . Polyubiquitination of MITA catalyzed by RNF26 was mapped to K150 , which is also targeted by RNF5 for K48-linked polyubiquitination and by TRIM56 or TRIM32 for K63-linked polyubiquitination [25]–[27] . These observations prompted us to hypothesize that polyubiquitin chains of distinct linkages might compete with each other at the same residue of MITA . Interestingly , virus-triggered K48- but not K63-linked polyubiquitination of MITA was enhanced by knockdown of RNF26 . Previously , it has been demonstrated that TRIM32 targets not only K150 but also K20 , 224 and 236 , whereas RNF5 targets only K150 of MITA [27] . This is consistent with our observations that RNF26 impaired K48- but not K63-linked polyubiquitination of MITA . RNF26-mediated K11-linked polyubiquitination of MITA protected it from RNF5-mediated K48-linked polyubiquitination and degradation . Consistently , degradation of MITA was accelerated in RNF26 knockdown cells compared to control cells after viral infection . These results suggest that RNF26-meidated K11-linked polyubiquitination competes with RNF5-mediated K48-linked polyubiquitination of MITA at K150 . These results are consistent with our observations that knockdown of RNF26 inhibited induction of type I IFNs at early phase of viral infection . The functions of K11-linked polyubiquitination are not well understood so far [29]–[33] . To our knowledge , our study represents the first report on the function of K11-linked polyubiquitination in virus-triggered signaling and innate antiviral response . Unexpected , although knockdown of RNF26 inhibited virus-triggered induction of type I IFNs at the early phase of viral infection , it had opposite effect at the late phase of viral infection . This led us to hypothesize that RNF26 temporally regulates virus-triggered type I IFN induction by distinct mechanisms . In this context , we observed that overexpression of RNF26 promoted degradation of IRF3 , whereas knockdown of RNF26 increased the level of IRF3 . Interestingly , RNF26-mediated degradation of IRF3 was blocked by the autophagy inhibitor 3-MA but not the lysosome inhibitor NH4Cl or the proteasome inhibitor MG132 . In addition , RNF26-induced degradation of IRF3 was markedly inhibited by knockdown of ATG12 , an essential component in the autophagic degradation pathway . These results indicate that RNF26 indirectly regulates stability of IRF3 protein in an autophagy-dependent manner . The exact mechanism on how RNF26 mediates autophagy-dependent degradation of IRF3 is currently unknown . Based on our findings , we come to a working model on how RNF26 temporally regulates virus-triggered type I IFN induction ( Figure 8 ) . At the early phase of infection , RNF26 mediates K11-linked polyubiquitination of MITA , which protects it from K48-linked polyubiquitination and degradation to facilitate the fast induction of type I IFN genes . In addition to its early phase function through K11-linked polyubiquitination of MITA , RNF26 constitutively down-regulates IRF3 level by autophagic degradation . This may contribute to the termination of type I IFN induction at the late phase of viral infection . Interestingly , since IRF3 activation is not required for virus-triggered induction of the proinflammatory cytokines such as TNFα and IL-6 , RNF26 positively regulates virus-triggered induction of the proinflammatory cytokines in a constitutive but not temporal manner . Therefore , our findings not only reveal the mechanisms on how RNF26 temporally modulate virus-triggered type I IFN induction , but also provide an explanation on how virus-triggered induction of type I IFNs and proinflammatory cytokines can be distinctly regulated .
Recombinant IFN-β , TNFα and IL-1β ( R&D Systems ) ; mouse monoclonal antibodies against FLAG ( Sigma ) , HA ( Covance ) , β-actin ( Sigma ) , AIF , KDEL ( Santa Cruz Biotechnology ) , β-tubulin ( Invitrogen ) , HSV-1 ICP27 , ATG12 ( Abcam ) , TBK1 , p-TBK1 and p-IκBα ( CST ) ; rabbit polyclonal antibodies against ubiquitin , IRF3 , p-IRF3 ( Santa Cruz Biotechnology ) , polyubiquitin K48-linkage and K63-linkage ( Millipore ) were purchased from the indicated manufacturers . SeV , HSV-1 , VSV , EMCV , ECTV , anti-SeV , anti-RIG-I , anti-VISA , anti-MITA anti-RNF5 and anti-IκBα sera were previously described [7] , [25] , [27] , [34] . Rabbit anti-RNF26 was raised against recombinant human RNF26 ( 241–433 ) . IFN-β , ISRE and NF-κB luciferase reporter plasmids , mammalian expression plasmids for HA- , Flag- , or GFP-tagged MITA and its mutants , ubiquitin , RIG-I , VISA , TRAF3 , TRAF6 , TBK1 , IRF3 , IRF7 , Sec61-β , GALT , RNF5 and β-actin were previously described [7] , [25] , [27] , [34] , [35] . Mammalian expression plasmids for human Flag- , GFP- , Cherry- or CFP-tagged RNF26 and its mutants were constructed by standard molecular biology techniques . The cells were seeded and transfected the following day by standard calcium phosphate precipitation method or by FuGENE ( Roche ) . Empty control plasmids were added to ensure that each transfection receives the same amount of total DNA . To normalize for transfection efficiency , pRL-TK Renilla luciferase reporter plasmids were added to each transfection . Luciferase assays were performed using a dual-specific luciferase assay kit ( Promega ) . Firefly luciferase activities were normalized on the basis of Renilla luciferase activities . The cells were lysed in lysis buffer containing 1% SDS and denatured by heating for 5 minutes . The supernatants were diluted with regular lysis buffer until the concentration of SDS was decreased to 0 . 1% . The diluted supernatants were subjected for immunoprecipitation as described [7] , [25] , [27] , and the immunoprecipitates and whole cell lysates were analyzed by immunoblots with the indicated antibodies . The tested proteins were expressed with a TNT Quick-coupled Transcription/Translation Systems kit ( Promega ) following instructions of the manufacturer . Ubiquitination was analyzed with an ubiquitination kit ( Enzo Life Science ) following protocols recommended by the manufacturer . The transfected cells were incubated with the ER-Tracker Blue/White or Mito-Tracker Red ( Invitrogen ) following protocols recommended by the manufacturer . The cells were then fixed with 4% paraformaldehyde for 10 minutes and observed with an Olympus confocal microscope under a ×60 oil objective . The cells were washed with PBS and lysed by douncing 30 times in 2 mL of homogenization buffer ( 10 mM Tris-HCl pH 7 . 4 , 2 mM MgCl2 , 10 mM KCl and 250 mM sucrose ) on wet ice . The homogenate was centrifuged at 500×g for 10 minutes twice . The supernatant ( S5 ) was centrifuged at 5 , 000×g for 30 minutes to precipitate crude mitochondria ( P5K ) . The supernatant ( S5K ) was further centrifuged at 50 , 000×g for 60 minutes to generate S50K and P50K . Double-stranded oligonucleotides corresponding to the target sequences were cloned into the pSuper . Retro RNAi plasmids ( oligoengine Inc . ) . The following sequences were targeted for human RNF26 cDNA: #1: 5′-GAGCAAGAGGAGCGGAAGA-3′; #2: 5′-GAGAGGATGTCATGCGGCT-3′ . The following sequences were targeted for murine Rnf26 cDNA: #1: 5′-GAGCGGAAGAAGTGTGTTA-3′; #2: 5′-GATCAACAGTCTAGTCAAC-3′ . Total RNA was isolated from cells using TRIzol reagent ( Takara ) and subjected to real-time PCR analysis to measure expression of mRNA . Gene-specific primer sequences were as described [27] or as follow: Human RNF26 ( Forward: 5′-CAGGACCATCAGAGTGACACCT-3′; Reverse: 5′-GCAACACTGTCTTGCTCTGGTC-3′ ) . Murine Rnf26 ( Forward: 5′-TGGCTGCTTTCCTCGCTCACAT-3′; Reverse: 5′-GCAACACCAATCCAGTGAGATGG-3′ ) . Human IRF3 ( Forward: 5′-TCTGCCCTCAACCGCAAAGAAG-3′; Reverse: 5′-TACTGCCTCCACCATTGGTGTC-3′ ) . Murine Irf3 ( Forward: 5′-CGGAAAGAAGTGTTGCGGTTAGC-3′; Reverse: 5′-CAGGCTGCTTTTGCCATTGGTG-3′ ) . The supernatants of cell culture medium were analyzed with a human IFN-β ( PBL ) , human TNFα or human IL-6 ( Boster ) ELISA kit following protocols recommended by the manufacturers . The 293 cells were transfected with two packaging plasmids ( pGAG-Pol and pVSV-G ) together with pMSCV-GFP-Flag-Ub-K11O retroviral plasmids by calcium phosphate precipitation . Twenty-four hours after transfection , cells were incubated with new medium without antibiotics for another twenty-four hours . The recombinant virus-containing medium was filtered with 0 . 22 µm filter ( Millex ) and then added into cultured THP-1 cells in the presence of polybrene ( 4 µg/mL ) . The infected cells were cultured for at least seven days and sorted by a flow cell sorter before additional experiments were performed . The 293 cells were transfected with pRK7-Neo-HA-Ub-K11O plasmid . Twenty-four hours after transfection , cells were selected with G418 ( 0 . 8 µg/mL ) . Single cell colonies were picked and identified by immunoblot analysis . The 293 cells were transfected with two packaging plasmids ( pGAG-Pol and pVSV-G ) together with a control , RNF26-RNAi or Rnf26-RNAi retroviral plasmids respectively by calcium phosphate precipitation . Twenty-four hours after transfection , cells were incubated with new medium without antibiotics for another twenty-four hours . The recombinant virus-containing medium was filtered with 0 . 22 µm filter ( Millex ) and then added into cultured 293 , Raw264 . 7 or THP-1 cells in the presence of polybrene ( 4 µg/mL ) . The infected cells were selected with puromycin ( 1 µg/mL for 293 and Raw264 . 7 cells or 0 . 5 µg/mL for THP-1 cells ) for at least seven days before additional experiments were performed . UniProtKB/Swiss-Prot accession numbers ( parentheses ) are indicated for proteins mentioned in text: RNF26 ( Q9BY78 ) , MITA ( Q86WV6 ) , RNF5 ( Q99942 ) , VISA ( Q7Z434 ) . | Virus infection induces the host cells to produce type I interferons , which are secreted proteins important for the host to clear viruses . Previously , we identified a cellular protein called MITA , which is essential for virus-triggered induction of interferons . In this study , we found an enzyme called RNF26 could covalently modify MITA with one type of polypeptide , called polyubiquitin . This modification caused increased stability of MITA after viral infection . RNF26 also caused disability of IRF3 , another important component required for virus-triggered interferon induction . Thus , RNF26 could temporally regulate virus-triggered interferon induction by two distinct mechanisms . This discovery helps to understand how the antiviral response is delicately regulated . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"disease",
"immunology",
"medicine",
"and",
"health",
"sciences",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"immunology"
] | 2014 | RNF26 Temporally Regulates Virus-Triggered Type I Interferon Induction by Two Distinct Mechanisms |
The establishment of apical-basolateral polarity is important for both normal development and disease , for example , during tumorigenesis and metastasis . During this process , polarity complexes are targeted to the apical surface by a RAB11A-dependent mechanism . Huntingtin ( HTT ) , the protein that is mutated in Huntington disease , acts as a scaffold for molecular motors and promotes microtubule-based dynamics . Here , we investigated the role of HTT in apical polarity during the morphogenesis of the mouse mammary epithelium . We found that the depletion of HTT from luminal cells in vivo alters mouse ductal morphogenesis and lumen formation . HTT is required for the apical localization of PAR3-aPKC during epithelial morphogenesis in virgin , pregnant , and lactating mice . We show that HTT forms a complex with PAR3 , aPKC , and RAB11A and ensures the microtubule-dependent apical vesicular translocation of PAR3-aPKC through RAB11A . We thus propose that HTT regulates polarized vesicular transport , lumen formation and mammary epithelial morphogenesis .
Epithelial cells in glandular and tubular epithelial systems are organized as one layer surrounding a lumen . The establishment of apical-basolateral polarity in these systems is characterized by the formation of cell–cell adherens and tight junctions and accompanies lumen formation ( reviewed in [1–3] ) . This organization provides a functional barrier that regulates the polarized secretion and intake of molecules . Cell polarity complexes , which were originally identified in model organisms such as yeast , worms , and flies , are highly evolutionarily conserved [4] . Three major polarity complexes have been identified . The PAR polarity complex , which includes Partitioning Defective 3 and 6 ( PAR3 and PAR6 ) , atypical protein kinase C ( aPKC ) and cell division control protein 42 ( CDC42 ) proteins , promotes the establishment of the apical-basal membrane border . The Crumbs ( CRB ) complex , which is required to establish the apical membrane , is composed of the transmembrane protein CRB and the associated cytoplasmic proteins , PALS1 ( also known as MPP5 ) and PALS1-associated tight junction protein ( PATJ; also known as INADL ) . Finally , the Scribble complex , which is composed of scribble homolog ( SCRIB ) , lethal giant larvae homolog ( LGL; also known as LLGL ) , and disc-large homolog ( DLG ) proteins , defines the basolateral plasma domain . In Drosophila , these complexes interact and establish the apical and basolateral surfaces of epithelial cells by a system of mutual exclusion [5 , 6] . The PAR complex is a master regulator of polarity and is involved in polarity and spatial organization in almost all metazoan cells [7] . Mammalian PAR3 is localized to tight junctions at the apical/lateral boundary [8] , and functions in their assembly [9] , whereas PAR6 and aPKC maintain the integrity of the apical domain [10] . All of these proteins interact directly with each other . PAR6 acts as a targeting subunit for aPKC , and it recruits the CRB complex [11 , 12] and LGL as substrates [13] . The binding of PAR3 to PAR6 , which forms a complex with aPKC , is required for the delivery of aPKC to the apical surface [14 , 15] . Moreover , the interaction of PAR3 with aPKC is essential for the restricted localization of these proteins to the apical region [15] . In the mammary gland , this interaction is essential for the regulation of progenitor differentiation and epithelial morphogenesis [15] . Formation of the apical surface , the first step of lumen morphogenesis , involves the coordination of the trafficking machinery and the polarity complexes . In mammalian cells , vesicles containing apical membrane components are delivered to a region named the apical membrane initiation site ( AMIS ) where the lumen begins [1 , 16] . This region is delineated by PAR3 , aPKC , and the exocyst subunit SEC8 . In polarized cells , trafficking from recycling endosomes is regulated by several members of the family of RAB GTPases . In particular , RAB11 controls vesicle trafficking in apical recycling endosomes and is necessary for epithelial morphogenesis [17 , 18] . Similarly , during lumen formation , the trafficking of vesicles containing apical membrane components depends on RAB11 [16] . The targeting of apical vesicles containing podocalyxin ( PCX ) to the AMIS is regulated by RAB11A together with RAB8 and RABIN8 , a RAB8-specific GEF that is activated by RAB11A [16] . The PAR complex targets SEC8-SEC10 to the AMIS , and recruits SEC15A-RAB8A-RAB11A vesicles to generate the pre-apical patch ( PAP ) [16] . This mechanism leads to the localization of CDC42 to the apical membrane , where it activates the PAR complex . Although the core complexes involved in these mutually interdependent processes are well characterized , regulatory factors that couple polarity proteins to the membrane transport machinery have not been identified . Huntingtin ( HTT ) , the protein mutated in Huntington disease , acts as a molecular scaffold and promotes intracellular dynamics . HTT associates with vesicles and microtubules . It is crucial for vesicular trafficking and affects axonal transport and endocytosis . HTT binds dynein and HAP1 directly [19] , and kinesin [20] and the dynactin subunit p150Glued [21] indirectly . HTT facilitates the transport of several cargoes along microtubules [22–24] . HTT also mediates vesicle recycling during endocytosis by activating RAB11 [25] . These functions have consequences for a wide variety of cellular events mostly described in the nervous system during both development and the maintenance of homeostasis in adults . For instance , through its function as a regulator of microtubule-based dynamics , HTT influences the division of progenitors at the ventricular zone during cortical development [26] , the maturation of newly generated neurons during adult hippocampal neurogenesis [27] and ciliogenesis in ependymal cells [28] . However , HTT expression is ubiquitous , and this raises questions concerning the functions of HTT in tissues outside the central nervous system . We previously showed that HTT is detectable in healthy mammary tissue and mammary tumors where it regulates tumor progression [29] . HTT is required in mammary basal progenitors for appropriate spindle orientation and for the determination of cell fate [30] . Here , we focused on the function of HTT in the establishment of apical polarity during the morphogenesis of the mouse mammary epithelium . We propose that HTT regulates apical vesicular transport , which enables the proper targeting of polarity proteins and the correct establishment of subsequent luminogenesis .
We recently showed that depletion of HTT from the basal compartment in the mammary gland results in altered morphological and functional differentiation [30] . However , the abundance of HTT is higher in luminal cells ( LCs ) than in basal cells ( BCs ) ( Fig 1A ) [30] . We sought to address whether HTT expression specifically in LCs is essential for epithelial morphogenesis; therefore , we deleted HTT from the luminal cell layer of the mammary epithelium by crossing Httflox/flox mice harboring floxed Htt alleles [31] with transgenic mice expressing Cre recombinase under the control of the mouse mammary tumor virus ( MMTV ) promoter [32] . Cre expression was mostly confined to the luminal cell population ( Fig 1A ) . The abundance of Htt transcripts was 72% lower in LCs from MMTVCre;Httflox/flox ( mutant ) epithelium than in those from control epithelium , whereas mammary Htt transcript levels were similar in control and mutant BCs . Thus , HTT is specifically depleted in luminal cells in MMTVCre;Httflox/flox mice . We then performed whole mount staining with fourth abdominal mammary glands isolated from mutant and control mice at the age of 5 , 6 , and 8 wk to measure ductal elongation and bifurcation . The direct visualization of ductal trees showed that ductal elongation and bifurcation were less extensive in mutant mice than in control mice ( Fig 1B ) . We quantified these effects by measuring the percentage of the fat-pad area covered by the ductal structures and the number of branches; both were significantly lower in mutant mice than in control mice at all stages analyzed ( Fig 1C and 1D ) . Interestingly , the number of terminal end buds ( TEBs ) in 6- and 8-wk-old glands ( Fig 1E ) was significantly higher in mutant mice than in control mice . At 12 wk , which marks the end of puberty in mice , ductal extension and branching were similar between mutant and control mice , and the effect of HTT deletion disappeared ( S1A and S1B Fig ) . These findings suggest that loss of HTT delays ductal elongation and bifurcation in the mammary tree during puberty . We performed hematoxylin and eosin staining on serial sections of mammary glands from 6- and 8-wk-old control and mutant mice ( Fig 1F ) . Although control TEBs showed a well-defined lumen at 6 wk , the structures from mutant mice were partially filled with cells . At 8 wk , control ducts were completely hollow , whereas mutant ducts displayed an aberrant architecture and contained many intraluminal cells ( Fig 1F ) . We hypothesized that these defects were linked to alterations in cell death and proliferation . Thus , we stained sections from control and mutant mammary ducts for cleaved caspase-3 to analyze apoptosis ( Fig 1G ) . Globally there were a high number of intraluminal cells in mutant TEBs; however , within this population , the proportion of apoptotic cells was lower in mutant TEBs than control TEBs at 6 wk ( 1 . 84% ± 0 . 05% in mutant versus 3 . 99% ± 0 . 1% in control mice; Fig 1G and 1H ) . In contrast with controls , ducts from 8-wk-old mutant mice still displayed apoptotic cells and a high number of intraluminal cells ( 15 . 2% ± 1 . 36% in mutant versus 1 . 4% ± 0 . 57% in control mice; Fig 1G and 1H ) . Furthermore , KI67 immunostaining showed that the percentage of proliferating cells was higher in ducts from 6-wk-old mutant mice than in those from control mice of the same age ( 20 . 4% ± 2% in mutant versus 8 . 6% ± 0 . 57% in control mice; Fig 1I and 1J ) . At 8 wk , the percentage of proliferating cells was similar in control and mutant mice ( S1C Fig ) . These in vivo data suggest that HTT may result in delayed apoptotic-mediated clearing of intraluminal cells . To confirm this hypothesis , we used the human MCF-10A cells , which form acini in 3-D culture by 20 d of morphogenesis by luminal cells clearing through apoptosis-mediated anoikis [33] . HTT deletion using specific shRNA blocked apoptosis-mediated luminal clearing , resulting in malformed acini filled with cells ( S2A–S2E Fig ) . The acini formed when HTT levels were lowered were significantly larger than in control condition ( S2B , S2D and S2F Fig ) . Thus , the loss of HTT alters ductal morphogenesis and results in delayed intraluminal cell death and a malformed lumen . We also investigated how the loss of HTT in luminal cells affected the differentiation of the mammary gland at day 18 . 5 of pregnancy and day 1 of lactation . Both the number of secretory alveoli and the percentage of epithelial cells were lower in mutant glands than in control glands ( Fig 2A and 2B ) . On day 18 . 5 of pregnancy and day 1 of lactation , there were fewer well-developed alveoli in mutant glands than in control glands . In controls , the large cytoplasmic lipid droplets in luminal alveolar cells on day 18 . 5 of pregnancy were replaced with small lipid droplets at the luminal surface on day 1 of lactation ( Fig 2A ) . In mutant mammary glands , the large cytoplasmic droplets remained in the alveolar cells on day 1 of lactation . We investigated the functional consequences of these epithelial defects by analyzing the subcellular location of signal transducer and activator of transcription 5A ( STAT5A ) on day 1 of lactation ( Fig 2C and 2D ) . The abundance of phosphorylated STAT5A in the nucleus ( the active form of STAT5A ) was lower in mutant alveolar cells than in control glands . The abundance of transcripts encoding the transcription factor ELF5 ( Elf5 ) , which is crucial for lobuloalveolar morphogenesis [34] , was significantly lower in mutant alveoli than in control alveoli ( Fig 2E ) . Consistent with these observations , immunolabeling showed that the abundance of the milk whey acid protein ( WAP ) was lower in mutant glands than in control glands , and the RT-PCR revealed that the same was true for RNAs encoding the milk proteins β-casein ( Csn2 ) and WAP ( Wap ) ( Fig 2F ) . Ultimately , mutant mice failed to nurse their pups , which displayed severe weight defects ( Fig 2G ) . Overall , these findings show that the loss of HTT in LCs alters lumen formation , ductal morphogenesis , and tissue architecture at different stages of mammary gland development and has functional consequences during lactation . We then analyzed how HTT deficiency in LCs affects epithelial cell polarity . We compared the localization of PAR3 and aPKC in LCs from 12-wk-old virgin mutant and control mice ( Fig 3A ) . In control glands , PAR3 and aPKC were localized at tight junctions and the apical surface of LCs , whereas in mutant ducts , PAR3 and aPKC labeling was more diffuse , and both proteins accumulated in the cytoplasm . We also examined the localization of E-cadherin , a marker of adherens junctions ( Fig 3A ) . As expected , E-cadherin was enriched at the lateral compartment in control LCs . By contrast , in mutant LCs , it accumulated abnormally with PAR3-aPKC at the apical surface and was also dispersed in the cytoplasm . This was associated with defects in epithelial architecture and lumen malformation . Apical localization of PAR3-aPKC was also altered when mutant epithelia formed lumens ( Fig 3A; arrows ) . We also determined the distribution of PAR3 , aPKC , and E-cadherin in LCs of control and mutant epithelia on day 18 . 5 of gestation and day 1 of lactation ( Figs 3A and S3A ) . Consistent with the morphological defects observed at all stages ( Figs 1 and 2 ) , the defects caused by the loss of HTT in pregnant and lactating mice were similar to those seen in virgin mice and included the mislocalization of PAR3 , aPKC , and E-cadherin and lumen malformation ( see asterisks ) . We also analyzed the Golgi distribution by immunostaining using an antibody directed against the Golgi matrix protein GM130 ( Figs 3B and S3B ) . In control epithelia , the Golgi apparatus localization was polarized in an apical position facing the lumen in most of LCs of control epithelia at all stages analyzed ( Figs 3B , 3C , S3B and S3C ) . In the majority of mutant LCs , however , we found that the Golgi apparatus was dispersed within the soma and did not show a characteristic polarized distribution . We confirmed these observations in 3-D cultures of MCF-10A ( S3D–S3F Fig ) . While the Golgi apparatus was dispersed in the absence of HTT , it still displayed a perinuclear distribution . In agreement , the microtubule network , which maintains the Golgi apparatus in the perinuclear area [35] , was comparable in control and shHTT-treated MCF-10A cells ( S4 Fig ) . Thus , the absence of HTT in luminal cells alters their polarization . We then asked whether HTT directly regulates the polarity complex . We determined whether HTT colocalizes with PAR3 and aPKC in mammary glands from 12-wk-old control mice ( Fig 3D ) . HTT colocalized with PAR3 and aPKC at the apical surface of LCs . In particular , HTT was enriched at tight junctions . Furthermore , PAR3 and aPKC coimmunoprecipitated with HTT in extracts of mammary epithelial MCF-10A cells ( Fig 3E ) . Consistent with these data , affinity-purification mass spectrometry previously showed that PAR3 and aPKC form a complex with HTT in cortical neurons [36] . Although PAR6 has not been reported to interact with HTT , it belongs to the PAR polarity complex [7] and also coimmunoprecipitated with HTT , PAR3 , and aPKC . Thus , HTT is associated with components of the PAR polarity complex and may regulate epithelial polarity through this interaction . We then investigated the mechanisms by which HTT regulates apical polarity during epithelial morphogenesis . MCF-10A and primary mammary epithelial cells are useful to assess several aspects of mammary epithelial morphogenesis in 3-D culture [33] ( S2 and S3D–S3F Figs ) . For instance , the localization of polarity markers such as GM130 can be assessed in MCF-10A cysts with already-formed lumen ( S3D–S3F Fig ) . However , MCF-10A and primary cells form lumen by apoptotic hollowing rather than initially setting up apical polarity; they form non-polarized early cell aggregates after plating , making them unsuitable to study the early steps of polarity establishment during epithelial morphogenesis [33 , 37] . We therefore used MDCK ( Madin-Darby canine kidney ) cells , which are widely used to model epithelial polarization in several tissues [16 , 38] . In 3-D culture , individual MDCK cells proliferate and assemble into cyst structures , to form a polarized spherical monolayer surrounding a central lumen [39] . After 24 h of plating , MDCK cells are an ideal system to directly visualize the process of polarity establishment . At this two-cell stage , cells undergo polarity inversion , which leads to the separation of the apical cortex from the lateral cortex [16] . Consistent with our in vivo observation ( Fig 3D ) , HTT was localized predominantly at the apical cell cortex and tight junctions ( Fig 4A ) . It was also present in the cytoplasm , where it was enriched in cytoplasmic vesicular-like structures that colocalized with PAR3 and aPKC ( Figs 4A , arrowheads , 4B , S5A and S5B , asterisks ) . We examined the extent to which HTT influences the apical translocation of PAR3 and aPKC during the first steps of lumen formation . We used lentiviral short hairpin RNAs ( shRNA ) to stably knock down HTT expression in MDCK cells . HTT expression was efficiently impaired with two lentiviruses ( shHTT1 and shHTT2 ) expressing shRNAs targeting different sequences of canine HTT ( Fig 4C ) . As expected [16] , we found that PAR3 and aPKC in control cysts were enriched at the apical surface , whereas the adherens junction marker β-catenin was restricted to the lateral cortex ( Fig 4D ) . The depletion of HTT impaired the cortical accumulation of PAR3-aPKC , which displayed diffuse cytoplasmic localization ( Fig 4D–4F ) . This impaired the transition from unpolarized epithelial cell aggregates to the establishment of the luminal PAP , where apical and basolateral plasma membranes are separated . Moreover , HTT-depleted cells displayed aberrant β-catenin localization , indicating altered specification of the basolateral cortex . Next , we introduced a construct encoding a full-length HTT tagged with mCherry ( HTTFL; Fig 4C ) [40] . The shHTT2 construct was designed to inhibit the expression of endogenous HTT but had no effect on the expression of exogenous HTTFL ( Fig 4C ) [40] . The expression of HTTFL restored the apical translocation of PAR3-aPKC and the lateral localization of β-catenin was similar to that observed in cells expressing endogenous HTT ( Fig 4D–4F ) . Thus , HTT is instrumental for apical localization of PAR3-aPKC during the first step of polarity establishment . We sought to investigate how HTT-mediated apical localization of PAR3-aPKC affects cystogenesis in MDCK cells . On day 4 of 3-D culture , most control cysts contained well-polarized cells that were organized around a central , single lumen ( 75% ± 2 . 74% of cysts; Fig 4G–4I ) . PAR3 and aPKC were localized at the apical cortex and at tight junctions , and E-cadherin was restricted to the lateral compartment ( Fig 4G ) . Only 31 . 5% ± 1 . 3% of cells expressing shHTT1 and 28 . 12% ± 3 . 41% of cells expressing shHTT2 formed normal structures , and most cysts contained several lumens and were significantly bigger than control cysts ( Fig 4G–4J ) . In HTT-depleted cysts , PAR3 and aPKC showed altered apical localization and abnormal colocalization with E-cadherin ( Fig 4G ) . Remarkably , the ectopic expression of HTT in HTT-depleted cysts restored normal cystogenesis and led to the apical accumulation of PAR3 and aPKC and the lateral localization of E-cadherin ( Fig 4G ) . By contrast , the expression of green fluorescent protein ( GFP ) -tagged PAR3 in shHTT2-expressing cells was not sufficient to rescue cystogenesis ( Fig 4H–4J ) . In this context , both PAR3-GFP and aPKC showed diffuse staining in the cytoplasm ( colocalization in white; Fig 4H ) . These observations suggest that HTT may act upstream from PAR3 to ensure the apical accumulation of PAR3-aPKC and proper cystogenesis . Apical vesicle trafficking during lumen morphogenesis depends on microtubule transport driven by motor proteins ( reviewed in [1–3] ) . HTT interacts with microtubule-based motors to promote vesicular transport in neurons [20 , 23 , 24 , 41] . We thus analyzed the role of HTT in the dynamics of apical vesicles ( Fig 5 ) . We performed live-cell imaging in 3-D culture with the lipophilic dye FM4-64 [39] . In control cysts , the basolateral membrane was labeled 30 min after the addition of the dye ( Fig 5A; S1 Movie ) . Two hours post–dye addition , both the apical membrane and the intracellular endocytic vesicles ( which accumulate underneath the apical surface ) were labeled ( Fig 5A; S1 Movie; see also magnification in Fig 5D ) . By contrast , in HTT-depleted cysts treated with FM4-64 , 2 h post-dye addition , endocytic vesicles failed to reach the apical membrane , accumulated in the cytoplasm , and the apical membrane was not labeled ( Fig 5A; S2 and S3 Movies ) . The ectopic expression of HTT in shHTT1/2-expressing cysts restored normal apical vesicular trafficking and cystogenesis ( Fig 5A; S4 Movie ) . However , the ectopic expression of PAR3 failed to do so ( Fig 5B; S5 and S6 Movies ) , reinforcing the hypothesis that HTT is upstream from the apical vesicular trafficking machinery . The trafficking defect observed in absence of HTT correlated with aberrant cystogenesis , suggesting that HTT could mediate its effect on cystogenesis , at least in part , by regulating apical vesicular trafficking . We then confirmed that cystogenesis was dependent on the integrity of the microtubule network and on molecular motors . We treated cysts with 10 μM of nocodazole for 90 min prior to the analysis of the trafficking of FM4-64-containing apical vesicles . Nocodazole treatment altered apical vesicle dynamics , and the vesicles accumulated in the cytoplasm ( Fig 5C and 5D; S7 and S8 Movies ) . Moreover , treatment with 5 μM nocodazole for 16 h impaired the apical accumulation of PAR3-aPKC and led to defects in cystogenesis ( Fig 5E and 5F ) . HTT interacts with the microtubule motor kinesin 1 to promote anterograde vesicular trafficking in neurons [24] . HTT also interacts with kinesin 1 to deliver the dynein/dynactin/NUMA/LGN complex along astral microtubules to the cell cortex during mitotic spindle orientation in mammary cells [30] . We asked whether HTT and kinesin 1 could act together during apical trafficking . Consistent with this idea , HTT and kinesin 1 colocalized and showed a punctate staining in control 24 h and day 4 3-D cultures ( Figs 5G–5I and S5C; asterisks ) . HTT depletion disrupted kinesin 1 localization ( Fig 5G–5I ) . Furthermore , kinesin 1 participated in apical trafficking: kinesin 1 depletion impaired the trafficking of FM4-64-containing apical vesicles , which correlated with defective cystogenesis ( Fig 5J and 5K; S9 and S10 Movies ) . Interestingly , the trafficking defects observed in the presence of nocodazole and in absence of HTT or kinesin 1 were associated to similar aberrant cystogenesis . Overall , these results show that HTT regulates apical vesicular trafficking ( Fig 5L ) . Our data also support the hypothesis that this may occur through a microtubule-based , kinesin 1-dependent process . HTT binds RAB11A and regulates its activity in neurons [25] . RAB11 participates in lumen formation in mammalian 3-D cultures [16 , 42] . We hypothesized that HTT regulates apical vesicle trafficking through a RAB11A-dependent mechanism . RAB11A coimmunoprecipitated with HTT-PAR3-PAR6-aPKC ( Fig 3E ) and HTT localized with RAB11A at the apical membrane of mammary epithelial cells in vivo ( Fig 6A ) . Similarly , in 24 h and 4 d 3-D cultures , HTT localized with RAB11A , showing a punctate staining which was consistent with localization on vesicles and accumulated at the apical membrane ( Figs 6C and S5D; asterisks ) . The apical accumulation of RAB11A was impaired when HTT was depleted in LCs in vivo ( Fig 6B ) or in 3-D cultures of MDCK cells ( Fig 6D ) . In shHTT2-expressing cysts , HTTFL was sufficient to restore the apical accumulation of RAB11A ( Fig 6D ) . Thus , HTT and RAB11A both localize at the apical membrane , and HTT is required for the apical localization of RAB11A . We then expressed different variants of GFP-tagged RAB11A in MDCK cells and analyzed the apical targeting of PAR3 and the subsequent effects on cystogenesis . In control cysts at 24 h , wild-type RAB11A ( RAB11AWT ) and the constitutively active RAB11AQ70L were localized , along with endogenous PAR3 , at the apical surface ( Fig 6E and 6F ) . By contrast , the dominant-negative RAB11AS22N accumulated in the cytoplasm and impaired the apical accumulation of PAR3 . Cystogenesis was altered at day 4 in cysts expressing RAB11AS22N , whereas the expression of RAB11AWT or RAB11AQ70L mostly resulted in cysts with a single lumen ( Fig 6E and 6G ) . We next analyzed whether the expression of the RAB11A variants rescues the defects in the apical targeting of PAR3 and cystogenesis induced by the loss of HTT . Remarkably , in contrast with RAB11AWT and RAB11AS22N , RAB11AQ70L expression in shHTT2-treated cysts was sufficient to rescue the apical translocation of PAR3 and cystogenesis ( Fig 6E and 6H ) . We obtained similar results with aPKC ( S6 Fig ) . We conclude that HTT regulates RAB11A to coordinate the apical vesicular trafficking of PAR3-aPKC . We then analyzed apical trafficking by live cell imaging of FM4-64-containing vesicles . The accumulation of FM4-64-containing vesicles at the apical surface was higher in control cysts expressing RAB11AQ70L than in those expressing exogenous RAB11AWT ( Fig 6I; S11 and S12 Movies ) . RAB11AS22N expression altered apical vesicle trafficking , which correlated with marked defects in cystogenesis ( Fig 6I; S13 Movie ) . In HTT-depleted cysts , RAB11AQ70L was able to recover FM4-64-apical vesicle trafficking and normal cystogenesis , whereas both RAB11AWT and RAB11AS22N failed to do so ( Fig 6I; S14–S16 Movies ) . These observations show that HTT is instrumental for RAB11A-mediated apical vesicular trafficking .
In this study , we propose a model in which HTT regulates RAB11A-mediated apical trafficking of the PAR-polarity complex in the mammary epithelium , with consequences for lumen formation and tissue architecture ( Fig 7 ) . Interestingly , loss of any of the components of the CDC42-PAR6-PAR3-aPKC complex also causes the formation of multiple lumens and thereby alters epithelial morphogenesis [10 , 43] . Disruption of the interaction between PAR3 and aPKC in the mammary gland induces malformations during mammary gland morphogenesis [15] . Remarkably , the epithelial architectural defects induced by the loss of HTT persisted during pregnancy and lactation and affected functional differentiation and milk production . Consistent with these findings , the expression of apical polarity proteins is essential for the differentiation of alveolar cells to milk secreting units [44] . We recently showed that the depletion of HTT from the basal compartment of the mammary gland alters luminal cell polarity [30] . In the K5Cre; Httflox/flox mouse model used in this study , HTT was depleted from basal cells but also partially from LCs . Thus , we were unable to conclude whether the effect of HTT on luminal polarity was direct or indirect . Here , we specifically removed HTT from LCs because HTT is strongly expressed in these cells and LCs are highly polarized . We show that HTT is important for the establishment of apical polarity during mammary morphogenesis . We provide evidence that one of the mechanisms by which HTT mediates its effect is the regulation of the apical trafficking of PAR3-aPKC . However , we cannot exclude that loss of HTT may lead to altered cell organization by another mechanism that would subsequently lead to a polarization defect . In particular , how HTT-dependent vesicular trafficking coordinates the segregation between apical and basolateral compartments remains to be determined . Early work in Drosophila melanogaster identified a Rab11-dependent trafficking of E-cadherin essential for epithelial junction maturation [45] . Furthermore , HTT forms a complex with β-catenin [46] . It is then tempting to speculate that HTT may also regulate basolateral trafficking through RAB11A during polarity establishment . The orientation of mitosis also regulates lumen formation; therefore , alteration in this process may also contribute to the phenotypes observed . Indeed , HTT regulates spindle orientation in MaSCs and controls the cortical accumulation of the mitotic complex , including LGN , NUMA , dynein , and dynactin [30] . Interestingly , RAB11A , PAR3 , and aPKC are also involved in spindle orientation [47 , 48] . Thus HTT could help localize RAB11A , PAR3 , and aPKC during lumen formation and mammary epithelium morphogenesis to ensure the coordination of spindle orientation and apical trafficking . RAB proteins cycle between GDP bound ( inactive ) and GTP bound ( active ) states and these cycles are controlled by guanine nucleotide exchange factors ( GEFs ) and GTPase-activating proteins ( GAPs ) . In their active form , RABs are associated with membranes and carry out their functions though effector partner proteins . RAB11 controls vesicle trafficking in apical recycling endosomes and is necessary for epithelial morphogenesis [17 , 18] . Our results suggest that HTT acts upstream from PAR3 by regulating RAB11 activity . These results are consistent with a previous study showing that HTT binds RAB11A and regulates its activity in neurons [25] . The authors of this study showed that the inhibition of HTT expression affects the attachment of RAB11 to membranes and the guanine nucleotide exchange activity on RAB11 . They also showed that HTT binds RAB11-GDP preferentially , suggesting that HTT either acts as a GEF for RAB11 or activates GEF activity on RAB11 . Nonetheless , other mechanisms besides the microtubule-based apical delivery of polarity proteins may be affected by the HTT-mediated regulation of RAB11 activity . Indeed , a recent study demonstrated that RAB11 localizes recycling endosomes to mitotic spindle poles by dynein-mediated transport [48] . Similarly , during mitosis , the interaction of HTT with dynein is required for the localization of spindle pole proteins [26 , 30] . The actin and the microtubule cytoskeletons and their associated motor proteins are critical for apical vesicle trafficking during lumen morphogenesis ( reviewed in [2 , 3] ) . Interestingly , previous studies suggest that HTT is a crucial link between the microtubule and the actin cytoskeletons . HTT forms a complex with dynein , dynactin , and kinesin 1 ( KIF5 ) in neurons to promote retrograde and anterograde microtubule-based axonal transport of several cargoes [19–24] . RAB11-containing vesicles are bidirectionally transported by HTT in vivo in whole-mount Drosophila larval axons [49] . During mitosis , HTT mediates the cortical localization of dynein , dynactin , LGN , and NUMA through kinesin 1-dependent transport along astral microtubules [30] . Here , we suggest that HTT acts with kinesin 1 to coordinate microtubule-based apical trafficking in a RAB11A-dependent pathway . The early endosomal trafficking effector , RAB5 , binds HTT through HAP40 , and RAB8 , which associates with the Golgi membrane , can also form a complex with HTT through the myosin VI linker , optineurin [50 , 51] . The HAP40-HTT complex also interacts with optineurin [52] . Thus , HTT may regulate actin-dependent dynamics when in complex with HAP40-Optineurin-MyosinVI , and it may regulate microtubule-dependent transport when in complex with dynein-dynactin-kinesin . Finally , the cell polarity machinery is perturbed during tumorigenesis with consequences for metastasis . For instance , PAR3 levels are significantly lower in human breast cancers than in non-malignant tissue , and this down-regulation correlates with the overactivation and mislocalization of aPKC [53 , 54] . In murine models of breast cancer , loss of PAR3 promotes breast tumorigenesis and metastasis [54] . Thus , the identification of new regulators of the apical vesicle trafficking machinery is critical for our understanding of both normal development of the epithelium and pathogenic pathways leading to metastasis .
pARIS-mCherry-HTTQ23 ( referred to herein as Q23HTTFL ) was previously described [40] . GFP-RAB11A wild-type ( WT ) , dominant-negative ( S25N ) , and constitutively active ( Q70L ) ( referred to herein as RAB11AWT , RAB11AS25N and RAB11AQ70L , respectively ) were obtained from B . Goud ( Institut Curie , France ) . PAR3-GFP ( referred to herein as PAR3 ) was provided by Dr . I . Mellman ( Genentech , CA , United States ) [55] . si-kinesin 1-sens ( 5′-GCAGUCAGGUCAAAGAAUA-3′ ) and si-kinesin 1-antisens ( 5′-UAUUCUUUGACCUGACUGC-3′ ) were used for siRNA against mouse/rat/human KIF5B ( si-kinesin 1 ) . siRNA negative control ( si-Control ) from Eurogentec ( OR-0030-neg05 ) was used . MCF-10A , a spontaneously immortalized , nontransformed human mammary epithelial cell line derived from the breast tissue [56] was maintained in DMEM/F12 ( Invitrogen , Carlsbad , CA ) supplemented with 5% donor horse serum , 20 ng/ml EGF ( Peprotech , Rocky Hill , NJ ) , 10 μg/ml insulin ( Sigma , St Louis , MO ) , 1 ng/ml cholera toxin ( Sigma ) , 100 μg/ml hydrocortisone ( Sigma ) , 50 U/ml penicillin , and 50 μg/ml streptomycin ( Invitrogen ) at 37°C in a humidified 5% CO2 atmosphere . MDCK cells were maintained in DMEM ( Invitrogen ) supplemented with 10% fetal calf serum , 50 U/ml penicillin and 50 μg/ml streptomycin ( Invitrogen ) at 37°C in a humidified 5% CO2 atmosphere . Cells were spread in 10 cm2 plate and transfected using Lipofectamin 2000 ( Invitrogen ) . After 24 h , cells were plated on Matrigel for 3-D cultures . Alternatively , after 48–72 h , cells were lysed or fixed and immunoprocessed . Three-dimensional cultures of MCF-10A and MDCK cells in Matrigel were performed as described previously [33] . In brief , MCF-10A and MDCK cells were trypsinized and resuspended to single cell suspension of 2 x 104 cells/ml ( MCF-10A ) and 4 x 104 cells/ml ( MDCK ) in 2% Matrigel ( BD ) . Four-hundred μl of cells were plated in each well of 8-well Lab-Tek II chamber slides ( Thermo Fisher Scientific ) precovered with matrigel ( 25 μl per well ) . MCF-10A cells were fed every 4 d and grown for 8–20 d . MDCK cells were fed every 2 d and grown for 1–4 d . Stable knockdown of HTT in MDCK cells was done as previously described for LGN [57] . Oligos containing target sequences were cloned in the pLKO . 1 vector . HEK293 cells were transfected with the RNAi vectors and the lenti-packaging mix ( Invitrogen ) . Virus supernatant was collected 48 h after transfection and used to infect MDCK cells ( plated in 12-well plates and transferred to P-100 plates 24 h after infection ) . Clones of interest were selected using puromycin ( 5 μg/ml ) and isolated 1 wk later . Target sequences for dog HTT were 5′-GTGCCTCAACAGAGTCATAA-3′ ( shHTT1 ) and 5′-GGTTACAGTTAGAACTCTATA-3′ ( shHTT2 ) . Empty pLKO . 1 was used as a control . Lentivirus-mediated stable knockdown of HTT in MCF-10A cells was described elsewhere [58] . Briefly , shRNA targeting the human HTT recognized a region within exons 8–9 and was transcribed from the polymerase III H1 promoter 5′-AGCTTTGATGGATTCTAA-3′ ( sh-HTT ) . The sh-Control recognized a sequence within the firefly luciferase gene 5′-CGTACGCGGAATACTTCGA-3′ . EGFP reporter gene under the control of the mouse PGK promoter allowed the selection of positive clones . The knockdown efficiency was analyzed by immunoblotting and immunostaining of HTT . Drugs were dissolved in DMSO and kept at -20°C as 10 mM stock solutions . To depolymerize microtubules , MDCK cells were treated with 10 μM or 5 μM nocodazole ( Sigma ) for 90 min or 16 h respectively . For live-imaging , MDCK cells were treated with 4 μM FM4-64 Dye ( N- ( 3-Triethylammoniumpropyl ) -4- ( 6- ( 4- ( Diethylamino ) Phenyl ) Hexatrienyl ) Pyridinium Dibromide ) ( Life Technology ) for 30 min . Anti-HTT antibodies used in this study were previously described: mAb 4C8 ( epitope 445–456 , clone HU-4C8-As , Euromedex ) , mAb D7F7 ( Cell Signaling ) [26] . For immunofluorescence , the primary monoclonal antibodies used were: anti-ß-catenin ( 1:200; BD Bioscience ) , anti-GM130 ( 1:100; BD Bioscience ) and anti-HTT 4C8 . The primary polyclonal antibodies used were: anti-PAR3 ( 1:200; Chemicon ) , anti-aPKC ( 1:200; Santa-Cruz Biothechnology ) , anti-RAB11A AT15 ( 1:200; Abcam ) and anti-cleaved caspase 3 ( 1:100; Cell Signaling ) . Rhodamin-conjugated Phalloidin was used for cortical actin ( F-actin ) labeling ( Molecular Probes ) . Secondary antibodies used were goat anti-mouse and anti-rabbit conjugated to AlexaFluor-488 or AlexaFluor-555 ( Molecular Probes ) at 1:200 . MCF-10A cells grown on chamber slides were fixed in 2% paraformaldehyde at room temperature for 20 min . Cells were washed three times in PBS:glycine ( 130 mM NaCl , 7 mM Na2HPO4 , 3 . 5 mM NaH2PO4 , 100 mM glycine; 15 min each ) , blocked first in IF buffer ( 130mM NaCl , 7mM Na2HPO4 , 3 . 5mM NaH2PO4 , 7 . 7 mM NaN3 , 0 . 1% BSA , 0 . 2% Triton X-100 , 0 . 05% Tween 20 ) containing 10% goat serum ( 1–2 h ) and then with a second blocking buffer ( IF buffer containing 10% goat serum and 20 μg/ml goat anti-mouse F ( ab′ ) 2; Jackson Immunoresearch ) for 30–45 min . Anti-cleaved caspase 3 , anti-β-catenin or anti-GM130 were diluted in the second blocking buffer and incubated overnight at 4°C . Acini were stained with anti-rabbit AlexaFluor-555 . MDCK cells grown on chamber slides were fixed with 4% paraformaldehyde in PBS and permeabilized with 0 . 5% Triton X-100 in PBS . Fixed cells were blocked with 10% normal goat serum/1%BSA in PBS for 2 h , and then incubated with anti-ß-catenin and anti-PAR3 or anti-aPKC overnight at 4°C . Alternatively cells were incubated with anti-HTT 4C8 and anti-PAR3 , anti-aPKC or anti-RAB11A 3 h at RT . Cells were stained with anti-mouse and anti-rabbit AlexaFluor-488 or AlexaFluor-555 . Cysts with actin staining at the apical surface of cells surrounding a single lumen were identified as cysts with normal lumens . For all immunostainings , the slides were counterstained with DAPI ( Roche ) and mounted in Mowiol . The pictures were captured with a Leica SP5 laser scanning confocal microscope equipped with a X63 oil-immersion objective . Z-stack steps were of 0 . 5 μm . Images were treated with ImageJ ( http://rsb . info . nih . gov/ij/ , NIH , US ) . To measure the relative fluorescence intensity at the apical surface , a 30-pixel line was drawn across the apical surface and the cytoplasm using ImageJ software . The Line Scan function of ImageJ was used to reveal the relative fluorescence intensity across the line . The quantification of the polarization of the Golgi in MCF-10A 3-D acini was done using a home-built macro ( ImageJ software , see below for details ) . For live-cell imaging , MDCK cells were grown for 4 d in 24 mm Matrigel-coated coverglass , mounted in 6-well plate ( TPP ) . 30 min before observation , acini were incubated in culture media containing 4 μM FM4-64 . Imaging was performed at 37°C in 5% CO2 using an inverted microscope ( Eclipse Ti; Nikon ) with a 60 x 1 . 42 NA oil immersion objective coupled to a spinning-disk confocal system ( CSU-X1; Yookogawa ) fitted with an EM-CCD camera ( Evolve; Photometrics ) . Exposure times were 200 msec and 10% laser power . Image stacks of 50 planes spaced 1 μm apart were taken at six stage positions every 5 min for 2 h . Maximum intensity projection of the fluorescent channels was performed . Images were treated with ImageJ . MCF-10A and MDCK cells were lysed in NP40 buffer ( 50 mM Tris , pH 7 . 4 , 250 mM NaCl , 5 mM EDTA , 50 mM NaF , 1 mM Na3VO4 , 1% Nonidet P40 ( NP40 ) , 0 . 02% NaN3 ) and centrifuged at 11 , 000 x g for 10 min at 4°C . MCF-10A 3-D cultures were treated with trypsin 0 . 25% for 15 min to break the Matrigel , then acinar structures were washed with PBS1X and resuspended in NP40 lysis buffer , containing protease inhibitor cocktail ( Sigma ) , and centrifuged at 11 , 000 x g for 10 min at 4°C . 20–30 μg of protein extracts were loaded onto SDS-PAGE ( polyacrylamide gel electrophoresis ) and subjected to Western blot analysis . Primary monoclonal antibodies used were: anti-HTT 4C8 ( 1:3 , 000 ) , anti-HTT D7F7 ( 1:500 ) , anti-α-tubulin ( 1:5 , 000 ) . Primary polyclonal antibodies used were: anti-PAR3 ( 1:1 , 000 ) , anti-PAR6 ( 1:500 ) , anti-aPKC ( 1:1 , 000 ) , anti-RAB11A ( 1:500 ) and anti-mCherry ( 1:1 , 000; Institut Curie , Paris ) . Secondary antibodies used were HRP-conjugated goat anti-mouse/anti-rabbit ( 1:10 , 000; Amersham ) . For immunoprecipitations , MCF-10A cells were lysed in IP buffer ( Tris 50 mM pH 7 . 4 , 250 mM NaCl , 5 mM EDTA , 50 mM NaF , 1% Na3VO4 , 1% NP40 , 0 . 02% NaN3 , 50mM KH2PO4 ) containing protease inhibitor cocktail . Lysates ( 500 μg at 1 μg/μl ) were precleared 1 h at 4°C with 50 μl of a 50% solution of protein A or G beads . Extracts were incubated for 1 h at 4°C with 5 μg of anti-HTT ( 4C8 ) antibody or anti-PAR3 prebound with 50 μl of a 50% solution of protein A or G sepharose beads ( Sigma ) . Beads were washed three times with IP buffer . Bound proteins were eluted with SDS loading buffer , resolved by SDS-PAGE and subjected to immunoblotting analysis . Mice expressing the Cre recombinase under the control of the MMTV promoter ( MMTVCre ) and Httflox/flox mice were previously described [31 , 32] . All mice were bred in a C57BL6 genetic background . Httflox/flox mice were used as controls and MMTVCre;Httflox/flox as mutants . All experiments were performed in strict accordance with the recommendations of the European Community ( 86/609/EEC ) and the French National Committee ( 87/848 ) for care and use of laboratory animals ( permissions 91–448 to SH and 76–102 to SE ) . Whole mounts were prepared as described elsewhere [59] . Glands were fixed with MethaCarn ( 60% methanol , 30% chloroform , 10% glacial acetic acid; overnight , room temperature ) and hydrated by incubation in ethanol solutions ( 100% , 70% , 50% , 30%; 15 min each ) and distilled water ( 2 x 5 min ) . Mounts were then stained overnight with carmine ( 2% ) and aluminum potassium sulphate ( 5% ) ( Sigma , Buchs , Switzerland ) , dehydrated in ethanol solutions ( 70% , 90% , 95% , and 2 x 100%; 15 min each ) , and cleared with xylene ( overnight ) . Images were captured with an Epson Perfection 3200 scanner . Mammary gland development was analyzed as described elsewhere [60] . Briefly , the degree of ductal invasion was determined by dividing the duct length by the mammary gland length from mid-point of lymph node , and the numbers of total branches and TEBs were determined on whole-mount images by the ImageJ program . Dissected mammary fat pads were fixed in MethaCarn and embedded in paraffin . Seven μm-thick sections were deparaffinized before staining with primary antibodies ( overnight , 4°C ) , and secondary antibodies ( 1 h , room temperature ) . Nuclei were counterstained with DAPI . Primary antibodies used were: rabbit polyclonal anti-PAR3 ( 1:200; Chemicon ) , anti-aPKC ( 1:200; clone C-20 , Santa Cruz Biotechnology ) , anti-RAB11A ( 1:200; Abcam ) , anti-pSTAT5 ( Tyr694 , 1:100; Cell Signalling ) , anti-cleaved caspase 3 ( 1:100; Cell Signalling ) , anti-WAP ( 1:300; clone R-131 , Santa Cruz Biotechnology ) and anti-keratin 5 ( K5 ) ( 1:2 , 000; Covance ) ; rabbit monoclonal anti-KI67 ( 1:100; clone SP6 , Neo Markers ) ; and mouse monoclonal anti-HTT ( 1:300; 4C8 ) , anti-E-cadherin ( 1:200; BD Bioscience ) and anti-GM130 ( 1:100; BD Bioscience ) . Antigen retrieval was performed by boiling the slides for 10 min in a microwave in 10 mM citrate buffer ( pH 6 ) for cleaved caspase 3 , Ki67 , WAP , and p-STAT5A , or in EDTA buffer ( pH 8 . 8 ) for 10 min for PAR3 , aPKC , RAB11A , HTT , GM130 , K5 , and E-cadherin antibodies . Secondary antibodies used were goat anti-mouse and anti-rabbit conjugated to AlexaFluor-488 or AlexaFluor-555 or Biotin ( Vector Laboratories ) . The isolation of mammary epithelial cells and the separation of basal and luminal cells were done as described elsewhere [61 , 62] . Once mechanically dissociated , mammary fat pads were digested ( 90 min , 37°C ) in CO2-independent medium ( Invitrogen ) containing 5% fetal bovine serum , 3 mg/ml collagenase ( Roche Diagnostics ) and 100 U/ml hyaluronidase ( Sigma ) . Cells were resuspended in 0 . 25% trypsin-EDTA ( 1 min ) , and then in 5 mg/ml dispase ( Roche Diagnostics ) with 0 . 1 mg/ml DNase I ( Sigma ) ( 5 min ) . Red blood cells were lysed in NH4Cl . Basal and luminal cells were isolated from mammary epithelial cells obtained from the inguinal glands of five 12-wk-old virgin MMTV Cre mice . Cells were stained with the following antibodies: anti-CD24-FITC ( clone M1/69; BD Pharmingen ) , anti-CD49F-PE ( clone GoH3; BD Pharmingen ) , anti-CD45-APC ( clone 30-F11; Biolegend ) and anti-CD31-APC ( clone MEC13 . 3; Biolegend ) . Basal ( CD24-low/α6-high ) and luminal ( CD24-high/α6-low ) cells were purified using FACSAria III ( SORP ) ( Becton Dickinson ) . RNA samples were retrotranscribed using the First-Strand cDNA Synthesis Kit ( Invitrogen ) . cDNAs were diluted 1:10 and submitted to RT-PCR with 7900HT Fast real time PCR system ( Applied biosystems ) using power SYBR Green PCR Master mix ( Applied biosystems ) with the following oligonucleotide pairs: Htt ( 5′-CTCAGAAGTGCAGGCCTTACCT-3′ , 5′-GATTCCTCCGGTCTTTTGCTT-3′ and 5′-CTCAGAAGTGCAGGCCTTACCT-3′ , 5′-GATTCCTCCGGTCTTTTGCTT-3′ ) [63] , Cre ( 5′-TTCCCGCAGAACCTGAAGAT-3′ , 5′- GCCGCATAACCAGTGAAACA-3′ ) [62] , Krt18 ( 5′-CGAGGCACTCAAGGAAGAAC-3′ , 5′-AATCTGGGCTTCCAGACCTT-3′ ) , Elf5 ( 5′-CCAACGCATCCTTCTGTGAC-3′ , 5′-AGGCAGGGTAGTAGTCTTCA-3′ ) , Wap ( 5′-AACATTGGTGTTCCGAAAGC-3′ , 5′-GGTCGCTGGAGCATTCTATC-3′ ) , Csn2 ( 5′-TGCAGGCAGAGGATGTGCTCCAGGCT-3′ , 5′-GGCCTGGGGCTGTGACTGGATGCT-3′ ) ( Primer3v . 0 . 4 . 0; http://bioinfo . ut . ee/primer3-0 . 4 . 0/primer3 ) . β-actin ( 5′-AGGTGACAGCATTGCTTCTG-3′ , 5′-GCTGCCTCAACACCTCAAC-3′ ) and hprt ( 5′-GCTGGTGAAAAGGACCTCT-3′ , 5′-CACAGGACTAGAACACCTGC-3′ ) [29] genes were used as internal controls . Fold changes were calculated using the ddCT method . This macro was developed on site by F . P . Cordelières at the Institut Curie Imaging Facility . // Macro angle measurement run ( "Set Measurements . . . " , "area mean min centroid center integrated redirect = None decimal = 1" ) ; run ( "Clear Results" ) ; roiManager ( "reset" ) ; run ( "Select None" ) ; setTool ( "freehand" ) ; Xsel = newArray ( 3 ) ; Ysel = newArray ( 3 ) ; waitForUser ( "Draw ROI around the cyst" ) ; run ( "Duplicate . . . " , "title = duplicate duplicate" ) ; run ( "Properties . . . " , "channels = 3 slices = 1 frames = 1 unit = pixel pixel_width = 1 pixel_height = 1 voxel_depth = 1 frame = [0 sec] origin = 0 , 0" ) ; run ( "Select None" ) ; setTool ( "point" ) ; waitForUser ( "Indicate the point A" ) ; run ( "Measure" ) ; Xsel[0] = getResult ( "X" , 0 ) ; Ysel[0] = getResult ( "Y" , 0 ) ; run ( "Clear Results" ) ; waitForUser ( "Indicate the point B" ) ; run ( "Measure" ) ; Xsel[1] = getResult ( "X" , 0 ) ; Ysel[1] = getResult ( "Y" , 0 ) ; run ( "Clear Results" ) ; Xsel[1] = ( Xsel[0] +Xsel[1] ) /2; Ysel[1] = ( Ysel[0]+Ysel[1] ) /2; run ( "Select None" ) ; setSlice ( 3 ) ; setAutoThreshold ( "Percentile dark" ) ; run ( "Analyze Particles . . . " , "size = 30000-Infinity pixel circularity = 0–1 . 00 show = Nothing display clear add slice" ) ; resetThreshold ( ) ; verif = roiManager ( "Count" ) ; if ( verif >1 ) exit ( "Many cysts detected" ) ; if ( verif = = 0 ) exit ( "No detected cyst" ) ; Xsel[2] = getResult ( "X" , 0 ) ; Ysel[2] = getResult ( "Y" , 0 ) ; roiManager ( "reset" ) ; run ( "Clear Results" ) ; setTool ( "angle" ) ; makeSelection ( "angle" , Xsel , Ysel ) ; run ( "Measure" ) ; angle = getResult ( "Angle" , 0 ) ; if ( angle >90 ) angle = 180—angle; print ( "Measured angle is " + angle + " degree" ) ; GraphPad Prism 6 . 0 software ( San Diego , CA ) was used for statistical analysis . Complete statistical analyses with number of measures are detailed in S1 Data . | In the adult mammary gland , tissue architecture is maintained through the regulation of the polarity of epithelial cells , which organize around a central cavity called the lumen . The mammary epithelium comprises a basal layer , which contains myoepithelial contractile cells and so-called mammary stem cells , and a luminal layer of cells organized around the lumen . The establishment of apical-basolateral polarity in luminal cells allows the separation of the apical and basolateral membranes and the maturation of cell–cell junctions . The protein complex composed of PAR3 , PAR6 , and aPKC regulates apical polarity in several tissues , including the mammary epithelium , and it is known that the loss of PAR3 and aPKC interferes with mammary gland development and promotes mammary tumor metastasis . RAB11A , a protein that regulates intracellular trafficking , coordinates apical translocation of PAR3-PAR6-aPKC . Huntingtin ( HTT ) , the protein mutated in Huntington disease , modulates RAB11A activity and also regulates the microtubule-based vesicular trafficking in neurons . Using MCF10A , MDCK 2-D and 3-D cell cultures , and mouse models , we demonstrate here that HTT coordinates the apical vesicular trafficking of PAR3-PAR6-aPKC through RAB11A . We show that loss of HTT in luminal cells alters apical polarity , tissue architecture and the maturation of luminal cells during pregnancy and lactation in the mouse . Together , these findings uncover HTT-mediated vesicular trafficking as a new pathway in the establishment of epithelial apical polarity , with potential implications for health and disease . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Huntingtin Is Required for Epithelial Polarity through RAB11A-Mediated Apical Trafficking of PAR3-aPKC |
Rift Valley fever ( RVF ) is a zoonotic arboviral disease that is a threat to human health , animal health and production , mainly in Sub-Saharan Africa . RVF virus dynamics have been poorly studied due to data scarcity . On the island of Mayotte in the Indian Ocean , off the Southeastern African coast , RVF has been present since at least 2004 . Several retrospective and prospective serological surveys in livestock have been conducted over eleven years ( 2004–15 ) . These data are collated and presented here . Temporal patterns of seroprevalence were plotted against time , as well as age-stratified seroprevalence . Results suggest that RVF was already present in 2004–07 . An epidemic occurred between 2008 and 2010 , with IgG and IgM peak annual prevalences of 36% in 2008–09 ( N = 142 , n = 51 , 95% CI [17–55] ) and 41% ( N = 96 , n = 39 , 95% CI [25–56] ) , respectively . The virus seems to be circulating at a low level since 2011 , causing few new infections . In 2015 , about 95% of the livestock population was susceptible ( IgG annual prevalence was 6% ( N = 584 , n = 29 , 95% CI [3–10] ) ) . Monthly rainfall varied a lot ( 2–540mm ) , whilst average temperature remained high with little variation ( about 25–30°C ) . This large dataset collected on an insular territory for more than 10 years , suggesting a past epidemic and a current inter-epidemic period , represents a unique opportunity to study RVF dynamics . Further data collection and modelling work may be used to test different scenarios of animal imports and rainfall pattern that could explain the observed epidemiological pattern and estimate the likelihood of a potential re-emergence .
Rift Valley fever ( RVF ) is a zoonotic arboviral disease ( Phlebovirus , Family Bunyaviridae ) primarily affecting domestic livestock ( cattle , sheep and goats ) . Epidemics of RVF in livestock mainly cause abortions and neonatal deaths . In humans , symptoms are usually non-specific causing an influenza-like syndrome , but sometimes infection can also result in meningo-encephalitis , haemorrhagic fever and death [1] . Since its first description in Kenya in 1931 [2] , RVF has been reported in Sub-Saharan Africa , as well as in Egypt ( 1977 ) , in Madagascar ( 1979 ) , in the Arabian Peninsula ( 2000 ) , and in the islands of the Comoros archipelago ( 2007 ) [3]; the latter being located in the North of the Mozambique Channel , between Mozambique and Madagascar ( Fig 1 ) . RVF is transmitted to mammals mainly by mosquito bites ( main vectors belong to the genera Aedes and Culex ) . The hypotheses underlying RVF virus emergence is the concomitance of ( i ) the presence of susceptible livestock , ( ii ) an increase in vector abundance ( e . g . due to heavy rainfall ) thereby facilitating virus amplification and transmission , and ( iii ) the presence of the virus itself . The virus may be imported through the movements of infectious animals or arthropod vectors , or be present locally . RVF virus may persist between epidemics at a low level in livestock causing sporadic cases , or be maintained in potential local reservoirs . The latter includes vertical transmission in mosquito eggs and a large range of wild mammalian hosts , such as buffaloes , rodents and bats , although no good evidence exists on the latter hypothesis [4–6] . Humans may acquire the virus mainly by contact with infectious animal tissues , and also possibly via mosquito bites [1] . In Mayotte , RVF virus was detected for the first time in 2007 in humans [7 , 8] . Sequencing related this virus to the 2006–2007 eastern African Kenya-1 lineage [9] , suggesting a recent import onto the island from the African mainland . Retrospective serological analyses conducted on livestock sera ( collected between 2004 and 2008 ) showed that RVF virus had been on the island at least since 2004 , so before the 2007 introduction , but no virus sequencing had been done [10] . Since 2009 , an on-going disease surveillance system ( SESAM , Système d'Epidémiosurveillance Animale à Mayotte ) has been monitoring livestock health status , and collecting livestock sera . This paper aims at presenting eleven years ( 2004–15 ) of RVF serological data , merging the 2004–08 dataset presented in Cetre-Sossah et al . [10] , the 2012–13 dataset presented in Cavalerie et al . [11] , with all other serological analyses conducted between 2008 and 2015 . This large dataset collected on an insular territory represents a unique opportunity to study RVF epidemiology . It will be used to describe the past and current RVF status in Mayotte and propose relevant further data collection and mathematical modelling work to study RVF virus dynamics .
The island of Mayotte is a French overseas department that belongs to the Comoros archipelago ( Fig 1 ) . It is a small island of 374km2 ( about 35km North-South and 10km East-West ) , with a high population density ( 212 , 600 inhabitants in 2012 [12] ) . The estimated livestock population is approximately 20 , 000 cattle and 13 , 000 small ruminants ( sheep and goats ) . Average herd size is rather small ( 5 animals ) ; with animals mainly kept outdoors year round and raised for family consumption or cultural ceremonies [13 , 14] . Livestock ( cattle and small ruminants ) serological data were collated from different sources . The 2004–08 data were retrospective serological surveys [10]; whilst prospective data collections were conducted in 2008–15 . The different surveys collated and the serological testing data are presented in Table 1 , and detailed hereinafter . Rainfall and temperature are known to drive vector abundance and competence . The climate in Mayotte is marine tropical , with little temperature variation ( average year round 24–34°C ) , but important rainfall ( 1500mm on average per year ) . Two main seasons are observed: hot rainy ( December to March ) , and dry cool ( June to September ) [15] . In order to compare serological results to rainfall and temperature , data from 2004 to 2015 were sourced [16] . Monthly total rainfall and average temperature values were plotted against time , together with their average values by calendar month . The total dataset included the results of 5720 samples from 3529 animals ( from 448 herds , average herd size = 7 . 4 , median = 5 , IQR = [2–10] ) , sampled from and collected over 11 years ( October 2004 to June 2015 ) . All 5720 blood samples were tested for IgG , and 26 . 5% ( n = 1513 ) were tested for both IgG and IgM . The dataset used the data from surveys 1 to 5 , 6c , 7a and 8 ( Table 1 ) , defined as cross-sectional or repeated cross-sectional studies . Surveys 7b and 7c ( the follow-up of the longitudinal study , totalling 155 sera ) were not included as they clearly indicated measures of disease incidence; neither survey 6a nor 6b , as they were both targeted surveys . Due to the small size of Mayotte , estimates were produced for the whole island , as a single spatial epidemiological unit .
After examining the annual IgG and IgM prevalences ( Fig 2A and 2B , S1 Table ) , and IgG prevalence by age group ( Fig 3A–3G , S2 Table ) we divided the study period into three phases . Phase 1 is the period with less information available . It includes the first four epidemiological years 2004–05 , 2005–06 , 2006–07 and 2007–08 , during which IgG annual observed prevalence remained at 15% or less ( Fig 2A ) . Newly infected animals were reported in 2007–08 ( Fig 2B , IgM positive animals ) , but the information was scarce , and no information on the age of animals was available . Phase 2 ( 2008–09 and 2009–10 ) suggests recent transmission of the virus . Indeed , IgG annual prevalences were significantly higher compared to the rest of the study period , reaching their maximum average value of 36% in 2008–09 ( N = 142 , n = 51 , 95% CI [17–55] ) . In addition , Fig 2B shows a very high proportion of recently infected animals during that time , with IgM prevalence of 41% ( N = 96 , n = 39 , 95% CI [25–56] ) in 2008–09 and 36% ( N = 77 , n = 28 , 95% CI [22–51] ) in 2009–10 . Finally , IgG prevalence was similar across all age groups in 2008–09 and 2009–10 ( Fig 3A and 3B ) . Phase 3 ( 2010–11 to 2014–15 ) suggests a decrease in RVF virus transmission . IgG annual prevalence was significantly lower than during phase 2 , at approximately 10–15% ( Fig 2A ) , and a steep drop in the number of new infections was observed , with IgM prevalence in 2010–11 being only 4% ( N = 109 , n = 4 , 95% CI [0–7] ) ( Fig 2B ) . In addition , young animals ( 1 to 4 years old ) were less affected over time , and the IgG seropositive animals were older than 5 years ( Fig 3C–3G ) , presumably those infected in 2007–10 . The 2013–14 and 2014–15 seasons suggest a very low intensity of virus transmission . IgG annual prevalence reached its lowest value in 2014–15 , 6% ( N = 462 , n = 29 , 95% CI [3–10] ) ( Fig 2A ) ; and very few young animals were positive between 2012–13 and 2014–15 ( Fig 3E–3G ) . Although no animals were IgM positive in 2014–15 ( Fig 2B ) , two animals in the one-year old group were found IgG positive , indicating that the virus may have been still circulating ( Fig 3G ) . Finally , monthly rainfall pattern , and average monthly rainfall are shown in Fig 2A and 2B . Monthly rainfall varied over the study period ( ranging from 2 . 40 to 540 mm , S3 Table ) . During phase 2 and in 2014–15 , peaks of above-average rainfall were observed; whilst high IgM prevalence was reported in May-July 2009 during the dry season ( Fig 2B ) . Temperature values were available for the period 2005 to 2015 ( Fig 2A and 2B ) . Their range was narrow , with average monthly values varying between 24 . 7 and 28 . 1°C over the entire study period , and extreme monthly average minimum and maximum temperature values between 19 . 8°C and 32 . 8°C ( S3 Table ) .
The analyses of serological data showed that Mayotte probably experienced an RVF epidemic in livestock around 2008–10 . Peaks of above-average rainfall were observed during the epidemic phase , while variation of temperature was limited . The RVF virus seems to have remained endemic at a low level since 2011 , causing few new infections . In 2015 , about 95% of the livestock population was susceptible . Serological data were collected throughout the period 2004–15 , although not in a standardised manner . Retrospective data were obtained from sera stored at the Veterinary Services office . These sera were collected prior to RVF detection on the island , and it was not possible to determine exactly how this sampling was conducted in livestock . Although these results are not of a comparable value to the SESAM dataset , it gives precious information on past RVF infections , and evidence of on-going RVF virus transmission in 2008 in animals ( IgM positive ) , shortly after the newly imported RVF strain was sequenced from humans . From 2009 onwards , farms were sampled from the CAPAM official registry [14] . Since Mayotte became a French department in 2011 and part of the EU in 2014 , official registration has become compulsory , expanding the official list . The first farmers to register were possibly more affluent ( with larger herds ) , and with an increased awareness of livestock health than those farmers who registered later . The average herd size during the study period was 7 . 4 , and was equal to 9 in the SESAM study only . The official census from 2010 reports an average herd size of 5 [13] , confirming that our sample tended to capture larger than average herds . This could have slightly biased our estimates , since animals in Mayotte are raised outdoors and therefore may share a similar exposure to mosquito bites . In addition , analyses were conducted at the scale of the island . Accounting for a smaller spatial resolution ( e . g . administrative communes ) would be of limited benefit for an island that is relatively small and that has a similar ecosystem throughout . Finally , in the SESAM dataset , the RVF status of an animal did not influence whether this animal would be resampled in the future . Therefore , the dataset presented is valuable to estimate RVF prevalence in Mayotte through time , especially after 2008 . Few animals were sampled in surveys 1–4 ( Table 1 ) , which resulted in large confidence intervals , giving limited knowledge for the period 2004–08 ( Phase 1 ) . The absence of age-stratified prevalence also precluded drawing any hypothesis on RVF virus transmission for that period . However , although illegal import of animals was quite common at that time , it is unlikely that all animals found positive were imported; and this supports the hypothesis that the RVF virus had been circulating on the island at least since 2004 ( Phase 1 ) , four years before the sequencing of the new virus lineage in 2008 . The data in Phase 2 suggest that Mayotte experienced a large RVF epidemic in livestock; but no clinical signs as usually described in animals ( abortions , high mortality in young animals ) were detected , probably because no formal surveillance system in animals was in place at that time . Indeed , in humans attending the hospital for dengue-like illnesses , RVF virus was detected by RT-PCR in 8 patients between September 2007 and May 2008 , confirming the presence of the virus on the island [7] . Since the implementation of the SESAM surveillance system in livestock in 2009 , RVF has been monitored in livestock . There is evidence of new infections ( IgM positive or one-year-old animals IgG positive ) ; but the virus has not been detected nor isolated during Phase 3 . There is seasonality in rainfall , with very dry months in June-September , and extremely wet months , especially from December to March ( average monthly rainfall from 223 to 321mm ) . In our dataset , however , high IgM prevalences in 2009 with IgM positive animals across the whole island were detected during the dry season . This suggested a large epidemic but does not support a direct correlation between rainfall and new cases . In other ecosystems , such as in the Horn of Africa or Southern Africa , unusually heavy rainfall or an increase in vegetation density were observed from one to six months before the emergence of new RVF cases [18–21] . Therefore , it may well be that the heavy rain observed in January-March 2009 prepared suitable conditions for mosquito breeding during the dry season , explaining the high rates of new infections in May-July 2009 . In addition , field studies conducted in Mayotte in 2007 showed that natural larvae habitats specifically in rural areas allowed Ae . aegypti to survive the dry season [22] . Finally , it is also possible that a high number of new infections also occurred during the dry season of 2008 following heavy rains , but unfortunately no IgM testing was done at that time . Very little variation was observed in temperature over the period 2005–15 . During the 2010 epidemic in South Africa , temperature above 25°C was the most important risk factor [23] , and experimental studies in Culex pipiens and Aedes taeniorhynchus , two RVF vector species , showed that temperature above 26°C favoured RVF virus amplification and transmission [24 , 25] . The high average temperatures observed in Mayotte year round may therefore provide almost constantly suitable conditions for RVF virus transmission; and RVF dynamics observed on the island maybe driven mainly by rainfall patterns . There is no information on the virus lineage that circulated in 2004–07 . The sequencing of the Mayotte 2008 lineage placed the virus into the East African clade that includes the Kenyan 2006–2007 and Madagascar 2008 lineages [9 , 26] . This suggests that the Mayotte 2008–10 epidemic might have followed , not only heavy rainfall , but also the import of infectious animals with an RVF virus lineage new to the Mayotte ecosystem . Trade of livestock exists from the African mainland and Madagascar , into the Comoros islands and then Mayotte ( Fig 1 ) , although the latter is illegal [27] . This import scenario was also supported by the detection of IgM positive goats illegally imported [10] from Anjouan ( Fig 1 ) , between November 2007 and March 2008 . Since 2008 , no virus has been isolated nor sequenced in animals . The Mayotte 2008 lineage could persist at a low level in livestock or also potentially in wildlife [1 , 6] , causing the latest sporadic new infections in 2014 and 2015 . Alternatively , as Mayotte still experiences regular animal illegal imports , introductions of other RVF virus lineages cannot be excluded , such as the Anjouan 2011 lineage detected in a zebu [28] . The working hypotheses underlying RVF virus re-emergence presented in the introduction are the concomitance of ( i ) the presence of susceptible livestock , ( ii ) an increase in vector abundance ( e . g . due to heavy rainfall ) , and ( iii ) the presence of the virus emerging from local reservoirs or newly introduced . The data presented here suggest that Mayotte currently meets two of the conditions for re-emergence , that are: ( i ) a high proportion of susceptible livestock that reached about 95% in 2015 , and ( iii ) the presence of the virus , evidenced by the new infections observed in phase 3 . Therefore , we hypothesize that with heavy rainfall , such as it was observed in 2008–10 , RVF virus could re-emerge . Modelling work was done to assess whether climate pattern could favor RVF virus persistence in Mayotte , which appeared to be true even under very low transmission assumption [11] . Further modelling work on RVF virus emergence can be implemented accounting for animal imports , wildlife and climate data . Different scenarios of animal imports and rainfall patterns could be tested to explain the observed epidemic dynamics and estimate the likelihood of a future epidemic . Further data collection would therefore be necessary , including ongoing climate data , surveillance in livestock , RVF prevalence in wildlife , RVF data on illegally imported animals , and virus detection , isolation , and sequencing when applicable ( S1 Text ) . In conclusion , this study has shown the value of repeated serological testing to explain RVF population dynamics in this island population despite limited resources . Linking these ongoing studies with additional data and modelling could also shed further light on the origin and re-emergence mechanisms of this virus . | Rift Valley fever ( RVF ) is a viral zoonotic disease mainly present in Sub-Saharan Africa , transmitted by mosquitoes and primarily affecting livestock ( cattle , sheep and goats ) . The epidemiology of the disease is not fully known , mainly because of data scarcity . In Mayotte , an island close to Madagascar , RVF has been present since at least 2004 , and retrospective and prospective surveys have been conducted to collect data from livestock over 11 years . Our work uses this 2004–15 dataset to describe the past and current epidemiology of RVF on the island . Results suggest that the disease was endemic between 2004–07 . An epidemic occurred between 2008 and 2010 . Since 2011 , the virus appears to be circulating at a low-level causing few new infections . Modelling work may be used to explain the observed epidemiological pattern , and estimate the likelihood of a potential re-emergence . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"livestock",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"infectious",
"disease",
"epidemiology",
"pathogens",
"tropical",
"diseases",
"microbiology",
"vertebrates",
"animals",
"mammals",
"viruses",
"rift",
"valley",
"fever",
"farms",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"bunyaviruses",
"veterinary",
"science",
"infectious",
"diseases",
"zoonoses",
"veterinary",
"epidemiology",
"serology",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"agriculture",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"cattle",
"amniotes",
"bovines",
"organisms"
] | 2016 | The Epidemiology of Rift Valley Fever in Mayotte: Insights and Perspectives from 11 Years of Data |
The pacific islands of Micronesia have experienced several outbreaks of mosquito-borne diseases over the past decade . In outbreaks on small islands , the susceptible population is usually well defined , and there is no co-circulation of pathogens . Because of this , analysing such outbreaks can be useful for understanding the transmission dynamics of the pathogens involved , and particularly so for yet understudied pathogens such as Zika virus . Here , we compared three outbreaks of dengue and Zika virus in two different island settings in Micronesia , the Yap Main Islands and Fais , using a mathematical model of transmission dynamics and making full use of commonalities in disease and setting between the outbreaks . We found that the estimated reproduction numbers for Zika and dengue were similar when considered in the same setting , but that , conversely , reproduction number for the same disease can vary considerably by setting . On the Yap Main Islands , we estimated a reproduction number of 8 . 0–16 ( 95% Credible Interval ( CI ) ) for the dengue outbreak and 4 . 8–14 ( 95% CI ) for the Zika outbreak , whereas for the dengue outbreak on Fais our estimate was 28–102 ( 95% CI ) . We further found that the proportion of cases of Zika reported was smaller ( 95% CI 1 . 4%–1 . 9% ) than that of dengue ( 95% CI: 47%–61% ) . We confirmed these results in extensive sensitivity analysis . They suggest that models for dengue transmission can be useful for estimating the predicted dynamics of Zika transmission , but care must be taken when extrapolating findings from one setting to another .
Many infections of humans are transmitted by mosquitoes . Dengue virus is one of the major pathogens infecting humans worldwide , causing an estimated 50–100 million cases resulting in about 10 , 000 deaths annually [1] . Confined mainly to tropical regions because of its reliance on transmission through Aedes mosquitoes , it is endemic in more than 150 countries across the world [2] . Its four circulating serotypes cause a wide range of clinical symptoms and severities , but most cases resolve without progressing to the more severe forms , dengue hemorrhagic fever and dengue shock syndrome . Upon infection following bite by an infectious female mosquito , the virus undergoes a period of incubation before progressing to disease in an estimated 20–50% of infected people [3 , 4] , with symptoms lasting approximately one week . The relative infectiousness of symptomatically and asymptomatically infected people remains a topic of active study , with recent evidence indicating that symptom-free people might be more infectious to mosquitoes than clinically symptomatic people [5 , 6] . Infection results in lifelong immunity to the same serotype but subsequent infection with heterologous serotypes is associated with higher rates of severe dengue [7] . Zika virus , a member of the Flaviviridae family like dengue and also transmitted by Aedes mosquitoes , was discovered in Africa in 1947 [8] . Formerly believed to be mostly confined to primate species , it has caused occasional cases in humans across Africa and equatorial Asia in the decades after its discovery , before sparking its first observed outbreak in humans on the Yap Main Islands , Micronesia , in 2007 [9 , 10] . Following further outbreaks on Pacific islands in 2013/14 [11–13] , cases of an illness characterised by skin rash were reported from Brazil beginning in March 2015 and Zika virus circulation confirmed in May 2015 [8 , 14 , 15] . Zika virus appears to largely cause asymptomatic infection or mild disease and a non-itchy rash . However , it has recently been linked to neurological issues in rare cases , particularly microcephaly when contracted in pregnancy [16] and Guillain-Barré syndrome [17 , 18] . A recent increase in reported occurrences of microcephaly in Brazil has led to the declaration of a Public Health Emergency of International Concern by the World Health Organization , to “reduce infection with Zika virus , particularly among pregnant women and women of childbearing age . ” [19] . In contrast to dengue , Zika virus has not been described in great detail , and its epidemiology in human populations remains poorly understood . Here , we characterise the epidemiology of dengue and Zika outbreaks in tropical island settings by comparing three outbreaks in Yap State , Micronesia: the outbreak of Zika virus on the Yap Main Islands in 2007 , a dengue outbreak on the Yap Main Islands in 2011 , and a dengue outbreak on the island of Fais . Island outbreaks are a particularly useful vehicle for understanding transmission dynamics as cases usually occur in episodic outbreaks , limiting interaction between pathogens and reducing the chances of misclassification . Moreover , all three outbreaks share particular characteristics: the two dengue outbreaks share the infecting agent; the two outbreaks on the Yap Main Islands the setting; and the Zika outbreak on the Yap Main Islands and the dengue outbreak on Fais that they probably struck immunologically naïve populations . Moreover , evidence suggest that both Aedes aegypti and Aedes hensili are important epidemic vectors in both settings , with the latter only recently having been implicated in outbreaks of arboviruses [20 , 21] . We exploit these relationships to comparatively study the three outbreaks by fitting a hierarchical transmission model to the three time series , holding parameters constant between the outbreaks where they represent a common element .
Yap State is one of the four states of the Federal States of Micronesia , consisting of the Yap Main Islands ( also called Yap Proper or simply Yap ) and fourteen outer atolls spanning an area of approximately 120 km2 . The Yap Main Islands consist of four major inhabited islands and six smaller ones that form a contiguous land mass of approximately 79 km2 . The 7 , 370 inhabitants of the Yap Main Islands ( 2010 census , population density 93/km2 ) live in villages , the largest of which is the capital of Yap State , Colonia ( population 3 , 126 ) , with the remaining villages mostly located along the shore line . Fais is one of the outer islands of Yap State which lies about 270 km to the East of the Yap Main Islands and has a much smaller land mass ( 2 . 6 km2 ) ( Fig 1 ) . The population of 294 ( 2010 census , density 113/km2 ) is concentrated in a single population centre that spans approximately a quarter of the island’s area . The Yap Main Islands have experienced several outbreaks of dengue in the past , including an outbreak of serotype 4 in 1995 [20] and an outbreak of serotype 1 in 2004 [22] . The outbreak of Zika in 2007 , on the other hand , was the first observed outbreak of Zika in any human population [9] . The outbreak of dengue in Fais , too , is believed to have been the first ever on the island [23] . Because of its stable climate , mosquito numbers are not believed to vary seasonally in Micronesia [24] . The dengue time series from the Yap Main Islands and Fais consist of clinically suspected dengue cases as identified by the Yap Department of Health [23] using the WHO ( 2009 ) case definition . A small proportion of cases ( 9% ) were reported on outer islands and included in the time series for the Yap Main Islands as we did not have access to a time series where the two were separated . Dengue virus serotype 2 was confirmed by reverse transcriptase polymerase chain reaction by the CDC Dengue Branch , Puerto Rico . The Zika time series from the Yap main islands consists of probable and confirmed cases as identified in a combination of prospective and retrospective surveillance at all health centres on Yap [9] . All three time series of cases are summarised in Table 1 . The outbreak of Zika on the Yap Main Islands had its first cases reported with onset in mid-April 2007 and the last in July 2007 . Overall , a total of 108 cases were classified as probable ( 59 ) and confirmed ( 49 ) in a population of 7 , 370 , and 73% ( 95% CI: 68%–77% ) were later found with evidence of recent Zika infection in a household survey [9] . The outbreak of dengue on the Yap Main ( and Outer ) Islands began with a case with disease onset on 1 September , 2011 , and two more onsets on the following day . The next case was reported with onset a week later , on 8 September , followed by another cluster around 15 September , and sustained spread beginning another week later , around 22 September , 2011 . The peak of the outbreak occurred in the week beginning 24 November , 2011 , with 142 cases reported with onset during that week . The last cases were reported with onset on 16 February , 2012 . The outbreak of dengue on Fais overlapped with the outbreak on the Yap Main Islands . It began on 10 November , 2011 , with onset of disease in the likely index case . No further case was reported for 16 days , before cases started increasing after the second reported case ( onset on 27 November , 2011 ) to a peak of 72 cases reported with disease onset in the week beginning 1 December , 2011 . The last reported disease onsets were 2 cases on 20 December , 2011 . Overall , 155 clinical cases were reported among the 294 residents . We implemented a variant of the Ross-McDonald model [26 , 27] , schematically depicted in Fig 2 . The human population of size NH was divided into susceptible ( SH ) , incubating or exposed ( EH ) , infectious ( IH ) and recovered ( RH ) compartments . The mosquito population of unknown size was divided into the proportion susceptible ( sM ) , incubating ( eM ) or and infectious ( iM ) . We assumed that the size of the human ( NH ) and vector populations did not vary over the course of the modelled outbreaks ( i . e . , we ignored birth and death rates in the human populations and assumed them to be the same in the vector populations ) , and further assumed that infection resulted in immunity that lasted for at least the duration of the outbreak , and that vertical transmission in the mosquito population could be neglected [28] . In our model , everybody who gets infected can transmit the virus to mosquitoes [6] . Any lack of symptomatic disease is reflected in the mean proportion of cases reported r , as defined in the likelihood function below . The system of ordinary differential equations ( ODEs ) governing the outbreaks are: d S H d t = - λ H S H d E H d t = + λ H S H - δ H E H d I H d t = + δ H E H - γ H I H d R H d t = + γ H I H d s M d t = + ν M - λ M s M - μ M s M d e M d t = + λ M s M - ( δ M + μ M ) e M d i M d t = + δ M e M - μ M i M ( 1 ) Here , λH and λM are the forces of infection acting on humans and mosquitoes , respectively , δH = 1/Dinc , H and δM = 1/Dinc , M are the incubation rates , defined as the inverse of the average incubation periods Dinc , H and Dinc , M in humans and mosquitoes , respectively , γH = 1/Dinf , H is the recovery rate in humans , defined as the inverse of the average duration of infectiousness , νM is the birth rate of female mosquitoes or number of susceptible female mosquitoes born per female mosquito per unit time , here assumed to be equal to the mosquito death rate μM = 1/Dlife , M , defined as the inverse of the average mosquito life span Dlife , M . This ensured that mosquito population sizes remained constant over the course of each outbreak . The forces of infection can be written as λ H = τ b H m i M λ M = τ b M I H N H ( 2 ) where τ is the number of human blood meals taken by a single female mosquito per unit time , bH and bM are the probabilities that a bite by an infectious female mosquito leads to infection in a human and a bite on an infectious human leads to infection in a mosquito , respectively , and m is the number of female mosquitoes per human . The human-to-human reproduction number of this model is R H → H = R H → M × R M → H = τ b M γ H × τ m b H μ M δ M μ M + δ M ( 3 ) The basic reproduction number of the system , or the average number of secondary infections ( in human or mosquito ) from a primary infectious bite can be calculated from the next-generation matrix [29] , and is the square root of the human-to-human reproduction number given in Eq 3 . The equilibrium generation interval , or the mean time between the infection of a primary case and its secondary cases , relates reproduction numbers ( which only describe reproduction per generation , without an explicit time scale ) to the time scale of transmission . For our model , in an equilibrium situation it would be [30]: Geq=Dinc , H+Dinf , H+Dinc , M+Dlife , M ( 4 ) In an outbreak situation , observed generation intervals deviate from the theoretical value at equilibrium and change over time . When new infections are generated at approximately exponential rate , observed mean generations interval are smaller than the equilibrium value as most infectious people will only just have been infected [31] . This issue has recently been generalised to the whole distribution of generation intervals , and beyond assumptions of exponential growth [32] . For Zika , the generation interval has been estimated to be between 10 and 23 days [33] , combining estimates for Dinc , H of 3–12 days , Dinc , M of 4–6 days , assuming Dinf , H = Dlife , M = 0 , that is that mosquitoes are infected by humans and vice versa just after their infectious period started , as well as an additional delay before symptomatic humans become viraemic of 3–5 days . If humans are , instead , taken to be viraemic for the first 3–5 days from symptoms onset [34] , the estimated range shortens to 7–18 days . This should be taken as a lower limit for observed generation intervals , as in reality some infections will be caused by humans/mosquitoes that have been infectious for some time . A second study estimated the equilibrium generation interval using all the components of Eq 4 and drawing from a systematic review of the natural history of the infection [35] . Assuming that humans and mosquitoes were equally likely to cause infection in mosquitoes or humans , respectively , the generation interval was estimated to be 20 days ( mean , 95% CI 15 . 6–25 . 6 ) , with a standard deviation of 7 . 4 days ( mean , 95% CI 5 . 0–11 . 2 ) , using an average mosquito life time of 5 days with standard deviation of 1 . 7 days [36] . To fit the model to the data sets , we used a Bayesian framework , generating samples from the posterior distribution using Markov-chain Monte Carlo ( MCMC ) . The observation likelihood at each data point was assumed to be distributed approximately according to a Poisson distribution with rate rZH , where ZH is the number of new human infections of Zika per reporting interval , and r is the proportion of these infections that were reported , estimated using a normal approximation with mean and variance both equal to rZH . We only had access to a weekly time series of Zika on the Yap Main Islands , and therefore aggregated the daily time series of dengue cases to weekly numbers to make estimates comparable between time series . We fixed the biting rate to 1 per day [37] . Since we did not have enough information on mosquito life span to inform a full prior distribution , we further fixed the life span of the mosquito to either 1 week [36] or 2 weeks [38] , and compared the two sets of fits using the Deviance Information Criterion ( DIC ) [39] . We modelled the other natural history parameters ( intrinsic and extrinsic incubation periods and infectious period in humans ) with dengue-like priors , assuming that infectiousness starts 1 . 5 days before symptom onset [36 , 40] and ends 1 . 5 days before their end . These prior distributions overlap with ones that have recently been estimated from the available data for Zika virus infections [35 , 36] . We estimated the remaining parameters of the model by fitting to all three time series simultaneously , with the following constraints: probabilities of infection from a potentially infectious bite , proportion reported , intrinsic and extrinsic incubation periods and human infectious periods were all to be disease-specific but the same across settings; mosquito densities , on the other hand were to be setting-specific but the same across the two pathogens , reflecting potential differences in the sizes of vector populations but also in human population density and behaviour . For the outbreak of dengue the Yap Main Islands , we assumed that only a proportion q of the population was susceptible to infection . For the Zika outbreak on the Yap Main Islands , we assumed that the whole population was susceptible to infection . In other words , our Zika model is the assumed equivalent of a single-serotype dengue model not incorporating cross-reactivity between heterologous viruses or serotypes . The dengue outbreak in Fais , too , was assumed to strike a fully susceptible population , as it was the first known outbreak of dengue on the island . All outbreaks were started with a single infectious case , and the date at which that case became infectious fitted as a separate parameter ( rounded to the week ) for all three outbreaks . The MCMC procedure for parameter estimation was implemented using the libbi software package [41] , run from the statistical package R [42] using the rbi [43] and rbi . helpers [44] packages . After adapting the size and shape of the multivariate normal proposal distribution in trial runs , the algorithm was run for 10 million iterations and convergence confirmed visually . All code and data used to generate the results are available at http://github . com/sbfnk/vbd . We fitted two modified models to a data set containing an additional data point included in the fit to reflect the final outbreak size observed in a serological study on the Yap Main Islands [9] . The likelihood at this data point was given by a normal distribution centred around the final size , with a standard deviation of 2 . 2% to reflect the 95% confidence interval reported in the serological study . In one model , the population size of Yap Main Islands would be reduced by a factor ρ [45] , whereas in the other one the initial proportion susceptible would be a proportion q of the whole population but everybody susceptible to mosquito bites , as in our model for the dengue outbreak on the Yap Main Islands . We further fitted a two-patch metapopulation model to the outbreaks on the Yap Main Islands . While we did not have any spatially resolved data to inform such a model , the outbreak of Zika on the Yap Main Islands could be interpreted to consist of two peaks , a structure that would be expected to reproducible by a two-patch model . In this model , the outbreak started in a patch which contained a proportion φ of the total population . This and another patch shared the same parameters , and humans in each patch exerted a force of infection on mosquitoes in the other ( representing human movement ) that was reduced by a factor σ with respect to the force of infection within each patch .
The models with mosquito life spans of 1 week vs 2 weeks fit the data equally well ( DIC difference <1 ) , with fits combining both models shown in Fig 3 . Assuming that both were equally likely to be true and combining the posterior distributions , the estimated disease-specific durations of infection and incubation largely corresponded to the given prior distributions ( Table 2 ) . There was , however , a more than twenty-fold difference in the proportion of infectious people reported , between a median estimate of 53% ( IQR 51%–56% , 95% CI 47%–61% ) for dengue and 1 . 6% ( IQR 1 . 5%–1 . 7% , 95% CI 1 . 4%–1 . 9% ) for Zika . Location-specific parameters indicated a considerable difference in the number of female mosquitoes per person , with a mean estimate of 1 . 0 ( IQR 0 . 69–1 . 5 , 95% CI 0 . 38–8 . 4 ) on the Yap Main Islands and 4 . 7 ( IQR 3 . 4–7 . 2 , 95% CI 2 . 1–30 ) on Fais . The proportion of the population initially susceptible to dengue on the Yap Main Islands was estimated to be 27% ( IQR 26%–29% , 95% CI 24%–32% ) . The median estimates of the human-to-human reproduction number , RH → H were 11 ( IQR 9 . 7–13 , 95% CI 8 . 0–16 ) for dengue on the Yap Main Islands , 7 . 6 ( IQR 6 . 3–9 . 6 , 95% CI 4 . 8–14 ) for Zika on the Yap Main Islands , and 51 ( IQR 40–71 , 95% CI 28–102 ) for dengue on Fais ( Fig 4 ) . By combining the estimated parameters between settings and disease , we estimated R0 for Zika on Fais to be 35 ( posterior mean , IQR 26–52 , 95% CI 18–79 ) . The differences in R0 between Yap and Fais are reflected in the different estimated differences in the number of female mosquitoes per person , which results in differences in the number of bites experienced per person . Much of the variation in R0 is explained by the different lengths of the generation interval which was poorly identified from the data ( Fig 4 , Table 3 ) . This is particularly the case for dengue in Fais , where all infections occurred in one to two generations , depending on the length of the generation interval ( Fig 5 ) . The alternative models with reduced population size or reduced susceptibility against Zika on the Yap Main Islands were both able to reproduce the observed proportion infected of 73% ( see Supporting Information S1 Text ) . In the model with reduced population size the initial proportion susceptible to dengue on the Yap Main Islands was estimated to 37% ( median , IQR 35%–39% , 95% CI 32%–44% ) , leading to a smaller human-to-human reproduction number of 8 . 7 ( median , IQR 7 . 3–10 , 95% CI 6 . 0–13 ) and greater proportion of Zika cases reported of 2 . 2% ( median , IQR 2 . 1%–2 . 3% , 95% CI 1 . 9%–2 . 7% ) . In the model where only a proportion of the population q was susceptible to infection with Zika on the Yap Main Islands , the estimate of the proportion susceptible to dengue and human-to-human reproduction numbers were unchanged , while the proportion of Zika cases reported increased to 2 . 2% ( median , IQR 2 . 1%–2 . 4% , 95% CI 1 . 8%–2 . 7% ) The two models described the data equally well ( DIC difference <1 ) . The alternative two-patch metapopulation model produced very similar parameter fits to the single-patch model . In particular , the fit to the outbreak of Zika on the Yap Main Islands produced a single peak unless it was fitted in isolation .
We have analysed three outbreaks of mosquito-borne disease on small islands of Micronesia using a mathematical model . We exploited the overlap between those outbreaks in setting and disease to constrain parameter values and used this to investigate differences in transmission dynamics . While we found large difference between the reproduction numbers for dengue in two different island settings , our estimates of the reproduction numbers for dengue within the same settings are similar . Our approach of fitting three time series concurrently in a hierarchical model with common parameters helped identify some parameters that would not be identifiable by observing the outbreak in isolation . For example , the parameters m ( ratio of female mosquitoes to humans , fixed across diseases ) and bH ( probability of infection of a susceptible human when bitten by an infectious mosquito , fixed across locations ) would not be separately identifiable when considering a single time series , but can , in principle , be identified when considering multiple locations and diseases . The proportion of cases of dengue reported was informed by the final size of the dengue outbreak in Fais which , in turn , enabled estimation of the initial proportion susceptible of the dengue outbreak on the Yap Main Islands , again from the final outbreak size . With these two parameters established , the reproduction number of dengue in the two settings could be estimated from the initial growth rate and outbreak duration , as a function of the generation interval . The generation intervals themselves were poorly identified in the data , and the corresponding marginal posterior distributions largely overlapping with the prior distributions . Parameters for the Zika outbreak on the Yap Main Islands were similarly identified . With the reproduction number given by the initial growth rate and outbreak duration , the proportion of cases reported could be estimated from the reported final outbreak size of the epidemic . In this context it should be noted that with the values of the reproduction number we estimated , one would expect nearly all of the population to get infected , in contrast to the 73% ( 68%–77% ) estimated to have been infected in a serological study after the outbreak [9] . It remains an open question how to best explain a rapidly growing epidemic that spreads through large parts of a population in a few generations without rendering everybody seropositive , a phenomenon also observed in the 2013–14 Zika outbreak in French Polynesia [13 , 45 , 48] . In the case of Zika on the Yap Main Islands , there might be several reasons for the discrepancy between modelled outbreak sizes and observed serology , such as the sensitivity of the used diagnostic test or lack of seroconversion at low-level exposure . If , on the other hand , the measured seropositivity reflects true infection history , its discrepancy with our modelled outbreak sizes could be because some individuals were not exposed to infectious mosquito bites due to spatial heterogeneity or because behavioural factors prevented them from getting bitten , which would not be captured in our model of a homogeneously mixing population . Fitting a model that included a factor to reflect this produced qualitatively the same results as the original model while lowering the reproduction number of dengue on the Yap Main Islands and increasing the proportion estimated to be initially susceptible to dengue infection on the Yap Main Islands well as the reporting proportion of cases of Zika that were reported . Lastly , the discrepancy could be because some of the population was protected from infection because of cross-immunity from prior infection with another virus , although current evidence points to an opposite effect of antibody-dependent enhancement due to prior dengue infection [49 , 50] . In the model fits in this scenario , the proportion of cases of Zika that was reported increased . In all cases , this proportion remained well below the equivalent number for dengue . The case series for Zika on the Yap Main Islands could be interpreted to consist of two peaks . In our basic model , we did not include a mechanism that could have produced these peaks , as we did not have access to any ( for example , spatially resolved ) data that could have informed such a choice . Whilst two peaks could be produced by a model with spatial heterogeneity , this would have been expected to produce a similar pattern in the dengue outbreak on the Yap Main Islands , which consisted of a single peak . Because this is not the case , fits with a two-patch model still yielded a single peak for Zika on the Yap Main Islands . Fitting the Zika outbreak on the Yap Main Islands in isolation using a two-patch model did reproduce two peaks , but ignored the additional information contained in the dengue outbreaks , giving less credence to the fits . In this context , it is worth noting that our model is deterministic and ignores any underlying stochasticity that may have played a role especially early and late in the outbreaks . All uncertainty in our model is in the likelihood which encodes the reporting process . The beginning of what could be seen as a second wave coincided with the arrival of the US Centres for Disease Control and Prevention ( CDC ) teams in Yap , which may have changed reporting rates [9] . With this in mind , our estimate of the proportion of cases reported should be interpreted as an average over the whole outbreak . Our estimates of human-to-human reproduction numbers for dengue on the Yap Main Islands are consistent with those previously reported in the literature [51] , and overlap with the range of 2 . 8–12 . 5 estimated from the exponential growth rate alone [52] . The estimate of the human-to-human reproduction number for dengue in Fais , on the other hand , is one of the largest ever observed in the literature , and larger than a previous estimate of dengue on a small island , although comparable in order of magnitude [53] . It is conceivable that on Fais , everybody was infected within a generation or two . The outbreak hit a population that occupies a small island ( confining geographical space both for people and vectors ) and is not believed to ever have been exposed to dengue previously , which could explain the rapid spread . More generally , the estimates for R0 are similar between dengue and Zika where they were observed in the same setting on the Yap Main Islands , but differ strongly between the dengue outbreaks on the Yap Main Islands and Fais . This suggests that outbreak setting and human population and mosquito densities are more important in governing transmission dynamics than differences between the pathogens . In other words , while our results suggest that insights from studying dengue transmission in one location can be used to predict the spread of Zika , care must be taken when extrapolating from insights on either of the pathogens in one location to another . Our results suggest that measuring mosquito densities and biting exposure in different settings could provide important information for estimating expected attack rates . In our case , Fais is a much smaller island , and one in which the assumption of random mixing is much more justified than on the Yap Main Islands , where spatial transmission dynamics may have diluted the potential for rapid spread . Our estimates of the reproduction number should be interpreted with caution as they could be influenced by heterogeneity . It has been shown if mixing is proportionate but heterogeneous ( which is to be expected for dengue or Zika ) , the reproduction number increases the stronger the heterogeneity [54] . This can cause difficulties in the interpretation of reproduction numbers based on homogeneous models applied to outbreak data [55] . This and other structural limitations of the modelling approach could be contributing in an unknown way to differences or similarities in the estimated values of the reproduction number , and experiments and observational studies will be required to corroborate our findings . In summary , we have studied three island outbreaks of vector-borne disease and elucidated on similarities and differences . We found that Zika transmission dynamics are similar to dengue when observed in the same setting , and that differences in human population structure and vector density are more important in determining transmission dynamics than difference between the two pathogens . For a new and yet understudied virus such as Zika , comparative studies like this one , especially when conducted on outbreaks in closed populations , can yield important insights into analogies that could be explored for interpreting observed transmission patterns and predicting future dynamics . Field studies on differences in vector density and biting exposure , as well as comparative modelling studies in other settings , would yield important further insights into the relationship between the transmission dynamics of Zika and dengue and the specific setting in which they occur . | Dengue and Zika are related viruses that are transmitted by the same species of mosquitoes . While dengue is well described and has affected people around the world for a long time , Zika has only recently caused outbreaks in human populations . To investigate whether the expected behaviour of Zika is similar to that of dengue , we compared three outbreaks in island populations of the pacific: two dengue outbreaks and one Zika outbreak . Island outbreaks are useful laboratories for understanding the spread of infections because they are usually short , well-identified episodes , whereas elsewhere it can be difficult to identify the properties of outbreaks when different viruses spread at the same time . In our investigation of the outbreaks in Micronesia we found that dengue and Zika virus did indeed behave similar in outbreaks they caused on the Yap Main Islands . A dengue outbreak on the smaller island of Fais , on the other hand , was different from the dengue outbreak on Yap in that transmission seems to have been much more intense . We conclude that dengue outbreaks are indeed a good model for Zika outbreaks when considered in the same setting , but that one must be careful when comparing outbreaks in different settings . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"death",
"rates",
"invertebrates",
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"infectious",
"disease",
"epidemiology",
"demography",
"pathogens",
"vector-borne",
"diseases",
"microbiology",
"animals",
"viruses",
"rna",
"viruses",
"population",
"biology",
"insect",
"vectors",
"infectious",
"diseases",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"population",
"metrics",
"mosquitoes",
"flaviviruses",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"population",
"density",
"organisms",
"zika",
"virus"
] | 2016 | Comparative Analysis of Dengue and Zika Outbreaks Reveals Differences by Setting and Virus |
Mammals show a wide range of brain sizes , reflecting adaptation to diverse habitats . Comparing interareal cortical networks across brains of different sizes and mammalian orders provides robust information on evolutionarily preserved features and species-specific processing modalities . However , these networks are spatially embedded , directed , and weighted , making comparisons challenging . Using tract tracing data from macaque and mouse , we show the existence of a general organizational principle based on an exponential distance rule ( EDR ) and cortical geometry , enabling network comparisons within the same model framework . These comparisons reveal the existence of network invariants between mouse and macaque , exemplified in graph motif profiles and connection similarity indices , but also significant differences , such as fractionally smaller and much weaker long-distance connections in the macaque than in mouse . The latter lends credence to the prediction that long-distance cortico-cortical connections could be very weak in the much-expanded human cortex , implying an increased susceptibility to disconnection syndromes such as Alzheimer disease and schizophrenia . Finally , our data from tracer experiments involving only gray matter connections in the primary visual areas of both species show that an EDR holds at local scales as well ( within 1 . 5 mm ) , supporting the hypothesis that it is a universally valid property across all scales and , possibly , across the mammalian class .
Understanding brain networks is arguably one of the major challenges of the 21st century [1] . The mammalian cortex is an extraordinary computational device , and analysis of its network properties with 107–1010 neurons and 1011–1015 synaptic connections is still largely unresolved . In the brain , activity of a single neuron encodes relatively little information; instead , that is achieved via population coding , through spatially distributed temporal activity patterns of cell assemblies . This contrasts with packet-switching information technology ( IT ) networks , which encode information directly into the packets and the network merely ensures routing between any two nodes . Since the spatiotemporal activity of cell populations is strongly determined by their connectivity and physical layout , cortical network structure and its spatial embedding play a significant role in the brain’s processing algorithm , in sharp contrast with IT networks . A purely bottom-up approach to deriving global brain function from local circuitry is currently intractable [2] . In contrast , a meso-scale approach is more feasible , focusing on the network of interactions between the elements of a mosaic of distinct areas representing the loci of function-specific computation ( visual , auditory , somatosensory , motor , etc . ) . As the mammalian brain is shaped by evolution , morphological and areal network level inter-species comparisons will help identify those features that are conserved across species from those that are species-specific . This will lead to a better understanding of network structural properties and provide valuable clues to the evolution of brain function [3] . However , progress in this direction has been hindered due to the absence of ( i ) the necessary data to address the physical properties of the network between areas and ( ii ) adequate theoretical network comparison methods . Published connectivity maps using consistent interareal tract tracing studies , first in the macaque [4] and more recently in the mouse [5 , 6] , allow consideration of the network as a directed , spatially embedded and weighted graph ( weights representing neuronal connection densities projecting between areas ) . The absence of full homology between the nodes ( areas ) and edges ( projections ) of the networks of the two species makes it difficult to determine commonalities and similarities between them . However , if generic , global organizational principles exist ( constraining the adaptation and growth of cortical connections in similar ways ) , then we expect to see similarities at the statistical level between the network features in the two species . Here we show that the cortical networks in the macaque and the mouse in fact do exhibit a common organizational principle despite their very different evolutionary trajectories and large differences in brain size . Supplemented by partial tract tracing data in the microcebus ( the mouse lemur ) we suggest that this principle and the associated network model is a universal determinant of the interareal network across mammals , allowing tentative predictions for the human brain . Expansion of the cerebral cortex is accompanied by an increase in the proportion of white matter relative to brain size [7–10] . However , this increase is not rapid enough to maintain a constant neuronal connection density ( defined as the fraction of neuron-to-neuron connections compared to all possible ones ) . Thus , an increase in brain size is expected to result in a reduction in the long-distance connectedness of cortical areas [11–14] . The reduction of the fraction of connections with cortical expansion and the minimization of the metabolic costs are important design features of the cortex [4 , 15–26] . One can hypothesize that this wire minimization constitutes a critical constraint for the optimal placement of areas in the cortex , serving to increase communication efficiency in larger brains [11 , 27–29] , and is supported by recent evidence suggesting reduction of long-distance connectivity with increases in brain size [28] . Recent retrograde tract tracing data in macaque [30] provides supporting evidence precisely of such a wiring constraint , in the form of an exponential decay of the wiring probability p ( d ) with projection distance d: p ( d ) ~e−λd , with a decay length ( ~1/λ ) that is short relative to hemispheric dimensions ( in the macaque λ ≅ 0 . 19 mm−1 , corresponding to a decay length of 1λ≅5 . 2 mm ) . A simple way to think of the decay length 1λ is that every increase by 1λ in projection length leads to a decrease in the number of projections by a factor of 1e≅0 . 37 ( i . e . , 37% ) . Note that using the base of the natural logarithm is convenient , as in this case 1λ is equal to the average projection length , providing a simple , intuitive interpretation . We refer to this decay property of connection density with distance as the Exponential Distance Rule ( EDR ) . Retrograde labeling using fluorescent tracers ( see Materials and Methods section ) is an accurate labeling method that reveals all incoming connections j→i to an injected ( target ) area i by labeling the cell bodies of the neurons in source area j whose axons make connections in area i . Importantly , there is no transneuronal labeling , so the retrograde labeling method used yields only one-step incoming connections to the injected nodes of the network . Note that the EDR is purely a property of the distribution of the physical lengths of individual axons , without regard to any network topological structure . The EDR states that there are many fewer long-range axons than short ones and quantifies this: the number of axons of length d that we find in the cortex is proportional to e−λd . In general , to experimentally establish the EDR , we do not need to work with brain areas as nodes of a network; we only need to be able to count neurons and measure the corresponding axon lengths . In this sense , the EDR is a more basic and general property than the description of cortical connectivity as a network at some coarse-grained ( e . g . , mesoscale ) level . Once the level of description is defined ( e . g . , areal ) , the network properties are , however , consequences of the distribution of the axonal lengths connecting the vertices . Since connectomes are embedded in physical space , the EDR property effectively constrains the topological structures that connectomes can form across different levels , ranging from the single neuron to the areal level [31] . In addition to the discovery of the EDR in the macaque , the consistency and completeness of this tract tracing data [32] has led to a deeper insight into the interareal network properties of the macaque cortex [30 , 33]: it revealed a much denser ( ρ = 0 . 66 ) interareal cortical graph than previously reported ( network density is defined as ρ = MN ( N−1 ) , where N is the number of areas and M is the number of connected ordered area pairs , see glossary ) . High density graphs have low specificity at the binary level ( areas connected or not ) , so that what distinguishes one area from another is the particular combination of areas it is connected to , combined with the weights of the connections , i . e . , their connectivity profile or fingerprint [33–36] . Because the range of weights spans many orders of magnitude ( five in the macaque ) , the specificity of individual connectivity profiles is actually very high [5 , 30 , 37] .
To what extent does the EDR , as a connectivity constraint , determine the properties of the interareal network ? To address this issue , one needs ( i ) a family of EDR-based network models and ( ii ) a method of comparison between the model-generated networks and the experimental data network . The exponential decay rule ~e−λd in the macaque was obtained from collating all the labeled neurons ( over 6 . 4 million ) following tract tracing experiments in different areas and constructing an interareal distance matrix , the latter estimated as the distances between the area barycenters through the white matter ( WM ) , along the shortest paths . Here axonal p ( d ) should be interpreted as an average property ( see Fig 1A ) , the probability that an axonal bundle projects to a distance d , independently of the specific functional nature of the areas . At this level of description , the strength of the connection between areas , expressed as the fraction of labeled neurons ( FLN ) , depends uniquely on their geometrical separation . Thus , the network is viewed as a spatial , directed , and weighted graph dependent on the matrix D = {dij} of interareal distances dij . We emphasize here that the EDR arises from the estimated probability distribution of axon lengths . Although the strength-distance relation is consistent with the EDR , the probability distribution of axons lengths provides a more compelling demonstration of the property and leads naturally to the parametric EDR model described below . The probability density function , q ( d ) , of the distances in the matrix D is typically a unimodal distribution ( Fig 1B ) , which , when combined with the exponential decay p ( d ) , leads to a log-normal distribution of edge weights , confirmed by the empirical FLN data [4 , 39] . The EDR distribution with the corresponding distance matrix D in a given brain naturally defines a parametric family of random graphs , called EDR random graphs ( Fig 1C ) , parameterized by the decay rate λ . For these model graphs we make the choice p ( d ) = λe−λd , where now λ is the ( only ) model parameter . To distinguish the decay rate parameters in these models from the experimentally measured ones , we denote the latter as λexp , e . g . , for macaque λexpmac = 0 . 19 mm−1 . We also employ , as a null model , the constant distance rule ( CDR ) family of random graphs , where there is no dependence of connection probability on distance , corresponding to the λ→0 limit , i . e . , to the choice p ( d ) = const . The EDR family of random graphs is defined via a simple algorithm [4] in the spirit of the Maximum Entropy Principle , i . e . , it is based only on the given information ( p ( d ) and D ) , while all else is uniformly random . The algorithm proceeds as follows: First , we randomly draw a connection length d from the distribution p ( d ) . Second , we choose uniformly at random an area pair whose separation distance in the matrix D falls in the same distance bin as d , according to some binning criterion ( bin sizes used in this study were typically 5 mm for the macaque and 0 . 4 mm for the mouse ) and finally , insert a randomly oriented connection between them . Multiple connections between the same area pair in the same direction generate the weights for the directed edges with a log-normal distribution . These steps are then iterated until the graph density in the model reaches the observed value in the experimental network . We denote the data network obtained from the experiments by Gexp ( e . g . , for the macaque we use Gexpmac , and for the mouse Gexpmac ) . Our goal is to compare the properties of the EDR model networks with the properties of Gexp . Since the model networks are only based on distance-dependent connection probabilities , one cannot expect perfect agreement ( edge-by-edge ) with the biological connectivity graph Gexp , however , if the distance rule is a strong determinant of the interareal network , the model graphs should be statistically similar to Gexp . The comparisons are performed via parameter matching of network properties [4]: for a given network property P , the interareal distance matrix D and parameter λ is used to generate a large ensemble of EDR graphs GEDR ( λ ) . By varying λ we determine the value λP via minimizing the deviation |P ( Gexp ) − 〈P ( λ ) 〉| , with the average 〈∙〉 taken over at least 103 EDR graph realizations from GEDR ( λ ) . Thus the model parameter is determined so that the average of P in the model is as close as possible with the value of P observed in the data network . We then compare the fitted value λP with λexp , the decay rate obtained directly from the experiments . If the two are close , then the EDR is a strong determinant for the measure P of the cortical network . Thus , the extent a particular measure in the EDR model and in the data network agree , i . e . , |P ( Gexp ) − 〈P ( λP ) 〉| with respect to the same comparison with the CDR model , i . e . , with |P ( Gexp ) − 〈P ( λ = 0 ) 〉| , expresses the degree to which the EDR influences that particular measure in the cortical network . This analysis is repeated with several local and global network measures . The more measures for which there is an agreement between λP and λexp , the stronger the effect of the EDR in shaping the interareal network . This method also has the added advantage of identifying those network properties that are not well described by the EDR , and thus , based on the nature of these measures , providing us with clues for additional network mechanisms . In the macaque , the EDR model predicts very well many local , global and weighted network properties of the interareal network ( see [4] for details ) , and thus it is a strong determinant for the large-scale network organization of the macaque cortex . It also captures its pronounced core-periphery organization ( i . e . , a densely connected set of areas—core , with feedback and feedforward links to/from a more loosely connected set of peripheral areas ) , with the core strongly dominated by associative areas [4 , 40] . The EDR network model of cortical connectivity represents a radical departure from previous , purely topological models of cortical networks , which do not take into account their physical , i . e . , weighted and spatially embedded nature , and this has now been well documented in the recent literature [41 , 42] . The spatial clustering and geometrical positioning of the nodes in the EDR model in the macaque is observed to strongly echo the functional layout of the cortex as revealed by numerous physiological and anatomical studies [36 , 43] . To determine whether a similar description is valid for the mouse ( Mus musculus ) cortex , we first conducted retrograde tracer experiments in the mouse neocortex ( S2 and S3 Figs ) , in order to determine the projection length distribution p ( d ) , which , indeed , shows a clear exponential decay ( Fig 2A ) . The decay rate , λexpmus , was determined from an exponential fit as λexpmus = 0 . 78 mm−1 , with a 95% confidence interval of ( 0 . 72 , 0 . 83 ) ( see Fig 2A , and inset ) . This exponential decay is to be compared with the same distribution for the macaque from Fig 2B in Ref [4] ( see Table 1 for the λ parameter estimates ) . The distance matrix Dmus was determined from flattened cortex measurements . The corresponding distance distribution q ( d ) is unimodal , as shown in Fig 2B , which is to be compared with the same distribution for the macaque from Fig 2C in reference [4] ( the consequences of the differences in these distributions are discussed in more detail in the section “Functional Layout in Terms of Spatial Clustering of Cortical Areas” ) . We then applied the network analysis described in the section above to the largest available edge-complete graph ( the status of connectivity between all pairs of nodes is known ) of 33 areas in one hemisphere of the mouse neocortex [5 , 6] , denoted by Gexpmus from here on ( S1 Fig ) . This mouse dataset contains 719 directed pathways and has an interareal network density of ρmus = 0 . 68 , similar to that reported in the macaque . Fig 3 shows the proximity of λPmus obtained using the parameter matching method to the decay rate λexpmus for several network measures including the number of area pairs connected uni- ( M1 ) or bidirectionally ( M2 ) , 3-motifs , clique distributions and the second largest eigenvalue of the symmetrized form AAT of the adjacency matrix A . These measures have been selected in part because they probe graph properties from local to global scales , and are of varying complexity . Additionally , these measures ( see glossary for definitions ) , and in particular the deviations from their values in random graphs carry functional significance: unidirectionally connected areas depict an asymmetric role in information processing ( driver versus driven nodes ) , the 3-motifs have extensively been studied as building blocks of functional organization in complex networks [38] , cliques identify maximally connected network regions usually representing activity-specific strongly correlated communities or clusters , and the second largest eigenvalue is related to the rate of spreading processes ( e . g . , epidemics or information ) on the network [4] . The different comparisons and fits based on these measures are highly consistent and indicate λ to be in the range 0 . 78–0 . 93 mm−1 ( purple vertical band in Fig 3A–3D ) . The broader range in mouse of |λP−λexpmus| compared to macaque might be due in part to the fact that the mouse connectivity matrix was generated by anterograde tracing [5 , 6] . Other statistical network properties , such as degree distributions , are likewise well captured by the EDR network model with λ = 0 . 78 mm−1 , ( see S4 Fig ) . A clique ( see glossary in S1 Text ) is a complete subgraph of a network , i . e . , it carries the maximum number of possible edges between its nodes . In dense graphs ( thus with many cliques ) the size ( number of nodes ) distribution of the cliques provides insight into the network’s heterogeneity [4] . The largest cliques in dense graphs can be used to define the cortical network core [4 , 40] . As in macaque [4 , 40] , the clique distribution analysis in the mouse ( Fig 4A ) reveals a distinct core-periphery structure . The mouse connectome , Gexpmus , includes a dense core of 12 nodes organized into the two largest cliques each of size 11 , plus a periphery of 21 nodes . There are a total of Mcc = 131 links within the core , Mcp = 190 links from the core to the periphery , Mpc = 170 from periphery to core , and Mpp = 228 links within the periphery . Densities for the mouse are the following: core 99% ( versus 92% in macaque ) , periphery 54% ( versus 49% ) and the links between the core and periphery , 71% ( versus 54% ) . The likelihood of a core having 12 nodes in a random graph on 33 nodes with the same density ρ = 0 . 681 as in Gexpmus is vanishingly small: ( 3312 ) ( 1321 ) p131 ( 1−p ) 1 = 2 . 07×10−12 ( versus 10−17 in the macaque ) . To see how well the EDR model reproduces the clique distribution , we define a scalar deviation measure σcl ( λ ) between the clique-size distributions in the data and the EDR model as the root mean square ( RMS ) of the clique-count log-ratios . The best agreement between the two distributions is achieved at λclmus = 0 . 93 mm−1 ( Fig 3C ) and the clique distributions in the model and data are rather close at this value ( Fig 4A ) . Anatomically , the cortical core in the mouse shows significant differences with that previously reported in macaque , the most striking being that the mouse core includes portions of primary somatosensory cortex ( SSp-ll and SSp-tr ) and primary motor cortex ( MOp ) ( Fig 4B ) . While additional injections may well expand the core membership in macaque , primary areas in the macaque core are extremely unlikely , given the rarity of connections linking primary areas [30] . This contrasts with the mouse where the inter-primary area subgraph has a density of over 80% [5 , 6] . In agreement with the presence of primary areas in the mouse core , the two-dimensional map of the flattened cortex ( Fig 4D ) shows that the mouse cortical core might be spatially more widespread across brain regions compared to that of the macaque , where the core appears concentrated in frontal and parietal areas [4] . Note that in both mouse and macaque , the core areas have overall , higher in-degrees than non-core areas ( Fig 4B for the mouse ) . The wider spatial spread of the mouse compared to the macaque core may reflect the relative expansion in primates of higher-level association cortex with respect to the primary areas [3] . These differences in the cortical core of the mouse and macaque need to be considered in light of the proposal that in primates at least , the core is related to cognitive architectures such as the global workspace , thought to be involved in consciousness [40 , 44] . Network motifs refer to the different possible connectivity patterns of a small , fixed number of nodes . For example , in the 33-node mouse cortical network there are ( 333 ) = 5456 triplets of nodes , each of which has one of the 16 connectivity patterns shown in Fig 5A . Three-node motifs have been proposed as the building blocks of network circuits and their pattern of variation in frequency to reflect functional properties of the networks [38] . For instance , motif 10 ( oriented 3-cycle ) is significantly under-represented in the cortex , while motif 3 is significantly over-represented ( lone , bidirectional link ) when compared to a random network , in both species ( see Fig 5D ) . As in macaque [4] , the mouse EDR predicts the observed motif frequency distributions in this species significantly better than does the CDR ( Fig 5B ) . Despite the marked quantitative differences in motif distributions between mouse and macaque ( Fig 5C ) , there could , however , be qualitative similarities . Testing this requires comparing the observed motif profiles to that of a randomized null model [38] consisting of an ensemble of random networks having the same degree sequence as the data . Graphs were uniformly sampled from this ensemble by repeatedly rewiring edges [45] . Fig 5D shows how the motif counts of the empirical connectomes differ from such a randomized null model . As similar patterns are observed in both species , these findings suggest that they are part of the same class of large-scale networks with similar architectural and functional constraints . Repeating this analysis for networks generated by the EDR model ( see Fig 6 ) we find a remarkable similarity in the motif profiles not just between the networks of the real data network and the EDR model but also between the model networks of the two species ( Fig 6 ) . This confirms the existence of a common network architectural invariant in these two species . This is unexpected , insofar as the decay rates and the distance matrices are very different between the two cortices . Since the motif profiles are binary measures , these findings indicate graph structural similarity between the two brains . In order to test whether there are significant similarities in the large-scale connectomes of the two species beyond the constraints imposed by the EDR , we used the EDR model as a null model [46] . S5 Fig shows that motif counts continue to look similar between the mouse and macaque , although their similarity is now less pronounced . To further probe the wiring similarity between the mouse and macaque connectomes , we next study the connectivity similarity profile measure . Elsewhere we have demonstrated that a quantitative measure of the similarity of the connectivity profiles of target cortical areas decreases in a regular fashion with increasing distance between them , i . e . , the closer two target areas are , the more their source areas overlap [4 , 33] . We have also shown that changes in similarity reflect the functional layout of the cortex [33] , and thus it is natural to compare the behavior of this measure between the mouse and macaque . A similarity index [4] can be defined for both incoming ( in-link similarity ) and outgoing ( out-link similarity ) connections . In order to compare macaque and mouse similarity indices , we focus here on the incoming connections , as those are the ones fully specified for all the injected areas in the macaque dataset . Next , we analyze the similarity between the connectivity profiles , for all possible target area pairs . The in-link similarity index for any two ( target ) areas is a measure describing the extent to which both targets receive/or do not receive in-links from the same source areas , compared to a fully randomized state of the network ( see Materials and Methods section for details ) . Fig 7 shows the distribution of in-link similarity indices as function of the distance between all area pairs in the mouse ( Fig 7A ) and the macaque ( Fig 7B ) . In both species , in-link similarity decreases with increasing distance between the area pairs , i . e . , areas that are further apart on the cortical sheet have increasingly dissimilar in-link connectivity profiles on average , while the opposite is true for areas that are closer to one another . The colored regions in both Fig 7A and 7B are the probability densities of in-link similarity indices generated by the corresponding EDR models , with red corresponding to higher , and blue to lower probabilities; in both cases the EDR model captures the average behavior rather well . In order to compare distance-dependent quantities between brains of very different sizes , all distances are rescaled by the average interareal distance in each species ( 〈d〉mus = 4 . 54 mm and 〈d〉mac = 26 . 35 mm ) . Interestingly , as the largest distances are dmaxmus = 10 . 1 mm and dmaxmac = 58 . 2 mm respectively , this fits both brains onto the same adimensional template , as dmaxmusdmus = 2 . 22 mm and dmaxmacdmac = 2 . 21 mm . Fig 7C shows the corresponding distribution of adimensional distances q ( d/〈d〉 ) . When plotting the in-link similarity indices against the rescaled distances ( Fig 7D ) , we find a remarkable overlap between the clouds of points in the two species . This is rather surprising given the fact that they have very different decay rates λ . They also have rather different interareal distance matrices as the macaque cortex is folded , resulting in it having a more peaked distance distribution than the mouse ( Fig 7C ) . Fig 8 shows the sensitivity of the in-link similarity indices using the EDR models in both the mouse ( panels 8A–8D ) and the macaque ( 8E–8H ) . For a given distance matrix , the point clouds are observed to rotate as a function of λ in both species , and hence there is no a priori reason for the overlap in Fig 7D . This overlap , however , is an indication of the existence of a network architectural invariant , present in both species , also reflected in the motif profiles discussed earlier . Further explanation for the significant , overall overlap between the similarity distributions for the two species is provided in the Discussion section . With the help of the common adimensional template defined above we now discuss species-specific characteristics in our comparison of cortical networks . The EDR decay p ( d ) can simply be recast in terms of adimensional distances , by writing ( d ) ~e−λd = e−γ dd , where γ = λ〈d〉 is the adimensional ( or normalized ) decay rate . Accordingly , γmus = 0 . 78 × 4 . 54 = 3 . 54 and γmac = 0 . 19 × 26 . 35 = 5 , showing that on the common template , the mouse has a shallower connectivity decay than the macaque . The distribution of distances in the mouse is broader compared to the macaque ( Fig 7C ) , which when coupled with the shallower connectivity decay contributes to the mouse cortex experiencing a less constraining effect of the EDR than does the macaque . This difference in the EDR between the two species explains some of the differences in the functional layout of the cortex in mouse and macaque . In Fig 9A and 9B we show the same similarity indices for all area pairs as before but also indicate which area pairs are connected ( black circles ) and which are not ( white circles ) and provide smooth estimates ( colored regions ) of connection probability as a function of similarity and adimensional distance . Comparing Fig 9A and 9B we see that in macaque , spatially clustered , presumably functionally related neighboring areas are heavily interconnected and share similar connectivity profiles , while more distant areas show weaker probability of connectivity and similarity index . This relationship between probability of connectivity , spatial separation and similarity is , however , weaker in the mouse . In both species , connection probability changes as a function of distance . Fig 9C and 9D show how this relationship differs in the two species . Consistent with a steeper EDR in the macaque , neighboring areas exhibit 100% connectedness , and the probability of connections ( density ) decreases smoothly and consistently with distance to around 10% density at maximum distances [47] . This contrasts with the mouse , in which neighboring areas do not quite reach densities of 100% and widely separated areas have densities in the region of 50% to 80% ( Fig 9C ) . Hence , these results show that compared to macaque , in the mouse , widely separated areas are more likely to be interconnected . These differences in the probability of being connected as a function of distance between the two species appear highly significant ( smooth curves in Fig 9C and 9D ) . Numerous studies point to the cost of long-distance connections as an inherent design challenge associated with differences in brain size [48] . One way to define total wire length is: Λ = Σi , jAijDij , where A denotes the binary adjacency matrix and D is the interareal distance matrix . Yoked permutations of the rows and columns of the adjacency matrix reassign the distances to each pair of areas while maintaining the connectivity unchanged . As in macaque [4] , the total wire length of the mouse inter-areal network is significantly shorter than a random permutation of the areas ( S6 Fig ) . Simulated annealing methods [4] showed that optimization of area placement can lead to a 12% reduction in total wire length in the mouse , significantly higher than the 5% reduction obtained in macaque [4] . Next , we address the strength of connections with the expectation that long-range connection strengths ( expressed as FLNs ) would decrease in the larger brain . Due to the EDR , the FLN clearly decreases with distance . Distinguishing interareal association and canonical connections allows an improved understanding of the effect of distance on connection weight ( for definition of associative and canonical connections see [3] ) ( Fig 9E and 9F ) . This suggests that the decline in FLN is steeper in canonical cortex compared to association cortex , so that the long-distance association cortex connections are one to two orders of magnitude stronger than the connections between canonical cortex areas with the same separation ( see [3] ) . However , the results suggest that the decline in weight with distance is steeper in the macaque compared to the mouse . Together these findings show that compared to the larger macaque cortex , in the smaller mouse brain long-distance binary interareal connections are marginally more numerous . By contrast there is a highly significant increase in the weight of the long-distance connections in the mouse , and this species difference is more pronounced in the projections of association than in the canonical connections of the primary areas .
The present meso-scale network investigation of the neocortex , with appropriate network comparisons , provides detailed information on a common organizational principle that explains numerous network features in two widely separated species , with distinct evolutionary histories . Based on phylogenic considerations , and the fact that evolution is essentially a tinkerer [49] , one expects to find evolutionarily preserved features embedded in these networks , i . e . , architectural invariants . Evolutionarily preserved features , however , often are expected to manifest themselves as organizational principles tied to biophysical constraints . The success of the mammalian class includes adaptation to diverse habitats and lifestyles , which is in part attributed to the behavioral flexibility ensured by the neocortex [50] . The modulation of corticogenesis [51] has led to extant mammals exhibiting a five-orders of magnitude range of brain size [52] , going from small-brained mammals that include miniaturization of ancestral forms to the expansion and additional arealization that characterize primates , especially humans . The present results suggest that the EDR plays a key role across the mammalian order to optimize the layout of the inter-areal cortical network allowing larger-brained animals to maintain communication efficiencies combined with increased neuron numbers . Our results indicate that the EDR and the associated network model provide a unifying framework to capture common network properties but also some of the differences across the mammalian branch and thus allow network comparisons between species . The EDR decay rate λ and cortical geometry ( interareal distances ) significantly impact on the structural heterogeneity of the cortical network with important consequences for the general functional layout and core-periphery structure , that we speculate , could be involved in higher cognitive processes [40] . The limitation of the EDR model stems from the fact that the EDR describes an overall , or average property . At this level , without additional determinants , it should not be used as a generative model of individual connections as we have emphasized elsewhere [4 , 40] . If we plot the decay of the probability of connections for several target areas , as shown in Fig 10A for the mouse , we find significant variability . The black line in Fig 10A , the average decay , is the same as that in Fig 2A . The fluctuations for a given target , however , are not noise , but rather they are part of a signal . This we illustrate in macaque: Fig 10B shows the consistency of fluctuations following repeat injections in area V1 , in five different individuals . There are numerous factors that one might need to take into account to better understand this variability . For example one may need to consider the observed systematic variation in neuron numbers across the cortex [53 , 54] , the anisotropy of axon outgrowth distributions [55] and possibly diverse developmental factors [56] . Overall , however , these considerations emphasize that the EDR network serves as a framework , upon which other details are imposed . Note that in order to assess the ability of the EDR model ( or any connectome model ) to reproduce properties of empirical network data , it is crucial that the data is as edge-complete as possible , i . e . , that the connectivity between any two nodes is known . Otherwise the lack of fit between model and data cannot be used to discard the model [57] . This holds for two reasons: ( a ) the EDR network model produces complete connectivity information between its nodes , it cannot generate “untested” connections , by default , and ( b ) many network measures can be sensitive to the absence or presence of an even a small fraction of connections in the network . It is also important to emphasize the roles of cortical geometry [58] and that of areal segmentation in shaping the network properties of the connectome . Since the connection probability depends on distance , network properties are influenced by the relative proximity of areas . In turn , the strength of connections between functionally defined areas correlate with the amount of signaling activity between them and therefore with their functional roles within the information processing hierarchy in the brain . Ad-hoc segmentations , however , will generate ad-hoc distance matrices for the EDR model , and accordingly , the model networks would no longer be interpretable from a functional circuitry point of view , and in this sense , it is important to use optimally defined functional parcellation of the cortex . Our comparative analysis of motifs and connectivity similarity indices demonstrates the existence of network architectural invariants , which in turn imply that the EDR parameter λ and areal positioning ( geometry ) are not independent parameters: while both change during evolution , the changes are orchestrated in such a way as to ensure that certain network/circuitry properties are preserved . As argued in the introduction , the network , i . e . , the graph connectivity ( form ) must play a significant role in the information processing algorithm itself ( function ) , and thus these network invariants are a reflection of common processing dynamics in the cortex . Our use of a normalized or adimensional distances facilitates comparisons across brains of different sizes . Fig 10C shows directly the fingerprint of such universal principles in neocortical organization: it shows the connection probability decay on the adimensional template brain from a common target area ( area V1 ) in macaque ( data from reference [30] ) , mouse as well as microcebus . At short to medium distances where the vast majority of neurons are located , decays are identical , but are observed to change in a species dependent fashion for the long-range connections . Microcebus belongs to a group that contains the smallest existing primates , with a brain under 2 cm in length . Although the microcebus data is only for V1 , it remarkably fits to the same adimensional template , with a decay rate λ between that of mouse and macaque , suggesting that the quantitative differences that distinguish the species are due to both brain size , and primate-rodent differences . The EDR could be the expression of the consequence of a universal information processing principle implemented in the cortex across several scales , specifically to include single neurons in the local circuit , which present over 80% of the cortical connectivity [31 , 39] . Hence , the two major ingredients for the EDR are found in the local circuitry , the log normal distribution of synaptic weights [59] and an exponential decay of connection distances as reported here . Further , the experimental evidence presented in Fig 11 , shows that p ( d ) follows a nearly identical , exponential decay out to within 1 . 5 mm for both mouse and macaque , with λexplocal≅4 . 54±0 . 08 mm−1 . These are gray matter , non-myelinated connections , and are observed to have a very different decay rate than white matter connections . Thus , at least in area V1 , the decay of connectivity with distance seems to behave in a very similar fashion in both mouse and macaque , and therefore surprisingly the decay rate in the gray matter does not appear to be related to brain size . Using the reported data in [59] for the rat visual cortex obtained from quadruple whole cell recordings , the local decay rate in the rat can be determined to be λexplocal≅4 . 96 mm−1 , a value consistent with the one found above in mouse and macaque , above . Table 1 summarizes the EDR related parameters in the mouse and macaque , for both white matter and gray matter connections . The universal character of the EDR is further supported by mathematical arguments . The exponential distribution ( EDR ) is memoryless ( Markov property ) , i . e . , in our case , the probability that an axon of some length grows by an additional amount is independent of its current length ( within cutoff limits ) . In this way , it has the property that f ( d+l ) = f ( d ) f ( l ) , where f ( l ) = ∫l∞p ( l′ ) dl′ = e−λl is the probability of an axon growing to a length beyond l . The exponential distribution is the only continuous distribution with this property [60]; for all other distributions , growth depends on the current length , i . e . on past growth history . This also implies that the EDR is the maximum entropy probability distribution for axonal lengths with given expectation value ( = 1/λ ) , see [61] . These properties are evolutionarily advantageous , conferring maximum adaptability during cortical expansion . Moreover , as more neurons are added , the probability distribution of the shortest connection among an arbitrary number of other connections also obeys an exponential distribution [62] , making the EDR an invariant property locally as well , supported by the experimental data quoted above . The present findings could have important consequences for understanding the human brain . The recognized limitations of current tractographic analysis of diffusion MRI data [63 , 64] , means that direct observation of long-distance connections in the human brain is not presently feasible . Given the specificity of long-range cortico-cortical connectivity [47] , this technical limitation has important consequences for understanding the human connectome , and we believe that comparative connectomics as developed in the present study will be a necessary step for determining universal principles of cortical connectivity . Hence , an in-depth understanding of the influence of changes in brain size will play an important role in better understanding the human brain . Since the EDR leads to a decrease in the strength of long-range connections in macaque compared to mouse , we may hypothesize that increase in brain size leads to increased reductions of weight in long-range projections for the whole mammalian branch . In the human brain the small number of fibers in such long distance connections will pose an acute problem for detection for some time . This could constitute an important limitation . For example , one could speculate that the low weight of human long-range connections may contribute to an increased susceptibility to disconnection syndromes , such as have been proposed for Alzheimer disease and schizophrenia [65–67] .
Experiments were performed in male and female PV-Cre [68] ( Jax: 008069 ) , x Ai9 reporter mice ( Jax: 007905 ) , harboring the loxP-flanked STOP cassette , which prevented the transcription of the tdTomato protein driven by the chicken β-actin ( CAG ) promoter [69] . The crossing produced Cre-mediated recombination , which resulted in the expression of the red fluorescent protein in the subset of parvalbumin ( PV ) -positive GABAergic neurons . All experimental procedures were approved by the institutional Animal Care and Use Committee at Washington University and conformed to the National Institutes of Health guidelines . Injections were made in Microcebus murinus in area 10 and area V1 . Surgical and experimental procedures were in accordance with European requirements 2010/63/UE and approved by the ethics committee CELYNE ( ref 00439 . 02 ) . For tracer injections , mice were anesthetized with of a mixture of Ketamine ( 86 mg · kg−1 ) and Xylazine ( 13 mg · kg−1 , i . p ) and secured in a head holder . The body temperature was maintained at 37°C . Intracortical connections within the left hemisphere were retrogradely labeled by inserting a glass pipette ( 20 μm tip diameter ) into the brain and injecting Diamidino Yellow ( 50 nl , 2% in H2O; EMS-Chemie , Gross-Umstadt , Germany ) by pressure ( Picospritzer , Parker-Hannafin ) . Injections were performed stereotaxically 0 . 35 mm below the pial surface , using a coordinate system whose origin was the intersection between the midline and a perpendicular line drawn from the anterior border of the transverse sinus at the posterior pole of the occipital cortex . The injections were made in the following areas: V1 , RL , AL , LM , P , RSD , ACAd , MOs , SSp-bfd , SSs , Au . Area AM was injected twice ( S3 Fig ) . The parcellation ( names and locations of the areas ) is based on Wang et al . [70] ( S2 Fig ) but differs from those used in Fig 4D and associated analyses . Four days after the tracer injection , mice were deeply anesthetized with an overdose of Ketamine/Xylazine and perfused through the heart with phosphate buffered saline , followed by 1% paraformaldehyde ( PFA ) in 0 . 1 M phosphate buffer ( PB , pH 7 . 4 ) . Immediately after , the cortex was dissected from the rest of the brain , completely unfolded , flat-mounted and post fixed overnight in 4% PFA at 4°C . Next , the tissue was cryoprotected in 30% sucrose and cut at 40 μm on a freezing microtome in the tangential plane . To survey the injection site and the distribution of labeled neurons across cortical areas , sections were wet-mounted in PB and imaged in St . Louis under a dissection scope equipped for UV- and red-fluorescence illumination . For plotting DY labeled neurons , the sections were permanently mounted onto glass slides and stored at 4°C . The distribution of DY-labeled neurons was analyzed in Bron . Plots of DY neurons were made at 20× under a fluorescence microscope equipped for UV illumination ( excitation: 387–398 nm , emission: 435–475 nm ) , using the Mercator software package running on ExploraNova technology . Labeled neurons were contained in 12–16 sections per hemisphere . Digital charts of the coordinates of DY labeled neurons across each section were stored in the computer . Next , the regional pattern in the density of PVtdT expression was imaged under fluorescence optics . Finally , the sections were stained for Nissl substance , imaged under bright field illumination and superimposed onto the digital maps of DY and PVtdT fluorescence . In Bron , all the images were acquired using MorphoStrider software ( ExploraNova ) . The digital charts were saved in PDF files and were scaled in Adobe Illustrator . The charts and the corresponding images were brought to a common scale , allowing reconstruction of the sections . Sections were stacked in order , and then aligned . The landmark for the alignment of the sections was the injection site , followed by rotation around this point , allowing a 3-D reconstruction of the flattened brain . The injected area was delimited , as were the borders of the neocortex . Automated processing was performed using in-house software , written in Python . For each case , the positions of labeled neurons inside neocortex , but outside the injected area ( i . e . , extrinsic neurons ) were extracted in digital format , The fraction of labeled neurons per area ( FLN ) was estimated as the number of labelled neurons extrinsic to the injected area expressed as a fraction of the total number of labeled neurons in the cortical hemisphere [39] . Unlike the template matching procedure used in previous studies [5 , 6] we parcellated each cortex individually based on multiple markers expressed across different tangential sections . In a stepwise procedure we first used density differences in the expression of PV-tdT- labeled cell bodies and processes to delineate borders of single areas such as V1 , S1 , S2 , Au , PD , UF , PV , GU , ORBI , MM . RSD , MOp , MOs , and ENTm ( S2 Fig ) . Next , the PVtdT-expression pattern was used to outline regions which from previous studies are known to include multiple areas . The list includes ( 1 ) LM , LI , P , POR , 36p; ( 2 ) AL , LLA , RL , A , AM , PM; ( 3 ) TEp , TEa , ECT , PERI; ( 4 ) AIp , AId , AIv; and ( 5 ) ACAd , ACAv , PL , ILA , ORBm , FRP ( S2 Fig ) . Each of these regions was further partitioned into areas based on the topographical distribution of DY labeled neurons , the size and location relative to readily identifiable areas , the rhinal sulcus , the crest of the medial wall [70–74] and the cytoarchitecture revealed by Nissl staining [5 , 75 , 76] . The segmentation shown in Fig 4D was carried out in Adobe Illustrator , combining Allen Brain Atlas boundary criteria ( visualized with Brain Explorer 2 ) with photos of PVtdT and Nissl staining , for one section of the flattened mouse brain . The contours of the cortical areas are non-self-intersecting closed polygons; therefore , computing their centroids is straightforward . The distances between areas were considered as the distances between their respective centroids . The Allen Brain Institute ( ABI ) dataset was collected on their website , offered as link in their original research publication [5] . The University of Southern California ( USC ) matrix was , on the other hand , extracted directly from their original article [6] . The ABI mouse atlas possesses 40 isocortical areas according to their Supplemental Table 1 . Out of these 40 areas , 2 did not correspond to any line or column in the data as structured in S1 Fig ( i . e . the connectivity matrix ) . An additional 4 areas were not considered as primary target of an injection by the authors , leading to our decision to exclude them from our analysis . We then extracted from the ABI data a 34 × 34 weighted and directed connectivity matrix . The larger USC dataset has a finer grained parcellation than that of ABI , although based on the same fundamental scheme . We contracted the USC final matrix down to a level of 42 × 42 by merging areas together in both rows and columns so as to obtain a squared matrix . At this point , the ABI and USC matrices had 33 areas in common , which corresponds to 97% and 79% of their full respective matrices . A similar parcellation scheme was extracted out of the two datasets , allowing complete , connection-by-connection comparison between the two matrices , see S1 Fig for the final connectivity matrix . The database in the mouse has been generated following tracer injections in all cortical areas . The macaque data , however , was obtained from 29 injections using a 91-area atlas . Because in macaque we are using an edge-complete subgraph , the statistical features are predicted to reflect those of the , as yet unavailable , fully connected graph . However , the presently available dataset cannot give complete information on detailed areal relationships , such as for example the full membership of the cortical core . Fig 7C shows the histograms of connection distances for mouse and macaque after normalization by the mean distance for each species . By construction , both distributions have mean equal to 1 and can be reasonably well described by truncated normal distributions . When fitting the distributions by maximum likelihood using functions from the truncnorm package [77] in R [78] , the variance of the macaque normalized distances appears smaller than for the mouse data with a ratio of 0 . 608 . Is this significant ? First , we examine this question with an F-test on the ratio of variances . The test is two-sided because we do not specify a priori which variance is greater . This is a more conservative approach . The F-statistic is the variance ratio with degrees of freedom ( 405 , 527 ) giving a highly significant p = 1 . 58 × 10−7 . The test assumes normality , however . To verify the conclusion , then , we performed a permutation test that does not make the normality assumption [79] . In short , we permute the macaque and mouse labels a large number of times and recomputed the variance ratio for each new permutation . Under the null hypothesis that both distance distributions are the same , we expect a large number of variance ratio estimates on the permuted datasets that are more extreme than the variance ratio computed on the data . The p-value is computed from the proportion of ratio estimates more extreme or equal to the obtained value . For ratios , the definition of more extreme is based on the values that are less than the estimate and greater than its reciprocal . We include the ratio estimate from the dataset in the distribution of permutation estimates . S7 Fig shows the value of the ratio of variances for 100 , 000 permutations of the two datasets . The vertical line indicates the value obtained from the data , which is lower than all of the other values of the permutation distribution , indicating that the obtained ratio is highly unlikely under the hypothesis that both distributions are the same . The p-value is indicated in the graph . The p-value is smaller than 10−5 , which is the resolution of the test for 100 , 000 permutations . Thus , the width of the distribution of distances for macaque is significantly narrower than that for mouse . To analyze the density of connectivity with distance , we estimate the probability of a connection with distance . This can be done with a logistic regression . By performing the analysis on the binary connectivity ( that is , presence/absence of a connection ) at each distance , no binning is involved . Standard logistic regression implemented via a Generalized Linear Model with a binomial family [80] specifies that the expected value of the connection probability is related to a linear predictor through a link function that is often taken to be the log of the odds ratio or logit function . The model fit would be g ( E ( Y = 1 ) ) = β0 , i+β1 , iDistance where Y is a binary variable indicating whether a connection is present between two areas , g is the link function , here log ( p/ ( 1 − p ) ) with p the expected value or probability of a connection , and β0 , i and β1 , i are intercept and slope , respectively , of the linear predictor , with i varying with the species . There is no a priori reason to suppose , however , that the sigmoid function of distance that this model implies will provide an adequate description of the change in probability with distance . To allow for a more flexible description of this relation , we fit the data with a Generalized Additive Model ( GAM ) using a binomial family [81] . The GAM model is given by g ( E ( Y = 1 ) ) = Imousefmouse ( Distance ) +Imacaquefmacaque ( Distance ) where fi are smooth functions of the covariates constructed from sums of spline curves with increasing complexity and Ii are indicator variables taking on the value of 1 for i = mouse ( or , respectively , macaque ) and 0 otherwise . The complexity ( or wiggliness ) of the fitted model is controlled by including a penalty in the fitting criterion based on the integrated square of the second derivatives of the f’s . The choice of degree of penalization ( or smoothness ) is controlled by minimizing a criterion related to prediction error ( i . e . , fitting some of the data and calculating the error on the remaining portion ) called the un-biased risk estimator ( UBRE ) that is closely related to Aikake’s Information Criterion ( AIC ) . Like AIC , UBRE favors a model that maximizes the predictability of future rather than the actual data and serves to minimize the tendency to overfit the data . The fits were performed with functions from the mgcv package [81] in R [78] . The estimates of the smooth curves for macaque and mouse are plotted in Fig 9C and 9D for macaque and mouse , respectively , with twice the estimated standard errors of the fits . To estimate the significance of the differences between the two estimates , we also fit the simpler nested model in which a single smooth curve described connectivity dependence with distance for both species . A likelihood ratio test of the nested models gave a χ2 ( 2 . 26 ) = 54 . 04 with p = 3 . 05 × 10−12 , strongly supporting that the differences in the curves are significant . Note that the generalization of degrees of freedom in the case of GAM fits are not necessarily integer valued . We extended this analysis to consider the connection probability as a smooth function of both distance and similarity . The GAM framework is used again but now to model surfaces of two variables , here giving the log-odds ratio of the connection probabilities as function of similarity and normalized distance . The model is given as g ( E ( Y = 1 ) ) = Imousefmouse ( Distance , Similarity ) +Imacaquefmacaque ( Distance , Similarity ) , where fi are now smooth 2D functions of the covariates constructed from sums of spline surfaces with increasing complexity and Ii , as before , are indicator variables taking the value of 1 for i = mouse ( or , respectively , macaque ) and 0 otherwise . Contour plots of the estimates of the connection probability as a function of normalized distance and similarity are shown in Fig 9A and 9B . The color gradient indicates connection probability , passing from high ( yellow , near 1 ) to low ( green , near 0 ) connection probability . The curves indicate estimates of contours of constant connection probability ( notated on the curves as probability values ) . To evaluate the significance of the species difference displayed in Fig 9A and 9B , we also fit the simpler nested model in which a single smooth surface described connectivity dependence with distance and similarity for both species . A likelihood ratio test of the nested models gave a χ2 ( 8 . 4 ) = 61 . 6 with p = 3 . 5 × 10−10 , strongly supporting that the difference in the surfaces are significant . Note that as above the generalization of degrees of freedom in the case of GAM fits are not necessarily integer valued . The method used to compute binary similarity indices with macaque data has been described previously [4] . Our published macaque database is made of 29 injected areas for a 91 parcellation scheme , thus giving a 91 × 29 connectivity matrix . In this context , only the in-degrees of injected areas are completely known , the out-degrees of source areas remaining incomplete . Therefore , if one wants to compare macaque and mouse using a degree-based binary similarity measure , one has to restrict oneself to in-degrees , in order to use complete data . For this reason , we detail here only the in-degree based similarity measurement calculations . The union of ABI and USC databases used here provides information about all 33 areas in terms of the in and out-going connections between 33 areas . We compared the similarity of the input pattern of pairs of areas by evaluating the number of sources areas from which both receive projections or neither do ( i . e . , similarity implies both projections exist or are absent; dissimilarity implies one is absent and the other is present ) . We define a normalized in-link similarity measure , Sxyin as follows: For any pair of areas ( x , y ) , let nxyin denote the number of projecting areas from which either both x and y or neither x nor y receive an incoming link . Because nxyin≤33 , we compute the ratio nxyin/33 for every area pair ( x , y ) . Clearly , this number will depend on the in-degrees of x and y , denoted by kxin and kyin ( 0≤kx ( y ) in≤33 ) . We define the in-link similarity as: Sxyin = nxyin33−pxyin , where pxyin is the expected value of the ratio ( nxyin/33 ) if the incoming connections of x and y were distributed uniformly at random across the 33 source areas . Thus: pxyin = ( kxin33 ) ( kyin33 ) + ( 1−kxin33 ) ( 1−kyin33 ) , where the first term is the probability that both x and y receive a link from a given source , and the second term is the probability that neither of them receive a link from a given source . | It was recently shown that the network of connections between different areas of the macaque cortex has strong structural specificity in terms of the strength of connections as a function of the distance between areas . This has led to a model of cortex connectivity that predicts many observed architectural features , including the existence of a strong core-periphery organization . When viewed across species , increases in brain size are accompanied by a relative decrease in connectivity , and thus an important question is whether there are architectural commonalities in the cortical networks within the mammalian branch . Here , based on tract tracing data from the folded macaque brain and the smooth mouse brain , we introduce a common model framework that allows network comparisons between species . We show that despite important differences in size , the cortices of both species share several network invariants , suggesting that the mammalian cortex exhibits universal architectural principals . This framework also captures differences between the two brains , including the fact that , unlike the macaque , the mouse core includes primary areas and that there is a relative decrease in the frequency of long-distance connections in the large macaque cortex compared to mouse . This approach allows network architectural extrapolations to the human cortex . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infographics",
"medicine",
"and",
"health",
"sciences",
"neural",
"networks",
"nervous",
"system",
"vertebrates",
"mice",
"neuroscience",
"animals",
"mammals",
"primates",
"network",
"analysis",
"old",
"world",
"monkeys",
"computer",
"and",
"information",
"sciences",
"monkeys",
"network",
"motifs",
"animal",
"cells",
"macaque",
"cellular",
"neuroscience",
"rodents",
"cell",
"biology",
"data",
"visualization",
"anatomy",
"central",
"nervous",
"system",
"neurons",
"graphs",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"amniotes",
"organisms"
] | 2016 | Spatial Embedding and Wiring Cost Constrain the Functional Layout of the Cortical Network of Rodents and Primates |
The natural reservoir of Influenza A is waterfowl . Normally , waterfowl viruses are not adapted to infect and spread in the human population . Sometimes , through reassortment or through whole host shift events , genetic material from waterfowl viruses is introduced into the human population causing worldwide pandemics . Identifying which mutations allow viruses from avian origin to spread successfully in the human population is of great importance in predicting and controlling influenza pandemics . Here we describe a novel approach to identify such mutations . We use a sitewise non-homogeneous phylogenetic model that explicitly takes into account differences in the equilibrium frequencies of amino acids in different hosts and locations . We identify 172 amino acid sites with strong support and 518 sites with moderate support of different selection constraints in human and avian viruses . The sites that we identify provide an invaluable resource to experimental virologists studying adaptation of avian flu viruses to the human host . Identification of the sequence changes necessary for host shifts would help us predict the pandemic potential of various strains . The method is of broad applicability to investigating changes in selective constraints when the timing of the changes is known .
Influenza A has the distinction of being an old disease , a recurring disease , and an ‘emerging’ disease . Influenza A viruses are found in humans as well as in other animals including swine , horses , sea mammals , and birds , of which waterfowl are considered the natural reservoir [1] . Subtypes of influenza A are distinguished by two surface glycoproteins; haemagglutinin ( HA ) , the primary target of the immune response , and neuraminidase ( NA ) . There are sixteen known types of haemagglutinin ( H1 to H16 ) and nine of neuraminidase ( N1 to N9 ) , all found in waterfowl . Only H1 , H2 , H3 and N1 , N2 , however , are known to have caused epidemic disease in humans . The predominant forms of influenza A currently circulating in humans are H1N1 and H3N2 . There are two distinct problems represented by influenza . Firstly , the various subtypes currently in circulation in humans cause significant morbidity and loss of life . Secondly , periodically a subtype of influenza can make the shift from aquatic birds to humans , possibly through an intermediate host , resulting in a widespread pandemic in an immunologically-naïve population . These ‘antigenic shifts’ can occur either through the transfer of an entire virus from one host to another , or through a re-assortment process where genomic segments of the avian virus mix with genomic segments currently circulating in humans . In 1957 three virus segments ( HA , NA , and PB1 ) from an avian-like source were combined with the other five segments already circulating in humans to create the H2N2 ‘Asian flu’ pandemic , while in 1968 two segments ( HA and PB1 ) from an avian-like source were combined with the other six from the already-present human H2N2 virus to form the H3N2 ‘Hong Kong flu’ pandemic [2] . It has been suggested that the 1918 H1N1 ‘Spanish flu’ virus was the result of a single host-shift event from birds to humans [3]–[5] but this remains controversial [6]–[9] . In recent years , a number of different avian subtypes have caused sporadic human infections , including H5N1 , H7N3 , H7N7 , and H9N2 [10] . While there is evidence for sporadic transmissions of these avian viruses between humans , the genetic changes necessary for widespread human-human transmission have , so far , seemingly not occurred . A number of different proteins have been implicated in determining host ranges . Influenza haemagglutinin binds to sialic acid linked to galactose on the surface of the targeted cell; the differing nature of the sialic acid-galactose linkages in birds and humans ( α2 , 3 sialic acid linkages in the bird gut , α2 , 6 sialic acid linkages of the upper human respiratory tract [11] ) provides an important barrier to host shift events . A number of amino acid substitutions have occurred in human influenza haemagglutinin ( e . g . Q226L and G228S in H2 and H3 , E190N/D and G225E/D in H1 ) to adjust to the different receptors [12]–[16] . Neuraminidase , the protein responsible for cleaving the haemagglutinin from the receptor surface , also seems adapted to the particular sialic acid linkages , as well as for the pH and temperature of the host tissues [17] . Proteins in the viral replication complex ( PA , PB1 , PB2 , and NP ) have also been implicated in limiting host range by restricting replication and intra-host spread in mammals ( for a review , see [18] . ) Of particular note is the PB2 gene , where one specific substitution , E627K , was identified and characterised experimentally as crucial for replication and intra-host spread in mammals [19]–[21] . As part of the widespread surveillance effort , it is important to understand the process of host shifts , and to identify the important changes that are necessary for the shift to occur , or that make the shift more likely . We currently have many examples of both avian and human viruses , so there have been a number of efforts at identifying ‘genetic signatures’ that characterise the virus as adapted to one or the other host . The most common method is to identify sites where the distribution of amino acids found in the virus in one host are sufficiently different from the distribution of amino acids found in the same site in viruses that affect the other host [22]–[24] . Unfortunately , there are two fundamental problems with this approach . Firstly , the observed changes could represent the result of neutral drift rather than anything specific to the nature of the different hosts . As the human viruses are more closely related to each other than they are to the avian viruses , it would be expected that there would be characteristic amino acids found in the human lineages that are distinct from those found in the avian lineages because of the ‘founder effect’ [25] , that is , the maintenance of the idiosyncratic properties of the particular virus that first infected humans . Comparisons of amino acid frequencies in viruses from the two hosts cannot easily distinguish between those that accidentally accompanied the host shift event and those that were actually associated with different selective constraints acting on the viruses in the two hosts . The second related problem is the use of inappropriate statistical tests to identify when these two distributions are sufficiently different . The statistical tests used generally assume that each of the observed sequences represent a set of independent measurements . But the underlying phylogenetic relationships will generate correlations in the amino acids at a site , confounding the signal due to the host shift event . This can be demonstrated by considering Figure 1 , which shows two possible situations where the avian viruses all have a leucine in a given position where all of the human viruses have a valine in the same position . In example A the results are statistically significant , in that the positions are independent , and it is unlikely that the simultaneous parallel changes in sequence occurred at random in the human viruses but not in the avian viruses . In example B there is much less statistical signal , as only one change of amino acid on the branch connecting the human and avian viruses is needed to explain the multiple observations . By neglecting the underlying phylogenetic structure , a single change of amino acid can be interpreted as a large number of independent events , grossly exaggerating the statistical significance . A number of the published approaches to this problem suffer from the above problems . For example , both Chen et al . [22] and Miotto et al . [24] employed an information-based approach to identify sites where host-specific amino acids can be identified . Their computations of entropy ( a measure of sequence diversity ) and mutual information ( the dependence of the observed residue distribution on host species ) are based on considering every observed sequence as an independent data-point , ignoring correlations between the evolutionarily related sequences . Different distributions in the two hosts can be explained due to the founder effect described above , independent of any role these sites have in host adaptation . That is not to say that their results are incorrect , only that these problems make it impossible to determine their statistical significance . Finkelstein et al [23] looked at sites with a significantly higher degree of conservation in human lineages than avian lineages , and identified 32 markers within the M1 , NP , NS , PA , and PB2 genes , 26 of them on the polymerase proteins NP , PA , and PB2 . This analysis did not consider the phylogenetic relationships explicitly in their calculation of conservation , choosing instead to base their calculation on the frequency of the different amino acids observed in that site in the different hosts . While they employed strict tests for , for instance , multiple hypothesis testing , it is difficult to determine how much their results were affected by considering only frequencies of amino acids to represent the selective constraints , again ignoring the underlying phylogenetic relationships . It is known , for instance , that such counting methods produce very inaccurate amino acid frequencies compared with phylogenetically-based methods [26] , and can not generally identify the rate of substitutions in the tree , but only the range of acceptable amino acids . As described above , the differences in the distribution of amino acids at a given site between avian and human viruses might represent neutral drift or , more interestingly , a change in the underlying selective pressure applied to the virus by the host . Rather than characterising only the difference in observed amino acid distributions , we can instead look directly for evidence of changes in the selective constraints by modelling the phylogenetics explicitly . These selective constraint changes will result in differences in the substitution process , as mutations that arise in one virus or another will have different probabilities of achieving fixation . Thus , changes in selection constraints will manifest themselves as changes in the observed substitution rates . This also allows rigorous statistical methods , such as the likelihood ratio test , to be used to establish statistical significance . The selective pressure acting on a site can be positive , negative , or neutral . Positive selection , also called adaptive , or more misleadingly [27] ‘Darwinian’ , refers to the acceptance of advantageous mutations; negative , or purifying selection involves the rejection of deleterious mutations . Neutral selection pressure involves the chance acceptance of mutations that do not have a significant effect on the fitness . Both positive and negative selection pressure represent strong constraints on the amino acids at a given site; the difference is that during purifying selection the current amino acids generally fulfil these constraints so change is restricted , while during adaptive evolution the current amino acids are not well suited , generally due to changes in the constraints or a selective advantage for diversification , enhancing the rate of evolution until more appropriate residues are found . Changes in the selective constraints can result in changes in the rate of substitutions at that location . If the initial amino acids do not match the current requirements of that site , there may be an adaptive burst of faster substitutions until the constraints are satisfied . Modifications of the stringency of the constraints , causing a given site to be more or less restricted , may cause a longer-term change in the substitution rate without necessarily causing an adaptive burst . Previous phylogenetic methods have generally focused on identifying changes in the absolute substitution rate [28]–[35] or ratio of non-synonymous to synonymous changes [36]–[38] . The latter method was used , for instance , to identify twelve sites on the influenza A nucleoprotein that seem to have undergone a change in selective constraints corresponding to the switch from avian to human host [39] . While these approaches are often useful , transient position-specific adaptive bursts are difficult to identify given the short duration of the effect . Sites can also undergo shifts in selective constraints without adaptive bursts or detectible changes in substitution rates , especially if the constraints in the two hosts overlap . Monitoring changes in the nature of the selective constraints has been much less common [40] and has not been applied to host shift events . In this paper we investigate the use of a phylogenetic method to detect changes in selective constraints that considers not only changes in the magnitude of selection constraints , but also changes in its nature , represented as the relative propensity for the different amino acids . We do this by considering two different models for each site , a homogeneous model where the selective constraints are independent of host , the other a non-homogeneous model where the selective constraints depend upon the host . The likelihood ratio test can then determine the level of statistical support for rejecting the null hypothesis of no such dependence .
We start our analysis with a set of human and avian influenza viral sequences and the associated phylogenetic trees for each influenza gene . We consider the different haemagglutinin and neuraminidase serotypes ( e . g . H1 , H2 , H3 , N1 , N2 ) separately . For each non-conserved site , we apply increasingly complicated substitution models , using the Likelihood Ratio Test ( LRT ) to evaluate the statistical support for each further complication . The substitution models are defined by a symmetric exchangeability matrix S , the equilibrium frequencies of the twenty amino acids π , and a rate scaling parameter ν representing the relative substitution rate at that site compared with other sites . The simplest model , Model 1 , consists of the WAG exchangeability matrix combined with the associated equilibrium frequencies for the different amino acids [41] , with one adjustable parameter per site representing the scaling factor ν . We then consider Model 2 where the equilibrium frequencies of the amino acids are optimised individually for each site [26] . The likelihood ratio test demonstrated that the use of site-specific equilibrium frequencies was justified for all sites ( P values ranging from 0 . 028 to 9 . 4×10−27 ) . We then created a non-homogeneous model , Model 3 where virus substitutions are modelled by one set of substitution rates in the avian host , and by a different set of substitution rates in the human host , as illustrated in Figure 2 . The two different substitution models shared the WAG exchangeability matrix S and a site-specific rate-scaling factor ν , but now the equilibrium amino acid frequencies were both host- and site-specific . We identified sites with statistical support for different substitution rates in the two hosts , using a false discovery rate ( FDR ) method to account for multiple hypothesis testing [42] . We identified 172 sites with an FDR<0 . 05 ( i . e . we would expect 5% of these sites to be false positives ) , and 518 sites with an FDR<0 . 20 . We will refer to the 172 higher-confidence locations as ‘A sites’ and the remaining 346 lower-confidence locations as ‘B sites’ . We then considered if modelling differences in the equilibrium amino acid frequencies was adequate , or whether we should include host-dependent rate scaling factors as well . We implemented a more complicated model ( Model 4 ) where the substitution rates were still defined with the WAG exchangeability matrix , but now both the equilibrium frequencies and the scaling factor ν were host- and site-dependent . Of the 2143 sites considered , few ( 37 ) had P values less than 0 . 05; after correcting for multiple hypothesis testing using the false discovery rate method , no site yielded any statistically significant improvement . The results described below will be based on Model 3 above . The list of 172 ‘A’ sites ( FDR<0 . 05 ) is shown in Table 1 . Sites were found on all of the genes considered . Supporting Table S1 shows the list of the 518 ‘A’ and ‘B’ sites with FDR<0 . 20 . Sites that have been identified experimentally are detected using this method , notably PB2 627 . HA sites H1 190 and 225 and H3 228 are also identified . Sites H2 226 and 228 are significant at the weaker FDR<0 . 20 level , while H3 226 is not statistically significant . To assess the performance of the technique describe here , we simulated each one of the 264 variable sites in the PB2 gene ten times ( 2 , 640 simulations in total ) . All sites were simulated using the same fixed tree topology . The 22 ‘A’ and ‘B’ sites identified as undergoing selective constraint changes ( FDR<0 . 20 ) were simulated under the non-homogeneous model , using the parameters obtained by optimizing model 3 . Similarly , the 242 locations with no evidence for change in selective constraints were simulated under the homogeneous model ( model 2 ) . We then applied the analysis described above to identify locations in the synthetic datasets that had undergone changes in selective pressure . On average , we observed that 1 . 5% of the locations identified with FDR<0 . 05 were false positives ( false positive rate of 0 . 08% ) ; this increased to 3 . 6% ( false positive rate of 0 . 2% ) for FDR<0 . 20 . This indicates that the FDR values are , at least for PB2 , likely conservative . Of the 22 locations modelled with changing selective constraints , 12 . 9 were identified with FDR<0 . 05 ( false negative rate of 41% ) , with 16 . 2 identified with FDR<0 . 20 ( false negative rate of 26% ) . The 13 ‘A’ sites were identified more consistently , with 10 . 1 found with FDR<0 . 05 and 11 . 0 found with FDR<0 . 20 . This suggests that there remain more locations undergoing changes in selective pressure than are being identified with the procedure described here . Our approach relies on the prior construction of an appropriate phylogenetic tree . In order to estimate the effect of phylogenetic uncertainty , we repeated the analysis of the PB2 gene segment with ten different phylogenetic trees obtained through non-parametric bootstrapping . The 13 ‘A’ sites were identified on 79% of the bootstrap trees with FDR<0 . 05 and identified on 90% with FDR<0 . 20 . 85% of the 22 ‘A’ and ‘B’ sites were similarly identified on the bootstrap trees with FDR<0 . 20 . Conversely , the bootstrap trees identified on average 2% ( with FDR<0 . 05 ) and 6% ( with FDR<0 . 20 ) of alternative locations that were not identified on the original tree . These might be false positives for the alternative trees , suggesting a similar amount of false positives on the original tree . Some of these locations , however , may be locations with changes in selective constraints , and thus represent false negatives for the original tree; most of these locations would have been so identified with a higher FDR threshold of 0 . 50 , although these points represent only about 12% of the otherwise unidentified locations . We constructed a simple model to help explain the lack of statistically significant improvement with adding host-specific scaling factors . This was based on considering a protein site where two amino acids ( A and B ) are present , where an organism with residue B has a fitness equal to 1−s relative to an organism with residue A . We used Kimura's fixation rate theory [43] to calculate the resulting substitution rates between A to B , and formulate these expressions in terms of a rate scaling factor ν and equilibrium frequencies πA and πB ( = 1−πA ) . We considered how ν , πA , and πB change as the relative fitness difference between A and B is altered . We also considered the overall rate at which substitutions occur in both directions , both for negative selection where the residues are at equilibrium ( Γ− ) as well as for positive selection ( Γ+ ) where the location contains the unfavourable residue B . Figure 3 shows the dependence of πA , πB , ν , Γ− , and Γ+ ( the latter three normalised by the mutation rate μ ) on the relative fitness difference s ( scaled by the effective population size Neff ) . As shown , under conditions of negative selection , increasing fitness differences result in a decrease in the overall rate of substitutions , but an increase in the rate-scaling factor . There is a relatively weak dependence of ν on s as long as the latter is not large relative to 1/Neff . Under conditions of positive selection , both quantities increase with larger fitness differences . The theoretically predicted weak dependence of ν on selective pressure and the lack of statistical support for host-dependent values of this parameter indicate that ν is not a good measure of the degree of selective constraints . To generate a more appropriate measure , we calculated the relative entropy between the equilibrium frequencies and what would be expected under no selection , π0 , estimating the latter by averaging the amino acid frequencies over our entire database . This measure of selective constraint magnitudes for the various sites in avian and human hosts are presented in Table 1 , Supporting Table S1 , and in Figure 4 .
Most methods that look for changes in the substitution rates model this as changes in ν , the scaling parameter , or in the related ratio of synonymous to non-synonymous substitutions . In our analysis , we find that , when we allow the equilibrium frequencies π to vary , there is no statistically significant variation in ν . This seems initially counter-intuitive , as there are some sites where there seems to be substantial changes in the degree of conservation; in site 274 in N1 , for instance , is almost universally tyrosine in avian viruses , while it varies between tyrosine , serine , and phenylalanine in human viruses . Yet the likelihood ratio test applied to this site rejects the inclusion of host-dependent scaling factors with a P value of 0 . 90 , suggesting that the relationship between rate scaling factors and site variation are not simply related . This observation motivated our simple model to try to gain insight into the relationship between equilibrium frequencies and rate scaling factors , by considering a protein site where two different amino acids , A and B , are found . We imagine that organisms possessing residue A at this location have a fitness advantage . Negative purifying selection would occur when the residues at this location are at their equilibrium value , while positive selection would occur when this location was filled by B , such as might occur when the selective pressure on the protein changes . By using Kimura's theory of fixation probability [43] , we can calculate the values of the rate scaling factor ν , the overall rate of substitutions for purifying ( Γ− ) and positive selection ( Γ+ ) , and the equilibrium frequencies of A ( πA ) and B ( πB ) , as a function of the different finesses provided to an organism with the two different possible amino acids at that location , as described in the Methods section below . Normalised values of ν , Γ− , and Γ+ are plotted as a function of 2Neff s in Figure 3 . As shown , ν varies surprisingly little with s as long as s is not much more than 1/Neff . This explains why including a host-specific ν never yielded statistically significant improvements with our data . When we consider adaptive substitutions , larger values of s correspond to higher selective constraints , larger values of ν , and faster evolution . The situation is quite different with purifying selection . As might be expected , larger values of s ( corresponding to larger degree of purifying selection ) result in a slower substitution rate , but this actually corresponds to larger values of ν . The reason why most phylogenetic programs use an inverted relationship , where larger values of ν correspond to faster substitution rates , is that they do not consider the value of π appropriate for each site . By assuming that the same values of π apply to all sites , a more extreme distribution of equilibrium frequencies , resulting in a decrease in the number of substitutions , is interpreted as a reduction in ν although this parameter is , in fact , increasing The magnitude of the selective constraints for the various sites in avian and human hosts are presented in Table 1 , Supporting Table S1 , and in Figure 4 . It is interesting to note the number of positions under changing selection constraints where the magnitudes of the selection constraints are relatively constant . Such sites would be difficult to detect by looking for changes in the substitution rate , especially in cases where the distributions of amino acids found in the two hosts have significant overlap . The methods described here are applicable for a wide range of problems involving changes in selective constraints . There are two particular factors , however , that make the technique especially well suited for influenza . Firstly , the branch along which the selective pressure changes can be identified a priori . Secondly , it is important to generate appropriate phylogenetic trees for the position under consideration . Generation of such trees can be complicated when there is incongruence between different locations . For influenza , incongruence between the various genomic segments results from the process of reassortment , where chimeric viruses containing genomic segments of different origin result from multiple infections . We are able to address this issue by considering each different genomic segment independently , constructing gene-specific phylogenetic trees . A more difficult problem is intra-gene homologous recombination , where different regions of a single genomic segment have different phylogenies . Such recombination is either extremely rare or non-existent in influenza ( as well as other negative RNA viruses ) , and has never been observed experimentally [47]–[49] . We have assumed that the transitions from avian to human hosts did not go through an intermediate species , such as swine . There is no evidence of involvement of swine in the 1957 Asian flu and 1968 Hong Kong flu host shift events . Based on his analysis of the 1918 Spanish flu sequences and the relative timing of the 1918 influenza outbreaks in swine and humans , Taubenberger concluded that the Spanish flu transferred in toto from birds to humans and from humans to swine [3]–[5] , although this conclusion has been challenged [6]–[9] . If an intermediate host species were involved , it would not be expected to affect the results if the selective constraints at any location in this intermediate host were to resemble either that of avian or human viruses , as this would only change the timing of the shift from one selective constraint to another . If there were an intermediate host and the selective constraints at some locations in this intermediate host were strong and substantially different from either avian or human viruses , the amount of evolutionary time in this intermediate host were sufficiently long , and the evolutionary time in humans sufficiently short so that the new equilibrium is not attained , the results of these calculations could be affected . There are two other important assumptions made in this work . Firstly , we assume that the selective constraints in human and avian viruses are constant , and that each location can be considered independently . We do not consider , for instance , that there may be different selective constraints in low-pathogenic and high-pathogenic avian viruses , or that compensatory changes can occur elsewhere in the protein or even in other proteins . The observation ( both here and experimentally [12]–[16] ) that different hemagglutinin subtypes undergo different patterns of change of selective constraints indicates that this assumption is not strictly valid .
For the following discussion we assume the evolution of a viral protein along a phylogenetic tree with two different host lineages , avian and human , where we consider the root of the tree to exist somewhere in the avian lineage . The evolution of amino acids in a site along a phylogenetic tree can be modelled as a continuous Markov process , described by a 20×20 substitution matrix Q . ( Standard phylogenetic modelling techniques are described in [50] . ) In order to provide for time reversibility ( that is , the expected number i to j transitions equalling the expected number of transitions from j to i ) , this is commonly represented as where S is a symmetric matrix representing the exchangeability of amino acids i and j , πj is the equilibrium frequency of amino acid j ( ) and ν is a scaling parameter that accounts for the overall rate of substitution at the site . S encodes the underlying codon structure as well as the relative similarities of the physicochemical properties of the amino acids , while the equilibrium frequencies represent the relative propensities for each of the amino acids at that site . We can calculate the likelihood of the data at this site given the model using Felsenstein's pruning algorithm [51] , [52] . We first consider a standard substitution model where S and π are given by the WAG substitution matrix [41] , where each site in the set of proteins is characterised by a distinct substitution rate scaling factor ν whose value is determined by maximising the log likelihood given the sequence data at that site and the input phylogenetic tree . This we refer to as Model 1 . We then considered the appropriateness of modelling each site in the set of proteins with a distinctive set of equilibrium amino acid frequencies [26] , what we refer to as single-site homogeneous Model 2 . We adjust the values of π simultaneously with ν to maximise the likelihood . To avoid over parameterisation , we still use WAG S values for all sites . The tree topology is assumed fixed , and branch lengths are the same for all sites . In order to reduce the number of adjustable parameters , πi = 0 for any amino acids not found at that site . As the equilibrium frequencies of the amino acids not observed are set to zero , this results in an increase in the number of parameters equal to the number of amino acids present at that site minus one ( due to the constraint that the equilibrium frequencies must sum to one ) . We then use the likelihood ratio test to see if site-dependent equilibrium frequencies can be justified with the data . As described in the Results section , the site-dependence of the equilibrium frequencies could be justified for all sites . Now let us imagine that upon inspection of the phylogenetic tree , we notice that amino acid preferences at a particular site seem different in the two host clades . We can incorporate this observation into our model by using two distinct Q matrices to describe the evolution of this site in the different hosts , as illustrated in Figure 2 . For the reservoir avian host we write and for the new human host where π and π′ represent the equilibrium amino acid frequencies at that site in avian and human viruses , respectively . ( In principle we could also have S depend upon the host , but this would result in a large increase in the number of adjustable parameters . We will consider host-dependence of ν below . ) The host shift event is defined as the midpoint of the branch connecting the common ancestor of the human viruses with its parent node . We can now calculate a new likelihood for this site using the same fixed topology , again adjusting π , π′ , and ν to maximise the likelihood . We call this the single site non-homogeneous model , Model 3 . Again , the increase in the number of adjustable parameters for Model 3 relative to Model 2 equals the number of amino acid types observed at that site minus one . Because the Model 2 is nested inside Model 3 , we can again use the likelihood ratio test to test the hypothesis of different selective constraints in different hosts at that site . In general , for a protein with N variable sites , we could repeat the procedure above for each site in the alignment , and perform N likelihood ratio tests . This would generate a list of those sites that show statistically different amino acid compositions , and hence distinctive selective constraints , in the different hosts . Following the calculation of the statistical significance for each site we can then use standard false discovery rate ( FDR ) methods to account for multiple hypothesis testing [42] . Finally , we consider if , in addition to host-dependent equilibrium frequencies , we also have statistical evidence for host-dependent rate scaling factors . We again use for the reservoir avian host but now use for the new human host where ν and ν′ represent the rate scaling factors at that site in avian and human viruses , respectfully . Again , Model 3 is nested inside Model 4 with an increase of one adjustable parameter , meaning that the statistical support for this extra factor can be evaluated with the likelihood ratio test . We do not observe support for this extra parameter in any of the sites after adjusting for multiple hypothesis testing . Human and avian viral sequences were collected from the NCBI Influenza Virus Resource [53] . Due to the frequency of reassortment , we cannot assume that the phylogenetic relationships for the various genomic segments are similar; they must be treated independently , including creating genetic-segment specific phylogenetic trees . The sequences for the various segments were treated as independent data sets , with separate datasets for the H1 , H2 , H3 , N1 , and N2 genes . Clusters of highly similar sequences ( approximately >99 . 5% ) were culled as to reduce the overall number of sequences to around 400 per dataset . It is common to find sporadic transmissions between avian , human , and other ( e . g . swine ) hosts; we eliminated all sequences resulting from such transmissions ( e . g . human H5N1 sequences ) , leaving us with a single connected set of avian sequences and separate monophyletic human clades corresponding to the host shift events of 1918 ( H1 , N1 , internal genes ) , 1957 ( H2 , N2 , PB1 ) , and 1968 ( H3 , PB1 ) . In order to generate more accurate phylogenetic trees , the culled sequences were aligned at the amino acid level ( MUSCLE , [54] ) , with these alignments then used to create nucleotide codon alignments ( PAL2NAL , [55] ) . The phylogenetic tree topologies were then created for the nucleotide data using PhyML ( [56]; HKY85 model [57] , Gamma-distributed rates ) . The resulting trees are included as Supporting Figure S1–S11 . Because amino acid distances are needed for the models developed here , branch lengths were then re-optimised for this fixed tree topology using the corresponding amino acid data ( PAML [58] , [59] , WAG substitution matrix [41] , Gamma-distributed rates ) . The analysis was then performed with each gene set , based on the phylogenetic tree for the genomic segment in which the gene is located . A computer program written in Java that implements and optimises the various models described above is available from the authors . The determinations of changes in selective constraints at each site is a separate hypothesis to be evaluated , so we must address the multiple-hypothesis question , that is , if we ask a suitably large number of statistical questions we are likely , at random , to obtain some statistically-significant results . We use the false discovery rate method , that is , specifying for each site the false positive rate that would have to be tolerated in order for that result to be statistically significant , following the Benjamini and Hochberg estimator [42] . We first choose an acceptable false discovery rate δ . If P ( k ) is the k-th smallest P value for a set of n sites , we choose the largest value of k so that . As different genes are evolving in different circumstances , we would not expect the fraction of sites in each gene undergoing changes in selective constraints to be the same . Combining all of the genes together in one dataset would result in an increase in false positives for the genes with fewer changes in selective constraints , and an increase in false negatives for the genes with more changes in selective constraints . For this reason we analyse the false discovery rate for each gene individually . Table 1 and Supporting Table S1 list , for each site , the smallest possible acceptable false discovery rate that would result in that site being labelled as statistically significant . These should not be interpreted as the probability that that given site is a false positive . Each site was simulated under the homogeneous ( Model 2 ) and non-homogeneous ( Model 3 ) models 10 times using the program Evolver [59] using the estimated tree topology and the WAG+F substitution matrix [50] . For each site , the tree was scaled according to the site-specific estimated rate-scaling parameter ν . Simulation under the non-homogeneous model was performed in two steps: the avian part of the tree was simulated using a randomly generated root sequence following the avian equilibrium frequencies for that location . The avian subtree contained a host shift tip that served as the root of the human clade . The human subtree was then simulated according the human equilibrium frequencies using the simulated avian sequence at the host shift . The PB2 sequence was bootstrapped 10 times and tree topology re-estimated for each boot sample . The homogeneous and non-homogeneous models were optimised for the observed data at each location , and the LRT was performed again for each one of the 10 new tree topologies so as to assess the effect of tree topology uncertainty on the identification of adaptive sites . Consider a protein site where two amino acids , A and B , are found . Let us imagine that that A is the more advantageous amino acid , that is , organisms with A at this site have a higher fitness , while organisms with B at this site has relative fitness . Let us also imagine that the mutation rate from A to B μAB is equal to the reverse mutation rate μBA = μ . We imagine a number of different lineages that have diverged , each with effective population size Neff . Assuming that the mutation rate relative to the population is reasonably small , A or B will become fixed in each lineage . For haploid organisms , the probability that A would become fixed in a given lineage is given by [43] ( 1 ) where we have recognised that this probability is simply the equilibrium frequency of A in the ensemble of diverged organisms , with . The substitution rate of A by B is just the mutation rate μ times the fixation probability , given by Kimura's formula for small s [43] . ( 2 ) We can compare these expressions with as used in phylogenetic analyses . As we are only dealing with two different residues , is a simple multiplicative constant and can be set equal to one , resulting in . Equating these two expressions for QBA and solving for ν yields ( 3 ) ( 4 ) Similar results are obtained , as would be expected , when we express . We can now consider the cases of neutral , adaptive ( positive ) , and purifying ( negative ) selection . Neutral selection is simply the case when Neffs is small and . For both neutral and negative selection , we can consider the overall rate at which substitutions occur , given by , which is equal to Neff μ in the case of neutral selection . Positive selection involves the situation where we are not at equilibrium , but rather , at least in this case , we have the less-fit residue occupying the given position . In this case , assuming again that A is the favoured residue , . We characterise the selection constraints by how far the equilibrium amino acid frequencies π differ from what would be expected under no selection π0 through the relative entropy , defined as ( 5 ) which is , as is desired , zero when π equals π0 . Unfortunately , it is difficult to estimate π0 , as there is little of the virus genome that is not under some degree of selective constraints . We estimate π0 by averaging the amino acid frequencies over our entire database , with the expectation that specific selection constraints will , at least approximately , average out . | Influenza A's natural reservoir is waterfowl . Sometimes avian virus genomic segments are able to shift to a human host , either in toto or by combining with those that underwent a previous host shift event . Such host shift events can cause worldwide pandemics in their immunologically naive hosts . In order for these host shifts to establish a stable lineage , the virus has to adapt to the new host . Identifying the changes that have occurred in the past can provide important clues about how this process happens , and how surveillance for new influenza threats should be targeted . Unfortunately , it is difficult to determine whether an amino acid has changed due to adaptation to the new host or whether the change occurred through random drift . Here we describe a novel phylogenetic approach to identifying locations where the nature of the selective pressure exerted on the location has changed corresponding to the host shift event . We identify a set of locations on a number of the genomic segments . The approach we describe is of wide applicability when the timing of the change of selective constraints is known in advance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"virology/virus",
"evolution",
"and",
"symbiosis",
"computational",
"biology/comparative",
"sequence",
"analysis",
"molecular",
"biology/molecular",
"evolution",
"computational",
"biology/evolutionary",
"modeling",
"infectious",
"diseases/viral",
"infections"
] | 2009 | Identifying Changes in Selective Constraints: Host Shifts in Influenza |
Many viral infections , including HIV , exhibit sex-based pathogenic differences . However , few studies have examined vaccine-related sex differences . We compared immunogenicity and protective efficacy of monomeric SIV gp120 with oligomeric SIV gp140 in a pre-clinical rhesus macaque study and explored a subsequent sex bias in vaccine outcome . Each immunization group ( 16 females , 8 males ) was primed twice mucosally with replication-competent Ad-recombinants encoding SIVsmH4env/rev , SIV239gag and SIV239nefΔ1–13 and boosted twice intramuscularly with SIVmac239 monomeric gp120 or oligomeric gp140 in MF59 adjuvant . Controls ( 7 females , 5 males ) received empty Ad and MF59 . Up to 9 weekly intrarectal challenges with low-dose SIVmac251 were administered until macaques became infected . We assessed vaccine-induced binding , neutralizing , and non-neutralizing antibodies , Env-specific memory B cells and plasmablasts/plasma cells ( PB/PC ) in bone marrow and rectal tissue , mucosal Env-specific antibodies , and Env-specific T-cells . Post-challenge , only one macaque ( gp140-immunized ) remained uninfected . However , SIV acquisition was significantly delayed in vaccinated females but not males , correlated with Env-specific IgA in rectal secretions , rectal Env-specific memory B cells , and PC in rectal tissue . These results extend previous correlations of mucosal antibodies and memory B cells with protective efficacy . The gp140 regimen was more immunogenic , stimulating elevated gp140 and cyclic V2 binding antibodies , ADCC and ADCP activities , bone marrow Env-specific PB/PC , and rectal gp140-specific IgG . However , immunization with gp120 , the form of envelope immunogen used in RV144 , the only vaccine trial to show some efficacy , provided more significant acquisition delay . Further over 40 weeks of follow-up , no gp120 immunized macaques met euthanasia criteria in contrast to 7 gp140-immunized and 2 control animals . Although males had higher binding antibodies than females , ADCC and ADCP activities were similar . The complex challenge outcomes may reflect differences in IgG subtypes , Fc glycosylation , Fc-R polymorphisms , and/or the microbiome , key areas for future studies . This first demonstration of a sex-difference in SIV vaccine-induced protection emphasizes the need for sex-balancing in vaccine trials . Our results highlight the importance of mucosal immunity and memory B cells at the SIV exposure site for protection .
Sex differences in the pathogenesis of numerous viral diseases , including HIV , are well-known [1] . HIV-infected women exhibit lower viral loads and higher CD4 counts than men , but progress faster to AIDS [2] . Women with similar viral loads as men exhibit a 1 . 6-fold higher risk of AIDS [3] . This sex bias is associated with differences in immune responses . Following viral infections , antigen recognition by pattern recognition receptors , induction of innate and adaptive immune responses , and production of inflammatory cytokines are higher in females than in males [1] . After viral clearance , immune responses in females can remain elevated , contributing to pathogenesis [1] . Less is known regarding sex differences following vaccination . Females have exhibited better immune responses to HSV-2 gD , HBV , and inactivated influenza vaccines [1] , but sex-based effects following HIV/SIV vaccinations have not been reported . Using a large number of female rhesus macaques in a pre-clinical SIV vaccine study we uncovered a sex bias in vaccine-elicited immunity and protective efficacy . Our vaccine strategy is based on mucosally-delivered replicating Ad-recombinants which target myeloid dendritic cells and persist in rectal macrophages , eliciting systemic and mucosal immunity [4] . Following Ad-priming we compared the immunogenicity and protective efficacy of regimens boosted with monomeric SIV gp120 or oligomeric SIV gp140 . gp120 immunogens are of interest as they were the form of antigen used as subunit boost for the RV144 clinical trial , the first to show modest protection [5] . Although that vaccine regimen failed to elicit neutralizing antibodies ( nAbs ) against primary HIV circulating isolates [6] , non-neutralizing antibodies exhibiting binding to V1/V2 and high ADCC activity in the presence of low serum IgA levels correlated with reduced infection risk [7–8] . Nevertheless , broadly neutralizing antibodies ( bnAbs ) are believed important for a highly efficacious vaccine . They develop in a small proportion of HIV-1 patients over prolonged infection , and contribute to maintenance of low viremia [9] . Passive transfer of bnAbs in non-human primates has protected against SHIV infection [10] . Thus , rational design of HIV Env antigens for elicitation of bnAbs is at the forefront of HIV research [11–12] . Native gp140 trimers are thought to be more promising for this purpose compared to monomeric gp120 [13–16] due to the presence of conserved conformational and quaternary epitopes . For example , the potent bnAb 35O22 targets an epitope shared across gp120 and gp41 [17] . The SIV rhesus macaque model is extensively used in pre-clinical vaccine research as SIV transmission and disease progression in macaques resemble human HIV infection [18] . However , SIV monomeric and oligomeric Env immunogens have not been directly compared in this model . We assessed both proteins as booster immunogens , focusing on systemic and mucosal humoral immunity , and evaluating protective efficacy following repeated low-dose SIV rectal challenges . Viremia reductions were modest post-challenge , but we discovered for the first time a sex bias in SIV vaccine outcome . Female but not male macaques exhibited significantly delayed SIV acquisition . These findings are timely in view of recent NIH policy requiring balancing of males and females in animal studies [19] . The mechanisms of acquisition delay point to local mucosal B cell responses .
Macaques mucosally primed with Ad5hr-SIV recombinants and boosted with monomeric gp120 or oligomeric gp140 as described in Materials and Methods and outlined in Fig 1A were challenged intrarectally with repeated low doses of SIVmac251 eight weeks following the last immunization . All macaques became infected by the 9th exposure except one gp140-immunized female . No difference in rate of infection was observed between all immunized macaques combined versus the controls ( Fig 1B ) or between either immunization group and the controls ( Fig 1C ) . As the study included only 12 contemporaneous controls , to achieve greater statistical power we combined these with an additional 53 historical controls ( 17 females , 36 males ) which had been challenged intrarectally at weekly intervals with the same low dose of the same SIV challenge stock . There was no difference in rate of SIV acquisition between the contemporaneous and historical control groups ( Fig 1D ) . A comparison of all the immunized macaques with these combined controls ( n = 65 ) showed a marginally significant difference in rate of SIV acquisition ( Fig 1E ) , suggesting a vaccine effect . To explore this effect further , we evaluated the rate of SIV acquisition among the immunized male and female macaques and the combined control males and females . We observed significantly delayed acquisition in all the immunized females when compared to the combined control females ( Fig 1F ) , whereas no acquisition delay was seen when all the immunized males were compared to the combined control males ( Fig 1G ) . A direct comparison of all the immunized females versus all the immunized males confirmed a significant acquisition delay in the females ( S1A Fig ) . We next explored the influence of the booster immunogen on acquisition delay . A significant difference was observed when gp120-immunized females were compared to the combined control females ( Fig 1H ) , whereas a marginally non-significant difference was seen when comparing gp140-immunized females versus the combined control females ( Fig 1I ) . In contrast , no delay in acquisition was seen when either gp120- or gp140-immunized males were compared with the combined control males ( Fig 1J and 1K ) . A direct comparison of the gp120- and gp140-immunized females versus the similarly immunized males again confirmed the delayed acquisition in gp120- but not gp140-immunized females ( S1B and S1C Fig ) . Overall the delay in SIV acquisition of the gp120 immunized females was clearly a vaccine effect and provides the first demonstration of a sex bias in SIV vaccination outcome . To understand the basis for the significantly delayed acquisition observed in gp120 immunized but not gp140 immunized females , we conducted a thorough analysis of systemic and mucosal humoral immune responses throughout the course of immunization and post-challenge . We first compared the two immunization groups . Oligomeric gp140 proved to be more immunogenic than gp120 as summarized in Tables 1 and 2 and detailed in the accompanying supplemental figures . Systemic Env-specific binding antibodies following Ad5hr-recombinant immunizations ( wk 14 ) were boosted to titers over 106 ( wk 53 ) in both immunization groups ( S2A–S2C Fig ) . The gp120 group exhibited similar antibody titers against gp140 and gp120 but gp140-immunized animals developed higher titers to gp140 with an overall higher titer to gp140 compared to gp120-immunized macaques ( Table 1; S2A–S2C Fig ) . Antibody levels were maintained between wk 53 post-vaccination and 2 weeks post-infection ( 2wkpi ) in both groups . The gp140 immunized macaques also developed higher cyclic V2-specific binding antibody titers than the gp120 group ( Table 1; S2D Fig ) . Serum nAb titers against tier 1 SIVmac251 . 6 were comparable in both immunization groups ( Table 1; S3A Fig ) . No neutralization of challenge-related tier 3 SIVmac251 . 30 developed . Higher ADCC activity ( S3B and S3C Fig ) was elicited by gp140 compared to gp120 immunization , regardless of whether gp140- or gp120-coated targets were tested ( Table 1 ) . Similarly , antibody-mediated phagocytosis of gp140-coated beads pre-and post-challenge was elevated in both immunization groups compared to controls ( p<0 . 0001; S3D Fig ) . gp140-immunized macaques phagocytosed gp120-coated beads significantly above control and gp120-immunized macaque levels ( wk 53 , Table 1; S3E Fig ) , whereas phagocytosis by gp120-immunized macaques was higher than that of gp140-immunized macaques 2wkpi ( p = 0 . 0034; S3D Fig ) . Mucosal binding antibodies were also assessed during the immunization regimen . Rectal gp120- and gp140-specific IgA and IgG were elicited following mucosal Ad-recombinant priming ( wk 14 ) in both immunization groups . After systemic Env immunization , mucosal IgG was significantly boosted ( wk 53 ) while Env-specific IgA was maintained at post-Ad levels ( S4A–S4D Fig ) . In most cases , IgA and IgG mucosal antibodies in both groups showed elevated reactivity to gp120 at wk 53 . gp140-immunized animals developed higher levels of Env-specific rectal IgG against gp140 ( wk 53 ) compared to gp120-immunized macaques ( Table 2 ) . Bone marrow ( BM ) antibody secreting cells ( ASC ) were next assessed by ELISpot . SIV Env-specific IgG and IgA memory B cells significantly declined after peak elicitation ( wk 53 ) , but rebounded 2wkpi ( S5A and S5B Fig ) . IgA memory B cells developed at higher levels than IgG memory B cells at wk 53 in both groups ( p = 0 . 0049 and 0 . 036 ) , and also 2wkpi ( p = 0 . 0001 and 0 . 023 ) ( S5C and S5D Fig ) . Env-specific plasmablasts ( PB ) and plasma cells ( PC ) exhibited a similar response pattern as memory B cells , but displayed smaller IgG and IgA ASC declines between wks 53 and 57 as expected for long-term memory cells ( S6A and S6B Fig ) . The gp140 group maintained higher levels of both IgG and IgA PB/PC prior to challenge ( wk 53 , Table 2; wk 57 , S6A and S6B Fig ) . In both groups IgG PB/PC were elevated compared to IgA PB/PC at wk 53 ( p = 0 . 0072 and 0 . 014 , respectively ) , but IgA was higher than IgG in gp120-immunized macaques 2wkpi ( p = 0 . 023; S6C and S6D Fig ) . With regard to cellular immune responses , we investigated SIV specific CD4+TM and CD8+TM T-cell responses in PBMC 2wkpi . SIVsmH4 Env-specific CD4+ and CD8+ T-cell responses , representative of env encoded in the Ad-recombinant , were comparable between immunization groups , and appeared post-infection in controls ( S7A and S7C Fig ) . In contrast , gp140- compared to gp120-immunized macaques exhibited a trend of elevated CD4+ and CD8+ T-cell responses following SIVmac239 Env stimulation , suggesting a more effective booster immunization ( S7B and S7D Fig ) . Similar results were seen after summing responses to Env , Gag , and Nef ( S7E–S7H Fig ) . Having shown that immune responses in general were elevated in the gp140 immunized macaques , but that SIV acquisition delay was observed in gp120 immunized female macaques , we next analyzed these data by sex . Systemic binding antibodies to the SIVmac239 Env boosting immunogens were higher in gp120-immunized males compared to females against both gp120 and gp140 targets prior to challenge ( wks 53 and 57 ) , and were maintained at higher levels against gp120 2wkpi ( Fig 2A ) . A similar result was not seen in the gp140-immunized animals ( Fig 2B ) . Males of both groups combined exhibited higher titers to gp120 than females at all time points ( Fig 2C ) . Antibody responses to SIV EnvE660 , representative of SIV EnvsmH4 in the Ad-recombinant , were higher in immunized males compared to females following priming ( wk 14; Fig 2D ) . However , no significant sex differences were seen in neutralizing antibody titers or binding titers to cyclic V2 ( S8A–S8C Fig ) ; BM ASC ( S9A–S9D Fig ) , or rectal Env-specific IgA and IgG ( S10A–S10D Fig ) . Although no significant sex difference by group was observed in phagocytic activity , gp120-immunized females maintained higher activity against gp140-coated beads compared to the gp140 group ( Fig 2E ) . Additionally , no sex differences were seen in ADCC activity by group; however , consistent with results of the group analysis ( S3B and S3C Fig ) gp140-immunized females and males maintained higher activity against both gp120 and gp140 targets ( Fig 2F ) . Female macaques displayed significantly higher rectal Env-specific memory B cell levels than males 2wkpi ( Fig 2G ) , regardless of immunization group . A similar trend was seen both prior to challenge ( wk 53 ) and 8wkpi . Env-specific CD4+TM and CD8+TM T-cell responses showed no sex-based differences ( S11A–S11D Fig ) , although females tended to exhibit higher responses following Env239 stimulation , indicative of the protein boosts derived from that strain ( S11B and S11D Fig ) . When CD4+TM and CD8+TM responses against Env , Gag , and Nef were summed , results were similar in animals stimulated with EnvsmH4 peptides , matched to the env gene in the Ad-recombinant , ( S11E and S11G Fig ) whereas females showed higher CD4+TM and CD8+TM T-cell responses than males in the animals stimulated with Env239 peptides , matched to the Env booster immunogens , significantly so for the CD4 responses ( p = 0 . 019; S11F and S11H Fig ) . Analysis of all the immunogenicity data showed that neither humoral nor cellular systemic immune responses , including serum binding antibodies , serum neutralizing or non-neutralizing activities , bone marrow memory B cells and PB/PC , and CD4+ and CD8+ T cell responses , correlated with SIV acquisition delay . With regard to mucosal immune responses , Env specific IgG in rectal secretions was not associated with acquisition delay in either gp120- or gp140-immunized male or female macaques ( S12 Fig ) . However , although present at lower levels , Env-specific IgA in rectal secretions significantly correlated with delayed acquisition ( Fig 3A ) . All immunized animals with rectal Env-specific IgA levels above the median ( 0 . 04ng/μg total IgA ) required more SIV exposures for infection . The difference remained significant in the gp140 group alone ( tested against gp140 , Fig 3C ) but not in the gp120 group ( tested against gp120 , Fig 3B ) . This same pattern was exhibited by immunized females . Higher Env-specific rectal IgA levels in all immunized females and in gp140-immunized females but not in gp120-immunized females were associated with an increased number of challenges ( Fig 3D–3F ) . Env-specific rectal IgA in vaccinated males did not correlate with delayed acquisition ( S13 Fig ) . As delayed acquisition of immunized females was most evident in gp120-immunized macaques ( Fig 1H and 1I ) , additional factors must have been involved . To pursue the role of mucosal immunity in delayed acquisition , we next examined Env-specific memory B cells and total PB and PC in rectal tissue by flow cytometry [20–21] ( S14 Fig ) . Consistent with the higher rectal Env-specific memory B cell levels 2wkpi in immunized females compared to males ( Fig 2G ) , the rectal Env-specific memory B cell levels 2wkpi were significantly correlated with challenge exposures in all immunized females , but not males ( Fig 3G and 3H ) . This correlation remained significant in gp120-immunized females and approached significance in gp140-immunized females ( Fig 3I and 3J ) . Total rectal PC levels were significantly correlated with acquisition delay in all immunized females but not males ( Fig 3K and 3L ) and in gp120- and gp140-immunized females analyzed separately ( Fig 3M and 3N ) . Overall , our data strongly implicate a local mucosal B cell contribution in delayed acquisition of vaccinated female macaques . A secondary outcome of this study was modestly reduced acute phase viremia in the immunized macaques compared to the controls . Median peak viremia for the gp120 and gp140 groups ( 1 . 79x107 and 2 . 16x107 SIV RNA copies/ml , respectively ) were reduced nearly one log compared to controls ( 1 . 71x108 SIV RNA copies/ml; p<0 . 05 ) . Viremia differences between gp120- and gp140-immunized macaques and controls were significant at 2 , 3 and 4wkpi , while the gp140 group also exhibited lower viremia at 6 and 8wkpi ( Fig 4A ) . Viral loads of the individual macaques are shown in S15 Fig . In contrast to SIV acquisition , no sex bias was observed in viremia reduction . Both females and males of both immunization groups as well as the controls exhibited similar viral loads during the acute phase of infection ( Fig 4B–4D ) . Similarly , CD4 counts over the period of follow-up were similar between the sexes ( Fig 4E–4G ) . We did observe a decrease in viral loads of males compared to females in the gp120 group over weeks 24 to 40 post infection ( Fig 4B ) . A similar difference was not seen in the gp140 immunized macaques , however , we cannot reach a firm conclusion regarding an immunization group difference as a number of macaques in the gp140 group had been euthanized prior to 40 weeks of follow up ( see below ) . Viral loads during the acute phase of infection for the historical controls were available for 41 of the additional 53 macaques , however , no acute viral load difference was observed between the current and historical controls or between males and females of the combined current and historical control groups ( S16 Fig ) . We next examined vaccine-induced immune responses associated with the modestly reduced acute phase viremia in the immunized macaques . We found that phagocytic activity prior to challenge ( wk 53 ) against gp140 targets by the gp140-immunization group , which displayed more prolonged viremia control than the gp120-immunization group ( Fig 4A ) , was significantly correlated with reduced viremia ( Fig 5A ) . Phagocytosis by all macaques was inversely correlated with peak viremia 2wkpi ( Fig 5B ) . No correlation with neutralizing antibody or ADCC activity was observed ( S17 Fig ) . CD8+ T-cell responses contribute to viremia control in natural infection [22–23] , confirmed in numerous pre-clinical vaccine studies [24–32] . SIVsmH4 Env-specific CD8+TM T-cells in all macaques significantly correlated with reduced peak and chronic viremia ( Fig 5C and 5D ) . By immunization group , a significant inverse correlation was only observed between SIVsmH4 Env-specific cytokine-producing CD8+TM T-cells of gp140-immunized macaques and viremia levels at peak and acute-phase time points and during the chronic phase of infection ( Fig 5E–5G ) . No correlations with SIVmac239 Env-specific CD8+TM T cells were observed . Overall , the contribution of cellular immunity to viremia reduction was most evident in gp140-immunized macaques . Although cellular responses in macaques overall and in gp140-immunized macaques were associated with better viremia control ( Fig 5C–5G ) , this outcome was not reproduced in females . SIVsmH4 Env-specific CD8+TM responses in all males but not all females correlated significantly with reduced peak , acute-phase , and chronic viremia ( p = 0 . 0055 , 0 . 0004 , and 0 . 0086 , respectively; S18A and S18B Fig ) . We observed that the number of challenges necessary to infect immunized females but not males correlated inversely with peak viremia ( Fig 6A and 6B ) . Thus we speculate that repeated exposures boosted immunity , leading to better acute viremia control . Over 40 weeks of follow-up , no group differences were seen in males or females with regard to viral loads ( S19A and S19B Fig ) or CD4 counts ( S19C and S19D Fig ) , with the exception as mentioned above , that gp120 immunized females exhibited higher viral loads than similarly immunized males over weeks 24–40 of the chronic phase ( Fig 4B ) . However , in spite of enhanced immunogenicity , significantly more gp140-immunized macaques ( n = 7 ) met established criteria and had to be euthanized before 40wkpi compared to gp120-immunized macaques ( n = 0 ) and the controls ( n = 2 ) ( Fig 6C ) . While 5 females and 2 males in the gp140 group were euthanized before 40wkpi ( Fig 6D ) this difference was not statistically significant .
Here we report for the first time a sex bias in SIV vaccine-induced protective efficacy . Delayed SIV acquisition in females was associated with local B cell immunity , including Env-specific mucosal IgA , Env-specific rectal memory B cells , and rectal PC . Our results highlight the importance of mucosal immunity and development of memory B cells at the site of viral exposure for an effective vaccine . The correlations of anamnestic Env-specific rectal memory B cell and total rectal PC responses with acquisition delay were obtained with samples obtained 2wkpi . It is possible that these B cell responses could have been boosted by the series of repeated low dose viral exposures necessary to infect the female macaques . However , in the absence of any detectable infection over the course of these weekly challenges , these responses , initially elicited by vaccination , even if boosted , were contributing to protective efficacy . Similar responses were not observed in control macaques . Future studies should investigate more fully the possibility of antigenic boosting by repeated low-dose challenge exposures . Our previous report of vaccine-induced rectal IgA correlating with delayed SIVmac251 acquisition [33] is confirmed here and extended by demonstrating the sex bias . Other reports have also associated mucosal antibody with protection . Vaccine-induced rectal antibodies mediating transcytosis correlated with decreased chronic viremia [34] . Macaques protected against repeated vaginal SHIV challenges exhibited vaginal IgAs that blocked transcytosis and vaginal IgGs with neutralizing and/or ADCC activity [35] . Following intravenous SIVmac251 challenge , aerosol-vaccinated macaques exhibited reduced CD4+ T-cell depletion in the lung correlated with viral-specific IgA in bronchoalveolar lavage and nasal fluid [36] . Thus the rationale for continued study of mucosal antibodies in vaccine efficacy is well-substantiated . We previously reported a correlation of vaccine-elicited HIV and SIV Env-specific IgG and IgA peripheral blood memory B cells with reduced viremia [37] . Here we extend this finding , demonstrating the importance of Env-specific memory B cells and PC at the mucosal exposure site for delayed SIVmac251 acquisition . It will be important to further explore how vaccine designs can foster homing of memory B cells to the mucosa and enhance their retention . Here we believe the replicating Ad-recombinants played a major role . We have previously shown that the biodistribution of this vector following administration to the upper respiratory tract is broad , and that it exhibits persistent expression in rectal macrophages [4] . Certainly , continued exploration of vaccine-elicited mucosal immune responses in males and females is warranted , along with pursuit of vaccine regimens that target the intestinal mucosa . Females exhibited a higher percentage of SIV Env-specific memory B cells in rectal tissue , consistent with higher basal immunoglobulin levels and greater humoral responses to antigens in women compared to men [1] . While mucosal antibodies correlated with significant acquisition delay in females , male macaques exhibited higher serum antibody binding titers than females at the time of peak response , 2 weeks after the second envelope boost . Nevertheless , no sex bias was seen in neutralizing or non-neutralizing antibody activities . The proportion of IgG subtypes in males versus females should be examined , as IgG3 V1V2-specific antibodies that mediate ADCC correlated with decreased risk of HIV infection in the RV144 trial , but exhibited a short half-life [38] . Recent development of reagents for use in subtyping macaque IgG should allow this question to be addressed . Additionally , Fc-receptor differences may exist between males and females . Polymorphisms in IgG Fc-receptors modulate antibody binding affinity for IgG subtypes , and affect antibody-dependent functions [39–40] . Moreover , differences in Fc glycosylation can affect antibody function [41] . Fucosylation modulates IgG1 binding to FcγRIIIa [42] . In the absence of fucose , binding is enhanced , resulting in improved ADCC activity [43] . A non-fucoslylated variant of bNAb 2G12 exhibited greater ADCVI activity against HIV and SHIV isolates [44] . Fc glycosylation differences also modulate ADCP activity [45] . Further , Fc agalactosylation and asialyation have been associated with better HIV control [46] . Differences in Fc glycosylation patterns between males and females have been established [47] and could have impacted our results . Delayed SIV acquisition in immunized females was greatest in gp120-immunized rather than gp140-immunized macaques that exhibited enhanced humoral immunity . Moreover , gp140-immunized animals met criteria for euthanasia earlier than gp120-immunized macaques although a sex bias was not observed . Although not excluding investigations of oligomeric gp140 , this result validates continued study of gp120 , the form of immunogen used in the RV144 trial , as a vaccine immunogen . The basis for the different outcome in gp120 immunized macaques while gp140 immunization appeared more immunogenic , however , is not known . Differences in antibody epitope specificities elicited by the different immunogens as well as IgG subtypes and Fc/Fc-R differences as discussed above might explain these outcomes . It may also be the case that higher antibody titers are not beneficial . This has been seen in other infectious diseases . For example , high titers and avidity of vaccine-elicited non-neutralizing antibodies against influenza have been associated with development of more severe disease [48] . Moreover , some non-neutralizing antibodies may be detrimental to protective efficacy . In the RV144 trial , V1V2-specific antibodies that mediate ADCC correlated with protection against acquisition , however high serum Env-specific IgA correlated with infection risk , possibly blocking protective ADCC responses [49] . We did not examine serum Env-specific IgA levels , but they warrant evaluation . Antibody-dependent enhancement of infection can also occur via complement and Fc receptors , dependent on antibody titer and receptor affinity [50] . Both FcγRIIa and FcγRIIIa receptor genetic polymorphisms increase receptor avidity for immune complexes [40] . Notably , the FcγRIIIa genotype was associated with HIV infection rate in the VAX004 trial [51] . Thus , genotyping receptors in females and males may also help explain our complex results . Viral loads exhibited in this study not unexpectedly inversely correlated with CD8+ T-cell responses . By immunization group , a significant correlation of these cellular responses with reduced viremia was only seen in gp140 immunized animals , perhaps due to additional epitopes present in gp140 . The gp140 immunized macaques also exhibited more persistent acute viremia reductions . It is possible that these cellular immune responses initially contributed to stronger acute viremia control in this immunization group while at the same time enhanced humoral immunity led to later detrimental effects as suggested above , resulting in the gp140 immunized macaques meeting euthanasia criteria earlier than the gp120 immunized animals . In this regard , significant correlations of CD8+ T cell responses with decreased viremia were exhibited in all male but not female macaques ( S18 Fig ) , perhaps reflecting a greater waning of vaccine-induced CD8+ T-cell responses during infection delay in females . This might have abbreviated the period of time during which the CD8 T cells were able to effectively control viremia . Among humoral responses , phagocytic activity 2wkpi correlated with decreased viremia in all macaques , but a significant correlation of ADCP prior to infection ( wk 53 ) was only present in gp140-immmunized macaques , a result possibly influenced by antibody quality as discussed above . The sex bias in immunity [52] , especially mucosal immunity [53–54] , is profound and can be attributed to both hormonal influences and contributions of X-linked genes . The microbiome plays a major role in shaping mucosal immune responses [55] and can impact mucosal infections . Steroid hormones can also modulate the microbiome , leading to distinct sex profiles [56] . Overall the microbiome composition is critical in HIV transmission and pathogenesis , can influence HIV acquisition [57] , and is a key area for further investigation of the sex bias in SIV acquisition . Female sex hormone changes throughout the menstrual cycle impact susceptibility to vaginal HIV infection by affecting all arms of the immune system . A “window of vulnerability” in the late secretory phase of the cycle during which risk of sexually transmitted infections is highest was postulated [58] , and corroborated by the demonstrations during the secretory phase of more frequent vaginal SHIV transmission to macaques [59] and better ex vivo HIV infection of human cervical explants [60] . We did not synchronize our female macaques , as rectal challenges were planned . However , fluctuations in female sex hormone levels could affect HIV/SIV acquisition by other than vaginal routes of exposure . Estrogen receptors ( ER ) are expressed by cells in a variety of tissues in addition to the reproductive tract . ERα is expressed in T and B lymphocytes , dendritic cells , macrophages , monocytes , natural killer cells and mast cells [61] , and influences intestinal levels of proinflammatory cytokines , including TNFα [62] . An examination of gut biopsies from men and women directly demonstrated that women have higher levels of immune activation and inflammation compared to men [53] . The profound effects of ERα on DC development and function greatly influence the quality of adaptive immune responses . ERβ is expressed predominantly in the brain , cardiovascular system , and colon and is found mainly on epithelial cells [63] . It plays an important role in cellular differentiation and maintenance of cellular homeostasis in the colon [64] . In addition , by suppressing chloride ion secretion across the colonic epithelium , estrogen controls fluid retention during different stages of the menstrual cycle [65] . Estrogen also increases mucin content of the protective mucus layer in the intestine and increases mucus viscosity and elasticity [66] . ERα and ERβ play different roles in controlling B cell maturation and selection . Engagement of both by estrogen can alter B cell maturation , whereas triggering of ERα influences development of autoimmunity [67] . In rhesus macaques the frequency of ASC in not only genital mucosal but also systemic lymphoid tissues , bone marrow , and PBMC exhibited profound changes throughout the menstrual cycle [68] . Overall , little is known regarding the influence of female sex hormones on other than vaginal viral exposures , however , as illustrated above , these hormones affect innate and adaptive immune responses , intestinal homeostasis and integrity , biophysical properties of protective mucus , and immune activation and inflammation in more than just reproductive tissue . Thus , it is reasonable to take into account potential hormonal effects in future vaccine studies . Our results showing a clear sex bias in vaccine challenge outcome correlated with local mucosal humoral immunity , is timely in view of recently formulated NIH policy requiring sex balancing in animal studies [19] . Such balancing will cause increased complexity in vaccine design and may require study of the microbiome and in-depth examination of immune responses beyond mere quantitation of functional activities . This approach may provide better understanding of vaccine protective mechanisms . The knowledge gained can be applied to future sex-balanced pre-clinical studies and clinical vaccine trials , critically important as women harbor ~50% of HIV infections worldwide [69] .
All animal experiments were approved by Institutional Animal Care and Use Committees prior to study initiation . During the course of this study , the study animals were housed in three facilities , each of which approved the work ( Bioqual , Inc . , Rockville , MD , Protocol No . 12-3507-15; Advanced BioScience Laboratories , Inc . ( ABL ) , Rockville , MD , Protocol No . AUP526; and the NCI Animal Facility , Bethesda , MD , Protocol No . VB007 ) . Each of these facilities is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International . The standard practices closely follow recommendations made in the Guide for the Care and Use of Laboratory Animals of the United States—National Institutes of Health . The rhesus macaques ( Macaca mulatta ) used in this study were housed in accordance with the recommendations of the AAALAC Standards and with the recommendations in the Guide for the Care and Use of Laboratory Animals . When immobilization was necessary , the animals were anesthetized with approximately 10 mg/kg of ketamine hydrochloride injected intramuscularly . All efforts were made to minimize discomfort of all animals used in the study , including provision of peri-operative and post-operative analgesia and strict accordance to humane endpoint criteria . Details of animal welfare and steps taken to ameliorate suffering were in accordance with the Guide and the recommendations of the Weatherall report , ‘‘The use of non-human primates in research” , as approved by the relevant IACUCs . Animals were housed in temperature controlled facilities with an ambient temperature of 21–26°C , a relative humidity of 30%– 70% and a 12 h light/dark cycle . Due to the nature of the experiment the animals were housed singly in stainless steel wire-bottomed cages and provided with a commercial primate diet and fresh fruit twice daily , with water freely available at all times . All animals were monitored twice daily for activity , food and water intake , and overall health . Enrichment in the form of rotating toys , visual and auditory stimuli , and foraging opportunities were provided daily . Animals that reached IACUC defined endpoints , including pain or distress , that could not be alleviated therapeutically , were humanely euthanized with an overdose of barbiturate consistent with the recommendations of the most recent American Veterinary Medical Association Panel on Euthanasia . Sixty Indian rhesus macaques ( Macaca mulatta ) aged 2 to 7 years and negative for SIV , SRV , and STLV were used in this study . Males and females ( see below ) were assigned to immunization and control groups to achieve similar mean ages , and balanced for Mamu A*01 and B*08 haplotypes ( 3 Mamu A*01 , 2 Mamu B*08 , and 1MamuA*01/B*08 ) . Experimental and control groups ( Fig 1A ) were divided in two for the vaccination phase of the study , and macaques were housed and handled at either Bioqual Inc , or ABL . The challenge of all 60 macaques was conducted at the ABL facility . Post-challenge monitoring after macaques had received up to nine challenges was carried out at the NCI Animal Facility . The number of macaques used was based on a power analysis which determined that using 18 additional historical controls previously challenged with the stock to be used , and the historical infection rate , the estimated power to detect differences between the experimental groups and the controls was 84% . Twenty four macaques were included in each immunization group and primed at weeks 0 ( intranasally and orally ) and 12 ( intratracheally ) with three replication-competent Ad5hr recombinants separately encoding SIVsmH4env ( gp140 ) /rev , SIV239gag and SIV239nefΔ1–13 ( Fig 1A ) . The recombinants were administered in PBS at 5 X 108pfu/dose/route as previously described [24] . The SIVmac239 monomeric gp120 and oligomeric gp140 boosting immunogens were produced in CHO cells , purified and characterized as previously described [70] , and administrated intramuscularly with MF59 adjuvant ( Novartis Vaccines and Diagnostics , Cambridge , MA ) at weeks 39 and 51 . The gp120 immunization group ( 16 females and 8 males ) received 100μg/dose of monomeric SIVmac239 gp120 in MF59 adjuvant , and the gp140 immunization group ( 16 females and 8 males ) received 100μg/dose of oligomeric SIVmac239 gp140 in MF59 . Control macaques ( 7 females and 5 males ) received equivalent doses of Ad5hrΔE3 empty vector and MF59 adjuvant only . At week 59 , all macaques were challenged intrarectally using a repeated low dose of SIVmac251 ( 1:500 dilution; 120 TCID50 ) , a challenge stock developed by Dr . Ronald Desrosiers and provided by Dr . Nancy Miller , Division of AIDS , NIAID . As SIV exposures were intrarectal , we did not synchronize females prior to initiating challenges . Challenges were continued weekly until the onset of infection determined by a plasma viral load of ≥50 SIV RNA copies/ml as assessed by the NASBA method [71–72] . Macaques were monitored for 40 weeks after infection or until euthanasia criteria were met . Samples from all macaques were included in each analysis except as specified in individual figure legends . Experiments were not blinded . By the time this study was completed , 53 additional historical control rhesus macaques , challenged intrarectally repeatedly with a 1:500 dilution of the same SIVmac251 stock , were available to provide greater statistical power for the analyses . Twenty-three of these macaques have been reported in previous publications [73 , 74] . Data on the remaining 30 have not yet been published . Additionally , rectal pinch biopsies were obtained at necropsy from 6 chronically SIV infected rhesus macaques for use in validating the staining of Env-specific rectal memory B cells . Serum antibody binding titers to monomeric SIVmac239 gp120 , oligomeric SIVmac239 gp140 ( Novartis ) and SIVsmH4 gp120 protein ( ABL ) were assessed by ELISA as described previously [75] . Antibody titer was defined as the reciprocal of the serum dilution at which the optical density ( OD ) of the test serum was two times greater than that of the negative-control serum diluted 1:50 . Binding antibody end point titers to variable region V2 of SIV gp120 Env were analyzed in serum samples collected prior to immunization and 2 weeks after the second protein boost ( wk 53 ) by ELISA as previously described [7] using a peroxidase-labeled γ chain specific goat anti-monkey IgG ( Catalog No 074-11-021 , KPL , Gaithersburg , MD ) and a custom-synthesized SIVmac251 cyclic V2 full-length peptide: CIAQNNCTGLEQEQMISCKFNMTGLKRDKTKEYNETWYSTDLVCEQGNSTDNESRCY ( JPT Peptide Technologies , GmbH , Berlin , Germany ) . Serum neutralizing antibody titers against SIVmac251 . 6 ( tier 1 ) and SIVmac251 . 30 ( tier 3 ) were assayed in TZM-bl cells as described [76] . Neutralizing titers were defined as the reciprocal serum dilution at which there was a 50% reduction in relative luminescence units compared to virus control wells which contained no test sample . Serum antibody-dependent cell-mediated cytotoxicity ( ADCC ) was evaluated using a rapid fluorometric assay [77] . Briefly , CEM-NKR cells coated with SIVmac239 gp120 or SIVmac239 gp140 ( Novartis ) were used as targets along with human effector PBMC at an effector-to-target ( E:T ) ratio of 50:1 , and serially diluted macaque sera . Controls included unstained and single-stained target cells . The percent ADCC cell killing was determined by back-gating on the PKH-26high population of targets cells that lost the CFSE viability dye . ADCC titers are defined as the reciprocal dilution at which the percent ADCC killing was greater than the mean percent killing of the negative controls plus three standard deviations . The maximum % killing for each serum was determined . Results were expressed as the 50% maximum killing titer: the reciprocal serum dilution at which 50% maximum killing was observed , and as endpoint titers . Antibody-dependent cellular phagocytosis ( ADCP ) activity was measured as previously described [78] , with minor modifications . Briefly , SIVmac239 gp120 or SIVmac239 gp140 was biotinylated with the Biotin-XX Microscale Protein Labeling Kit ( Life Technologies , Grand Island , NY ) , and 3–5 μg of gp120 or gp140 was incubated with a 100-fold dilution of 1μm Yellow-Green streptavidin-fluorescent beads ( Life Technologies ) for 25 min at room temperature in the dark . Serial dilutions of each serum sample ( 1:50 to 1:3000 ) were added to 250 , 000–300 , 000 THP-1 cells in wells of a 96-well U-bottom plate . The bead-gp120/gp140 mixture was further diluted 5-fold in RPMI 1640 medium containing 10% fetal bovine serum ( R10 ) and 50 μl was added to the cell/serum mixtures and incubated for 3 h at 37°C . Cells were then washed at low speed , fixed in 2% PFA , and assayed for fluorescent bead uptake by flow cytometry using a BD Biosciences LSRII . The phagocytic score of each sample was calculated as follows: ( % phagocytosis x MFI ) /106 . The values were standardized to background values ( cells and bead only without serum ) by dividing the phagocytic score of the test sample by the phagocytic score of the background sample . Rectal secretions were collected using cotton swabs and stored in 1 ml of PBS containing 0 . 1% bovine serum albumin , 0 . 01% thimerosal , and 750 Kallikrein inhibitor units of aprotinin at -70°C until analyzed . Samples were tested for blood contamination using Chemstrips 5 ( Boehringer Mannheim ) prior to assay . To remove fecal contaminant sample was passed through a 5μm PVDF microcentrifugal filter unit ( Millipore , Billerica , MA ) . Briefly , SIVgp120 and gp140-specific IgA and IgG antibodies were measured by ELISA as previously described [79 , 80] . Env-specific IgA and IgG standards derived from IgG-depleted pooled serum or purified serum IgG , respectively , obtained from SIVmac251-infected macaques and quantified as previously described [27] were used to generate standard curves . HRP-conjugated goat anti-monkey IgA and IgG ( Nordic Immunology ) and TMB substrate were used in sequential steps , followed by the addition of phosphoric acid prior to reading the OD at 450 nm . Total IgA and IgG antibodies were measured in each sample and used to standardize gp120 or gp140-specific IgA and IgG concentrations . Results are reported as Env-specific IgA or IgG/total IgG or IgA ( ng specific/μg total ) . Rectal biopsies were collected at different time points and single cell suspensions were obtained from fresh samples as previously described [21] . Cells obtained were stained with a mixture of fluorescent-conjugated monoclonal antibodies . Env-specific memory B cells were identified using a biotinylated SIVmac239 gp120 or gp140 with the Biotin-XX Microscale Protein Labeling Kit ( Life Technologies , Grand Island , NY ) followed by APC-conjugated Streptavidin ( Life Technologies ) as previously described [20] . Briefly , staining was carried out at 4°C in the presence of unconjugated anti-CD4 antibodies to block reactivity to CD4 . Representative gating is illustrated in S14A Fig . gp120/gp140-specific B cells were detected within the memory B cell subpopulation ( CD27+/-IgD- ) . Rectal plasmablasts and plasma cells were similarly assessed in fresh rectal biopsies as previously described [21] . Plasmablasts were identified as CD19+CD20+/-IgD-IRF4+CD138-HLA-DR+Ki67+ and plasma cells as CD19+CD20+/-IgD- IRF4+CD138+HLA-DR-Ki67- ( S14B Fig ) . The Env-specific memory B cell staining was validated using rectal pinch biopsies from 6 chronically SIV infected rhesus macaques ( not a part of this study ) in analyses by both flow cytometry and B cell ELISPOT . A significant correlation was obtained ( S14C Fig ) . Bone marrow samples were collected at different time points , and lymphocytes were purified as previously described [75] and frozen until analysis . Lymphocytes were thawed and both total and SIVgp120 or gp140-specific IgG and IgA secreting B cells were quantified by ELISpot as previously described [37] . Briefly , plasmablasts and plasma cells were quantified on unstimulated samples while memory B cells were enumerated following 3 days of polyclonal stimulation with CpG ( ODN-2006 ) ( Operon ) , 0 . 5 μg/ml recombinant human sCD40L ( Peprotech ) , and 50 ng/ml recombinant human IL-21 ( Peprotech ) . In both cases , Env-specific IgA and IgG antibody secreting cells ( ASC ) were standardized to the total number of IgA and IgG ASC and are reported as the percentage of SIVgp120 or gp140-specific ASC relative to the number of total ASC . Peripheral blood mononuclear cells ( PBMC ) were isolated from EDTA-treated blood by ficoll gradient [79] and frozen until assay . Cellular immune responses were evaluated by intracellular staining for SIV-specific IFN-γ , IL-2 and TNF-α cytokine secreting cells . After thawing , PBMC were stimulated with peptides at 1μg/ml final concentration . SIV peptide pools were made up of 15 mers overlapping by 11 amino acids and included EnvsmH4 ( Advanced BioScience Laboratories , Inc ) , Envmac239 , Gagmac239 ( AIDS Research Reference and Reagents Program ) and Nefmac251 . Control tubes included a non-stimulated and a Leucocyte activation Cocktail ( BD Pharmingen ) as a positive control . Anti-CD28 PE/Texas red ( clone CD28 . 2; Beckman Coulter ) and anti-CD49d ( clone 9F10; eBioscience ) were also added during stimulation along with a protein transport inhibitor ( BD Pharmingen ) . After 6h incubation at 37°C , cells were washed with PBS , then stained as previously described [33] with the following antibodies: Anti-CD4 PerCP/Cy5 . 5 ( clone L220 ) , Anti-CD8 Qdot655 ( clone RPA-T8 , eBioscience ) , and Anti-CD95 PE/Cy5 ( clone DX2 , eBioscience ) . A viability dye ( Life Technologies ) was added to the antibody cocktail to exclude dead cell background . Following incubation for 30 min at 4°C in the dark , intracellular staining was performed . Cells were washed twice , resuspended in 250μl fix/perm solution ( BD Pharmingen ) for 20 min at 4°C , washed twice with BD perm/wash buffer and resuspended in 100μl wash buffer plus the following antibodies: Anti-CD3 Pacific blue ( SP34-2 , BD Pharmingen ) , Anti- IFN-γ APC ( B27 , BD Pharmingen ) , Anti-TNF-α FITC ( Mab11 , BD Pharmingen ) and Anti-IL-2 PE ( MQ1-17H12 , BD Pharmingen ) . After 30 min at 4°C in the dark , cells were washed twice with BD perm/wash buffer and pellets were resuspended in 2% formaldehyde solution for acquisition on an LSRII . CD3+ T cells were used as a gate for CD4+ and CD8+ T cells , and each population was further divided into CD28+CD95+ central memory ( CM ) and CD28-CD95+ effector memory ( EM ) cells . The percent of cytokine-secreting cells in each memory cell subset was determined following subtraction of the values obtained with non-stimulated samples . Both subsets were summed to give the total memory ( TM ) T-cell population . Flow-cytometric analysis was performed using FlowJo V9 . 8 . 1 . ( ThreeStar , Ashland , OR ) . All tests of quantitative data are rank-based and thus distribution-free , so the weak assumptions of the tests are met . Rank-based tests do not require similar variances . Grouped , continuous , and discrete data were analyzed using methods appropriate to each of those types . The Wilcoxon rank-sum analysis was used to test for differences between immunization groups for binding antibody titers , neutralizing and non neutralizing antibody activities , rectal SIV-Env specific B cells , mucosal Env-specific IgG and IgA , Env-specific antibody secreting B cells and cytokine responses . The Wilcoxon signed-rank test was used to test for differences in paired samples within immunization groups . The Cochran-Armitage test was used to analyze V2 peptide titers and ADCC titers . The Spearman rank correlation test was used to assess the relationships of antibody and cellular responses with number of challenges and viral loads . Acquisition and survival data were analyzed using the exact logrank test . For all comparisons a two-sided p<0 . 05 was considered statistically significant . Adjustments for multiple comparisons were not made . Estimates of variation are provided as needed in individual figure legends . Analyses were conducted using SAS/STAT software version 9 . 3 and GraphPadPrism V6 . | Viral infections can have different disease courses in men and women . Following HIV infection , women generally exhibit lower viral loads and higher CD4 counts than men , but paradoxically progress faster to AIDS . Sex differences result from effects of X-linked genes and hormonal influences , and are believed to be largely based on immune response differences . Nevertheless , little is known about potential sex differences following vaccination . Here we report for the first time a sex bias in response to a SIV vaccine in rhesus macaques , showing that female animals were better protected against acquisition of SIV compared to males . The vaccine-induced immune responses that contributed to this better protection were viral-specific antibodies and immune antibody-secreting B cells , both at the local rectal site of SIV exposure . These results suggest that HIV/SIV vaccines should be better designed to target mucosal exposure sites . Additionally , they indicate that more vaccine studies should include animals of both sexes to address potential differences . Our study also illustrates that inclusion of both sexes can lead to greater complexity in vaccine trial outcomes , necessitating more in depth analyses . However , we believe sex balancing to be particularly important , as approximately 50% of HIV infections worldwide occur in women . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Mucosal B Cells Are Associated with Delayed SIV Acquisition in Vaccinated Female but Not Male Rhesus Macaques Following SIVmac251 Rectal Challenge |
Kaposi's sarcoma-associated herpesvirus ( KSHV ) interacts with human dermal endothelial cell surface tyrosine kinase EphrinA2 ( EphA2 ) and integrins ( α3β1 and αVβ3 ) in the lipid raft ( LR ) region , and EphA2 regulates macropinocytic virus entry by coordinating integrin-c-Cbl associated signaling . In contrast , KSHV enters human foreskin fibroblast ( HFF ) cells by LR-independent clathrin mediated endocytosis . The present studies conducted to identify the key molecules regulating KSHV entry in HFF cells showed that KSHV induces association with integrins ( αVβ5 , αVβ3 and α3β1 ) and EphA2 in non-LR regions early during infection and activates EphA2 , which in turn associates with phosphorylated c-Cbl , myosin IIA , FAK , Src , and PI3-K , as well as clathrin and its adaptor AP2 and effector Epsin-15 proteins . EphA2 knockdown significantly reduced these signal inductions , virus internalization and gene expression . c-Cbl knockdown ablated the c-Cbl mediated K63 type polyubiquitination of EphA2 and clathrin association with EphA2 and KSHV . Mutations in EphA2's tyrosine kinase domain ( TKD ) or sterile alpha motif ( SAM ) abolished its interaction with c-Cbl . Mutations in tyrosine kinase binding ( TKB ) or RING finger ( RF ) domains of c-Cbl resulted in very poor association of c-Cbl with EphA2 and decreased EphA2 polyubiquitination . These studies demonstrated the contributions of these domains in EphA2 and c-Cbl association , EphA2 polyubiquitination and virus-EphA2 internalization . Collectively , these results revealed for the first time that EphA2 influences the tyrosine phosphorylation of clathrin , the role of EphA2 in clathrin mediated endocytosis of a virus , and c-Cbl mediated EphA2 polyubiquitination directing KSHV entry in HFF cells via coordinated signal induction and progression of endocytic events , all of which suggest that targeting EphA2 and c-Cbl could block KSHV entry and infection .
During the initiation of infection of target cells , viruses bind to the cellular receptors and utilize a plethora of cellular signal molecules . The utilization of receptors , adaptors and signal molecules largely depends on the nature of the target cells [1] . Animal viruses can utilize different internalization and trafficking pathways that allow specific localization within the cells upon entry for a successful infection . Besides fusion of the viral envelope with the host plasma membrane , receptor mediated endocytosis , an essential biological process mediating cellular internalization events , is often exploited by many enveloped and non-enveloped viruses for their entry into target cells [2] , [3] . KSHV , etiologically associated with Kaposi's sarcoma ( KS ) , primary effusion lymphoma ( PEL ) and multi-centric Castleman's disease ( MCD ) , manifests a wide range of receptor ( s ) and signal molecules utilization that varies according to the target cell type , serving as an excellent model to determine virus entry associated events [4] , [5] , [6] . KSHV has a broad range of in vivo tropism of target cells such as B , endothelial , epithelial , fibroblast cells , CD34+ stem cell precursors of dendritic cells ( DCs ) , monocytes and macrophages [7] . Although KSHV-infected “spindle cells , ” are likely of endothelial origin , fibroblast cells are also found in the KS microenvironment , support de novo KSHV infection and represent the characteristic component of KS lesions [8] . Following de novo infection of skin-derived fibroblasts , KSHV induces the production of pro-inflammatory and pro-migratory factors and promotes endothelial cell invasion of extra cellular matrix ( ECM ) through paracrine mechanisms [9] . In addition , latent KSHV infection of oral cavity derived primary human fibroblasts enhances the secretion of KS-promoting cytokines and intrinsic invasiveness through VEGF-dependent mechanisms [10] , which highlight the potential role for KSHV-infected fibroblasts in promoting KS pathogenesis . KSHV entry into adherent target cells is a multi-step complex process , involving various viral envelope glycoproteins and multiple cell surface molecules , which overlaps with the induction of pre-existing host signal molecules followed by entry into the cytoplasm , release of viral capsid and transport towards the nucleus via dynein mediated transport along the KSHV induced acetylated thickened bundles of microtubules [6] . KSHV utilizes endocytosis for its entry into human endothelial cells , fibroblasts , B cells and monocytes with different modes of entry depending on cell type [6] , [7] . Actin-dependant macropinocytosis and lipid rafts ( LRs ) are utilized by KSHV to enter human microvascular dermal endothelial ( HMVEC-d ) cells , while LR-independent clathrin mediated endocytosis is used to enter primary foreskin fibroblast ( HFF ) cells [6] . Our earlier studies have also demonstrated that during its infection of HMVEC-d and HFF cells , KSHV first binds to cell surface heparan sulfate ( HS ) molecules and subsequently utilizes integrins α3β1 , αVβ3 , and αVβ5 [6] , [8] , [11] , [12] , [13] , [14] , [15] , [16] . Studies utilizing virus pre-incubated with soluble integrins and anti-integrin antibodies have shown that KSHV interaction with integrin induces a cascade of associated signal molecules such as FAK , Src , PI3-K , RhoGTPases , NF-κB and ERK1/2 which mediates virus endocytic entry , transport towards the nucleus and viral gene expression [6] , [8] , [11] , [12] , [13] , [14] , [15] , [16] . Our studies have demonstrated that KSHV macropinocytosis in endothelial ( HMVEC-d ) cells is controlled by c-Cbl , an adaptor protein , and actin-myosin II A [17] . c-Cbl promoted rapid integrin receptor ( s ) ( α3β1 and αVβ3 ) translocation into LRs , where monoubiquitination of translocated α3β1 and αVβ3 receptors facilitated KSHV internalization via macropinocytosis , while non-LR bound αVβ5 receptor was polyubiquitinated and targeted towards a non-infectious clathrin mediated pathway to lysosomes [18] . Further studies conducted to determine the potential candidate ( s ) that coupled KSHV induced integrin signaling with macropinocytosis identified the receptor tyrosine kinase EphA2 as a key molecule mediating KSHV entry [19] . Ephrins have been implicated as a hub for signaling events [20] and ephrin receptors control macropinocytosis and clathrin dependant endocytosis in various cell types [19] , [21] . However , the key molecules that regulate KSHV internalization in HFF cells were not characterized . In addition , whether the factor ( s ) involved in KSHV entry differ based on cell type leading to such varied modes of KSHV entry in endothelial and fibroblast cells is not known . Studies presented here demonstrate that EphA2 regulates KSHV entry into HFF cells via clathrin mediated endocytosis by coordinating integrin-associated signaling and endocytic events . EphA2 knockdown disrupted KSHV induced signal induction and impaired virus association with clathrin adaptor . Upon KSHV infection c-Cbl ubiquitin ligase with its tyrosine kinase binding ( TKB ) and RING domains facilitated the interaction with the tyrosine kinase ( TK ) and sterile alpha motif ( SAM ) domains of EphA2 , thereby promoting EphA2 polyubiquitination ( K63 type ) critical for clathrin mediated internalization of associated virus . These studies provide the first important insight into understanding the c-Cbl mediated modulation of EphA2 regulating clathrin mediated virus entry and productive infection .
Since endothelial and fibroblast cells harbor latent KSHV infection and represent cellular components of KS , it is relevant to characterize the entry components in these model systems [22] , [23] . Earlier studies revealed the involvement of EphA2 as one of the important entry receptors for KSHV in endothelial and epithelial cells [19] , [24] . To establish a potential role of EphA2 related events during KSHV infection of primary fibroblast cells , we first analyzed whether EphA2 lentivirus-encoding shRNAs which were shown earlier to be effective for maximal EphA2 knockdown [19] could inhibit viral entry . Compared to control shRNA transduced cells , HFF cells transduced with EphA2 shRNA #3 and #4 showed significant reduction in KSHV entry ( ∼70% and ∼80% , respectively ) as determined by measuring the internalized KSHV viral ( ORF73 ) DNA copies ( Figure 1A ) . The greater efficiency of EphA2 shRNA #4 led us to use this shRNA in our subsequent studies . We next determined the effect of EphA2 shRNA on KSHV latent gene expression in HFF cells since KSHV latency establishment in these cells represents successful virus infection [25] . Control shRNA or EphA2 shRNA-transduced HFF cells were infected with KSHV for 48 h and viral latent gene expression was determined by real-time RT-PCR for the latency associated ORF73 ( LANA-1 ) gene . Compared to control shRNA transduced cells , HFF cells transduced with EphA2 shRNA showed ≥90% inhibition of ORF73 gene expression ( Figure 1B ) . In conjunction with this we also observed ≥80% reduction in the punctate nuclear staining of LANA-1 at 48 h post infection ( p . i . ) in EphA2 shRNA transduced HFF cells compared to control sh-RNA cells ( Figure 1 , C1 and C2 ) . Taken together , these results suggested that inhibition of viral gene expression in EphA2 knockdown cells was most likely due to blocking entry into the fibroblast cells , and EphA2 plays a significant role in KSHV entry . EphA2 , a receptor tyrosine kinase ( RTK ) , is functionally activated through tyrosine phosphorylation essential for receptor signaling and internalization [26] , [27] . Since previous studies demonstrated that KSHV infection induced EphA2 phosphorylation in endothelial and epithelial cells [19] , [24] we determined the activation of EphA2 during primary KSHV infection of HFF cells . As shown in Figure 2A , EphA2 was phosphorylated as early as 5 min p . i . , increased at 10 min p . i . and reduced by 30 min p . i . . At 5 min and 10 min p . i , colocalization between EphA2 and KSHV ( glycoprotein gpK8 . 1A ) was observed in virus infected HFF cells compared to mock infected ( Un ) control ( Figure 2B ) . Since EphA2 interacted with integrin receptors during KSHV infection in endothelial cells [19] , we determined their associations in KSHV infected fibroblast cells through coimmunoprecipitation . We observed considerable interactions of EphA2 with integrins α3β1 , αVβ3 or αVβ5 in KSHV infected cells at 10 min p . i . which decreased by 30 min p . i . ( Figure 2C , first , third and fifth panels ) . However , there was no interaction between EphA2 and integrin β6 in virus infected HFF cells ( Figure 2B , seventh panel ) indicating KSHV induced specific association of those integrin molecules with EphA2 . Immunofluorescence analyses further supported these findings where we observed significant colocalizations of integrins αVβ3 or αVβ5 with EphA2 at the initial time of infection which were more pronounced at the peripheral regions of KSHV infected cell ( Figure 2 , D1 and D2; white arrows ) . We also observed similar colocalization between EphA2 and α3β1 intergrin ( data not shown ) . These results demonstrated that early during HFF cell infection , KSHV induces the activation of EphA2 which interacts with virus particles and entry associated integrin receptors . Our earlier studies have shown that KSHV utilizes clathrin dependant endocytosis to enter HFF cells without the involvement of lipid rafts [8] , [22] . However , the factors involved in clathrin-mediated internalization of KSHV were not determined . Since we observed the activation and association of EphA2 with virus particles and KSHV entry receptors , we hypothesized that KSHV-induced EphA2 could be modulating this internalization pathway in HFF cells . To test this hypothesis , we first identified the activities of major endocytic factors required for KSHV entry in HFF cells based on our earlier studies that c-Cbl , myosin IIA and clathrin are implicated in endocytosis of KSHV in various cell types [17] , [18] , [22] . Following virus infection , a 3- to 7-fold increase in c-Cbl phosphorylation and 11- to 17- fold increase in p-myosin IIA was observed at 5 , 10 , and 30 min p . i . , whereas a 14- to 18-fold increase in phosphorylation of clathrin heavy chain was observed at 5 and 10 min which was subsequently decreased at 30 min p . i . ( Figure 3A ) . There was no significant change in basal levels of c-Cbl , myosin IIA or clathrin which demonstrated that KSHV modulates activation of these entry effectors . Association of KSHV entry receptor integrins with EphA2 and activation of virus induced endocytic effectors led us to investigate whether EphA2 could also associate with those effector molecules essential for clathrin mediated virus entry . Lysates from mock infected HFF cells or infected with KSHV for 5–30 min were immunoprecipitated with anti-EphA2 antibody followed by Western blotting for c-Cbl , clathrin and clathrin adaptors AP2 and Epsin15 . In contrast to uninfected cells , KSHV infection increased the association of EphA2 with c-Cbl , clathrin , AP2 and Epsin15 , which maximized within 5 min p . i . and subsequently reduced by 10 and 30 min p . i . ( Figures 3 , C and D ) . When cells were infected for 5 min with KSHV pre-incubated with heparin ( 100 µg/ml ) , we observed the near absence of AP2 and Epsin15 interactions with EphA2 ( Figure 3D ) , which not only demonstrated the specificity but also suggested that KSHV infection induced association of EphA2 with endocytic effectors . Compared to uninfected cells we observed substantial colocalization of c-Cbl and myosin IIA in KSHV infected HFF cells from 5 and 10 min p . i . , which decreased by 30 min p . i . ( Figure 3E1 ) . This result demonstrated that this is a virus infection induced event . With respect to uninfected cells mean pixel intensities of colocalization between c-Cbl and myosin IIA were significantly higher in KSHV infected HFF cells at 5 and 10 min p . i . ( ≥81% ) which was less increased at 30 min p . i . ( ≥42% ) ( Figure 3E2 ) . Efficient tyrosine phosphorylation of EphA2 , c-Cbl and clathrin and the presence of c-Cbl and clathrin with EphA2 co-immunoprecipitates ( Figures 2A , 3B and 3C ) led us to further determine whether phosphorylated EphA2 associates with activated effector molecules . Compared to mock infection , early during KSHV infection ( 5 and 10 min p . i . ) , a significant colocalization was observed between p-EphA2 and p-c-Cbl or clathrin predominantly at the periphery of infected HFF cells ( Figures 4A and 4B , white arrows ) with concomitant reduction of association at 30 min p . i . ( Figures 4A and 4B ) . Taken together these studies suggested that c-Cbl , myosin IIA , and clathrin with its adaptor molecules AP2 and Epsin15 are recruited by activated EphA2 to form a KSHV-macromolecular internalization complex essential for clathrin mediated virus entry . The above results demonstrated the association and colocalization of EphA2 with KSHV entry receptors and endocytic effectors preferentially at the periphery of HFF cells early during infection . Together with previous studies which suggested that KSHV enters into HFF cells via clathrin mediated endocytosis without involving lipid rafts [22] prompted us to next assess the localization of EphA2 in KSHV infected HFF cells . In uninfected and virus infected cells , EphA2 was predominately located in the NLR portions of plasma membranes ( Figure 5A ) . In addition , compared to mock-infected NLR fractions , we observed about a 7-fold increase in EphA2 phosphorylation at 5 min p . i . which gradually decreased by 10 min and 30 min p . i . ( Figure 5B ) . To further confirm the significance of EphA2 in NLR associated KSHV entry , NLR fractions prepared from mock or KSHV infected HFF cells at 5 , 10 and 30 min p . i . were immunoprecipitated ( IP-ed ) with anti-αVβ5 or -αVβ3 antibodies . Western blotting of the IP-fractions for EphA2 demonstrated a strong association of EphA2 with these integrins at 5 min p . i . which gradually decreased by 10 or 30 min p . i . ( Figure 5C , first and third panel ) . Interestingly , a similar pattern of interaction was also observed between EphA2 and clathrin from NLR fractions of infected HFF cells ( Figure 5C , fifth panel ) . These results corroborated our early evidences in Figures 3C and 4B and suggested that EphA2 mediated KSHV entry in HFF is a non-LR associated event . Our earlier reports suggest that interaction of KSHV with integrins and other receptors activates the host's pre-existing signal molecules such as FAK , Src and PI3-K which are known to regulate endocytosis [6] . Efficient tyrosine phosphorylation of EphA2 and its association with the activated endocytic effectors c-Cbl , clathrin , clathrin adaptors and myosin IIA as revealed from our data prompted us to elucidate whether EphA2 interacts with these KSHV-induced signal molecules . Mock infected or KSHV-infected HFF cell lysates were IP-ed with anti-EphA2 antibodies followed by Western blotting for FAK , Src and PI3-K . Compared to mock infection , EphA2-immunoprecipitates showed increased association with FAK , Src and PI3-K-p85 as early as 5 min p . i . which decreased by 10 and 30 min p . i . ( Figure 6A ) . The association of EphA2 with FAK was greatly decreased in infection with heparin pretreated KSHV which demonstrated the specificity of virus induced association of these signal molecules with EphA2 ( Figure 6A ) . These were further verified by immunofluorescence analyses which demonstrated the colocalization of p-EphA2 with p-FAK ( Figure 6B1 ) , p-Src ( Figure 6B2 ) and p-PI3-K-p85 ( Figure 6B3 ) predominantly at the peripheral region of HFF cells early during KSHV infection ( Figures 6 B1 , B2 and B3 , yellow spots indicated by white arrows ) . These results strongly suggested that increased levels of pFAK , pSrc and pPI3-K molecules are recruited by activated EphA2 to the KSHV-macromolecular internalization signaling complex critical for endocytosis . The role of functional EphA2 in the regulation of KSHV entry associated signaling in relation to efficient virus internalization was ascertained by analyzing the effect of EphA2 knockdown on the activation of effectors such as c-Cbl , myosin IIA and clathrin as well as on selective activation and amplification of FAK , Src and PI3-K-p85 signal molecules . HFF cells transduced with control or EphA2 shRNA were infected with KSHV for 5 min and analyzed by Western blotting for the phosphorylation of the above molecules involved in entry associated events . Compared to control shRNA , ≥90% of EphA2 was knocked down with sh-EphA2 in HFF cells ( Figure 7A , seventh panel ) . While the uninfected cells showed minimal activation , control shRNA cells upon KSHV infection for 5 min showed 6 . 4 , 8 . 9 and 4 . 2-fold increase in p-FAK , p-Src and p-PI3-K-p85 levels , respectively ( Figure 7A ) , and 11 . 3 , 18 . 8 and 18 . 7-fold increase in phosphorylation of c-Cbl , myosin IIA and clathrin heavy chain , respectively ( Figure 7B ) . In contrast , knocking down EphA2 reduced the activation of FAK , Src and PI3-K by >60% ( 4 . 5 , 5 . 5 and 2 . 7-fold reduction , respectively ) ( Figure 7A ) and also reduced by >75% ( 7 . 7 fold ) , >90% ( 17 . 5 fold ) and >70% ( 10 . 9 fold ) in phosphorylation levels of c-Cbl , myosin IIA and clathrin , respectively ( Figure 7B ) . However , there were no marked changes in the respective total expression levels of these molecules which also showed that there were no off target effects by EphA2-shRNA . KSHV interaction with integrin in the absence of EphA2 could be the potential reason for the observed low levels of FAK , Src and PI3-K phosphorylation in EphA2 knockdown cells . Nevertheless , the observed significant reduction in p-FAK , p-Src , p-PI3-K , p-c-Cbl , p-myosin IIA and p-clathrin due to EphA2 knockdown suggested that EphA2 must be acting as the master coordinator to amplify these signals to regulate virus entry , and reduced levels of these signals could be the potential reason for the dramatic reduction in KSHV entry and infection . These studies clearly demonstrated the importance of functional EphA2 in the modulation of signal amplifications critical for c-Cbl-myosin- dependent clathrin mediated endocytosis of KSHV in HFF cells . Since KSHV-induced signaling is highly coupled to its endocytosis [6] and as we observed EphA2 dependent activation and association of endocytic effectors ( c-Cbl , myosin IIA and clathrin ) and regulation of KSHV-induced signaling , we further characterized the role of EphA2 in endocytic trafficking of KSHV via a clathrin dependent route . We tracked the internalization of KSHV in EphA2 knockdown HFF cells with respect to control shRNA transduced cells . Within 5 min p . i . , a double labeled immunofluorescence assay showed substantial colocalization of KSHV particles with clathrin adaptor AP2 in control shRNA transduced cells ( Figure 8A1 , third panel , white arrows ) with ≥80% colocalization frequency ( Figure 8A2 ) which is similar to virus only infected positive controls ( Figure 8A , second panel , Unt; white arrows ) . However , in shEphA2 transduced cells , such colocalization was not significantly present ( Figure 8A1 , bottom panel ) ( <15% colocalization frequency , p<0 . 01 , Figure 8A2 ) suggesting that initiation of clathrin mediated endocytosis of KSHV is mostly dependent on EphA2 . A similar colocalization was also observed within 5 min p . i . between KSHV particles and Alexa 594-conjugated transferrin , a physiological ligand known to enter cells specifically via clathrin mediated endocytosis , in control shRNA transduced or untransduced cells infected with KSHV ( Figure 8B1 , Unt; white arrows ) ( ≥80% colocalization frequency , Figure 8B2 ) which was less evident in virus infected EphA2-shRNA transduced cells ( Figure 8B1 , bottom panel ) ( <15% colocalization frequency , p<0 . 01 , Figure 8B2 ) . This result suggested that KSHV was internalized with clathrin dependent endocytic marker transferrin and further validated the involvement of EphA2 in clathrin mediated entry of KSHV in HFF cells . We next investigated the role of EphA2 in KSHV trafficking by tracking its uptake into Rab5 positive endosomes in control and EphA2 shRNA-transduced HFF cells . In control shRNA transduced cells KSHV localized predominantly in Rab5 positive endosomes which is similar to untransduced virus infected positive control cells ( Figure 8 C1 , Unt; white arrows ) with >80% colocalization frequency ( Figure 8C2 ) . In contrast , cells transduced with EphA2 shRNA showed significantly reduced ( <20% ) colocalization ( p<0 . 01 ) of virus with Rab5 ( Figures 8 C1 and C2 ) . Taken together , these studies demonstrated that EphA2 clearly has a positive role in clathrin mediated entry of KSHV and its trafficking towards early endosomes presumably for productive infection . In endothelial cells , KSHV induced the association of EphA2 with c-Cbl , a critical endocytic effector molecule with a well characterized E3-ubiquitin ligase activity [28] . It is also well documented that many receptors , including receptor tyrosine kinases , undergo ligand-dependent ubiquitination mediated by c-Cbl [29] , [30] . Receptor ubiquitination has been recognized as an internalization signal . Moreover , the fate of internalized receptors also varies with the pattern of ubiquitination [31] , [32] . Furthermore , our studies already showed that EphA2 interacts with the adaptor molecule Epsin15 ( Figure 3 ) which is known to mediate the interaction between ubiquitinated cargo molecules and clathrin , an inducer of internalization [33] , [34] . Therefore , we theorized that early during KSHV infection of HFF cells , EphA2 may undergo c-Cbl mediated ubiquitination to assist in clathrin-dependent virus internalization . To test this hypothesis , we downregulated the expression of c-Cbl in HFF cells by si-RNA and observed >90% inhibition in c-Cbl expression relative to controls ( Figure 9A , upper two panels ) . Next we checked the ubiquitination of EphA2 after immunoprecipitation of the uninfected , control or si-c-Cbl treated KSHV infected HFF whole cell lysates with EphA2 antibody followed by Western blotting with P4D1 monoclonal antibody known to efficiently recognize both mono- and polyubiquitin [35] . We also used FK-1 monoclonal antibody which specifically detects polyubiquitin but not monoubiquitin [35] . Cell lysates prepared under very stringent conditions as described in the materials and methods section were used in these experiments to detect ubiquitination of proteins . Compared to uninfected cells , as early as 5 min p . i . in infected cells , increased ubiquitination of EphA2s was observed with P4D1 antibody in control si-RNA treated cells , which was similar to cells infected with KSHV only ( Figure 9A , third panel , lanes 2 and 3 ) . In contrast , very poor or negligible ubiquitination was detected in si-c-Cbl treated HFF cells ( Figure 9A , third panel , lane 4 ) . When experiments were conducted to discriminate the nature of ubiquitination , a similar result was observed with polyubiquitin specific FK-1 antibody where an intense smeary pattern resembling polyubiquitination was seen in control si-RNA treated KSHV infected cells similar to virus only infection ( Unt ) , whereas negligible or no ubiquitination was observed in si-c-Cbl treated cells ( Figure 9A , fourth panel , lanes 2 and 3 vs . lane 4 ) . This result clearly demonstrated the involvement of c-Cbl promoting polyubiquitination of EphA2 early during KSHV infection . To prove that EphA2 polyubiquitination is a result of KSHV infection , we used heparin-treated KSHV for infection . As expected , EphA2 ubiquitination was significantly reduced by heparin treatment of KSHV ( Figure 9A , third and fourth panels , lane 5 ) , suggesting that KSHV binding in infected cells triggers polyubiquitination of the EphA2 receptor . The typical polyubiquitin chains reported include either K48-linked chains which have been studied in the context of protein degradation [36] or K63-linked chains which have been implicated in a variety of non-proteolytic functions including mediators of novel signaling [37] . Therefore , we examined the composition of the EphA2 polyubiquitin chains in a similar experiment as before by probing with monoclonal antibodies specific for either K48–linked or K63-linked polyubiquitin chains . As shown in Figure 9B second panel , the Lys63 linkage but not the Lys48 linkage ( Figure 9B , first panel ) was most strongly detected in control siRNA treated cells similar to KSHV infection alone ( 5 min p . i . ;Unt ) but negligible or very low levels in cells treated with c-Cbl siRNA . We also observed faint lys 48-linked polyubiquitination of EphA2 ( Figure 9 B , first panel ) which could be due to autouiquitination of c-Cbl and/or due to the induction of EphA2 by its ephA1 ligand . Nevertheless , these results clearly suggested Lys63-linked polyubiquitination of EphA2 directed by c-Cbl early during KSHV infection . In search of the regions responsible for physical association between EphA2 and c-Cbl that have functional relevance with EphA2 ubiquitination we used specific deletion mutants of Myc-tagged EphA2 namely EphA2-Kinase Dead ( Eph-ΔKD ) and EphA2-sterile alpha motif ( Eph-ΔSAM ) mutants , and HA-tagged c-Cbl mutants such as c-Cbl RING Finger ( RF ) domain mutant ( c-Cbl-C3HC4C5 ) and c-Cbl Tyrosine Kinase Binding ( TKB ) mutant ( c-Cbl-G305E ) along with their respective Wild-type ( Wt ) constructs ( Figure 9C ) . Since HFF cells are not easily transfectable , we used 293 cells for this study . Each of the Myc-tagged EphA2 constructs was cotransfected with each of the HA-tagged c-Cbl constructs and the cell lysates from the KSHV infected or uninfected cells were used for ubiquitination analysis of Myc-EphA2 . Cell lysates prepared under stringent conditions were immunoprecipitated using anti-Myc antibody followed by immunoblotting with polyubiquitin specific FK1 antibody ( Fig . 9D , first and second panels ) . To delineate the interaction between Myc-EphA2 and HA-cCbl , lysates from HEK293 cells overexpressing different combinations of Myc-and HA-tagged Wt and mutant proteins were subjected to coimmunoprecipitation with anti-Myc antibody followed by western blot with anti-HA antibody ( ( Fig . 9D , third panel ) . Expression of Wt and deletion mutant EphA2 proteins was verified by immunoblotting with Myc antibody of the same blot ( Figure 9D , panel 4 ) , and expression of Wt and mutant c-Cbl proteins was verified by immunoblotting the cell lysates with anti-HA antibody ( Figure 9D , panel 5 ) . Strong polyubiquitination was detected only in Wt-EphA2-Myc precipitated with Wt-HA-c-Cbl ( Figure 9D , second panel , lane 3 ) . For clarity , a low exposure of the same blot is shown in which we observed the smeared bands of Myc-tagged EphA2 ( Figure 9D , first panel , lane 3 ) . In contrast , negligible or no polyubiquitination was observed in each of the other EphA2 precipitates ( Figure 9D , first and second panels , lanes 4–11 ) compared to the negative control ( Myc-+HA- vectors only ) ( Figure 9D , first and second panels , lanes 1–2 ) . Interestingly , only Wt-c-Cbl appeared to be coimmunoprecipitated with Wt-EphA2 ( Figure 9D , third panel , lane 3 ) but very little or no c-Cbl , whether Wt or c-Cbl-C3HC4C5 or c-Cbl-G306E , was observed to be coimmunoprecipitated with EphA2-ΔKD or EphA2-ΔSAM ( Figure 9D , third panel , lanes 6–11 ) . Similarly , very poor or no association of Wt-EphA2 was detected with c-Cbl-C3HC4C5 or c-Cbl-G306E ( Figure 9D , third panel , lanes 4–5 ) . The intense polyubiquitinated Myc-EphA2 bands observed in Figure 9D , first and second panel lane 3 could be due to strong interaction between overexpressed EphA2 and c-Cbl upon KSHV infection . This is similar to the intense smeary bands of polyubiquitinated bands of EphA2 seen in Fig . 9 A and B . These results demonstrated that physical association of EphA2 with c-Cbl occurs probably with the aid of both the EphA2 tyrosine kinase ( TK ) and SAM domains and facilitated by the presence c-Cbl RF and TKB domains . Other than the Wt-EphA2 and c-Cbl cotransfection , the absence of polyubiquitination in KSHV infected cells corresponding to cotransfection with each of the other Myc-EphA2 and HA-c-Cbl mutants clearly suggested the functional significance of c-Cbl mediated KSHV induced polyubiquitination of EphA2 involving association of c-Cbl –TKB and –RF domain along with EphA2 TK and SAM domains . The fact that receptor ubiquitination serves as a better internalization signal for clathrin adaptors during clathrin mediated endocytosis [38] , [39] , [40] and c-Cbl directed polyubiquitination of EphA2 in our results prompted us to investigate the functional implications of this event in clathrin mediated endocytosis . Therefore , we verified the consequence of c-Cbl knockdown affecting the association of EphA2 with clathrin during KSHV infection of HFF cells . Upon KSHV infection ( 5 min ) , when cell lysates of HFF cells pretreated with control or c-Cbl siRNA were immunoprecipitated using anti-EphA2 antibody , strong co-IP of EphA2 and clathrin heavy chain was observed in control siRNA treated cells similar to virus only infected cells ( Figure 10A , upper panel , lanes 2 and 3 ) . However , such co-IP was significantly reduced ( ∼8 fold ) in cells treated with c-Cbl siRNA ( Figure 10A , upper panel , lane 4 ) . Consistent with this result , in immunofluorescence studies , knockdown of c-Cbl showed no appreciable colocalization between p-EphA2 and clathrin as evident from the non-coherent pattern of green ( p-EphA2 ) and red ( clathrin ) signals ( Figure 10B , panel 4 ) . In contrast , substantial colocalization of EphA2 with clathrin was observed in cells treated with control siRNA similar to virus only infection ( Figure 10B , panels 2 ( Unt ) and 3 ( si-cont ) , white arrows ) . Altogether , these results clearly demonstrated the active participation of c-Cbl in regulating KSHV induced association of EphA2 with clathrin and thus influencing clathrin mediated endocytosis . We rationalized that if EphA2 that is critically associated with KSHV entry undergoes c-Cbl directed polyubiquitination , and c-Cbl knockdown disrupts association between EphA2 and clathrin , then c-Cbl ‘loss of function’ should also impair clathrin mediated endocytosis of KSHV in HFF cells . Hence , we sought to characterize the functional effect of c-Cbl on the association of KSHV with clathrin upon c-Cbl knockdown in HFF cells . KSHV colocalized appreciably with clathrin by 5 min . p . i with some viruses located inside in control siRNA treated cells similar to virus only infected cells ( Figure 11A1 , second ( Unt ) and third ( si-cont ) panels ) ( ≥80% colocalization frequency , Figure 11A2 ) . In contrast , no significant colocalization between KSHV and clathrin was observed on c-Cbl siRNA treatment as suggested from the unsynchronized green and red signals ( Figure 11A1 , fourth panel ) ( <15% colocalization frequency , Fiure 11A2 ) ; moreover , the majority of the viruses were present at the outer cell periphery ( Figure 11A1 , fourth panel ) . However , we still observed significant colocalization between EphA2 with KSHV ( Figure 11B1 ) even in c-Cbl knockdown HFF cells , similar to control siRNA treatment or virus only infection in the absence of any si-RNA as validated from synchronous green ( EphA2 ) and red ( KSHV ) signal ( Figure 11B1 , second , third and fourth panel ) ( ∼80% colocalization frequency , Figure 11B2 ) . These results demonstrated the potential involvement of c-Cbl in regulating clathrin mediated KSHV entry into HFF cells without disturbing the association of KSHV with EphA2 , one of its receptors , and suggested that interaction of virus with EphA2 is followed by subsequent c-Cbl directed clathrin dependent entry . Given the fact that c-Cbl knockdown impaired ( i ) EphA2 polyubiquitination ( K63 type ) , ( ii ) association of EphA2 with clathrin , ( iii ) association of KSHV with clathrin without any disturbance of prior interaction of KSHV with EphA2 and the observation that EphA2 knockdown resulted in ∼80% KSHV entry inhibition but >90% inhibition of viral gene expression , we speculated whether c-Cbl directed EphA2 polyubiquitination may be registered as an internalization signal for clathrin mediated virus entry into HFF cells . Since we observed ∼15–20% entry of input virus in the presence of EphA2sh-RNA , it prompted us to investigate the fate of the entered viral particles in the context of c-Cbl knockdown . We tracked the internalized KSHV particles , by 30 min . p . i . , in control and si-c-CblRNA treated HFF cells with LysoTracker , a basophilic lysosomal marker . While in si-control or virus only infected cells , KSHV ( red ) did not colocalize with LysoTracker ( green ) ( Figure 11C , second and third panels ) , in c-Cbl knockdown cells , the majority of the virus particles remained at the cell periphery and the very minimal amount of internalized virus was found to colocalize with LysoTracker . This result suggested entry and trafficking of virus towards the lysosome and consequently a nonproductive infection possibly through a non-c-Cbl mediated mechanism . Taken together , our results clearly demonstrated that c-Cbl dependent EphA2 polyubiquitination is essential for clathrin mediated entry of KSHV in HFF cells which results in a productive infection leading to establishment of latency .
The Eph family of receptor tyrosine kinases ( RTK ) and their membrane bound ligands , known as Ephrins , exert bidirectional signaling where signaling through ligands is “reverse signaling” and through Eph receptors represents “forward signaling” [41] . They are known to mediate diverse activities such as integrin-associated signaling , effects on the actin cytoskeleton , cell-substrate adhesion , intracellular junctions , cell shape and cell movement [41] , [42] , [43] , [44] , with broad implications in neovascularization and oncogenesis [45] , [46] . Ephrin receptors have also been shown to be the center of signaling crosstalk between integrins , PI3-K , and Rho-GTPases [20] , [47] , which are also induced early during KSHV infection [6] . Our current study shows that EphA2 is predominantly a non-LR partitioned molecule in HFF cells which is activated by KSHV infection in association with KSHV entry receptors ( integrins α3β1 , αVβ3 , αVβ5 ) at the cell periphery . Very poor or absence of integrin association with EphA2 in uninfected cells signifies this to be a virus induced event . This finding is consistent with the previous reports where Ephrin-EphR has been shown to be involved in regulating an assortment of integrin signaling pathways [42] , [43] , [47] , [48] . Our studies demonstrating the activation of c-Cbl , myosin IIA along with clathrin , substantial colocalization of c-Cbl and myosin and the association of EphA2 with c-Cbl and clathrin during KSHV infection particularly at early time points ( 5 min and 10 min p . i . ) indicate an active macromolecular complex comprised of integrin-c-Cbl-myosin-EphA2 , critical for clathrin mediated internalization of KSHV . The formation of clathrin-coated pits ( CCP ) at the plasma membrane during the process of receptor-mediated endocytosis requires the interaction of clathrin with adaptors which serve important roles in the regulatory and catalytic mechanism at specific stages in clathrin mediated endocytosis [49] . Clathrin is recruited to CCP by AP2 [50] , which has two large subunits involved in the recruitment of accessory proteins [51] . One such protein is Epsin15 , which is required for CCP assembly and invagination during CME [52] , [53] . Colocalization of activated EphA2 with clathrin ( Figure 4B ) and association of EphA2 with AP2 and Epsin15 ( Figure 3D ) very early during KSHV infection further provides strong evidence that the initial internalization of KSHV may proceed via EphA2 dependent clathrin mediated endocytosis . This observation is consistent with the previous finding where ligand induced EphA2 has been implicated in endocytosis and receptor clustering [54] , however , involvement of EphA2 in clathrin mediated endocytosis of a virus was not reported before . Predominant activation of clathrin and its association with EphA2 very early during virus infection with concomitant decrease overtime indicates that the process is rapid and occurs at the very initial stage of infection . Since the non-LR region constitutes most of the cellular area enriched with all of the KSHV entry receptors , distribution of EphA2 at the non-LR region of HFF cells probably provides a good platform for the initiation of infection . Selective modulation of cellular signaling is crucial for endocytosis regulation [55] . However , there is little information regarding the molecules involved in signal assembly and regulation of virus endocytosis . We observed significant association of EphA2 with FAK , Src and PI3-K as early as 5 min post KSHV infection in HFF cells which decreased from 10 min and 30 min p . i . indicating EphA2 mediated recruitment of these signal molecules . EphA2 knockdown significantly reduced not only the activation of FAK , Src and PI3-K signal molecules but also the endocytic adaptors c-Cbl , myosin IIA and clathrin . Though tyrosine phosphorylation of clathrin heavy chain was implicated in the internalization of bacteria [56] , our studies report for the first time that EphA2 influences the tyrosine phosphorylation of clathrin heavy chain during KSHV infection . EphA2 mediated regulation of these molecules along with clathrin further suggests that EphA2 is a master regulator of signal molecules during KSHV entry . Since Src has been shown to be involved in the activation of clathrin and AP2 during clathrin mediated endocytosis [57] , the role of EphA2 in the amplification of KSHV-induced Src activation early during KSHV infection suggests that this amplification is essential for clathrin/AP2 activation and formation of CCP . As shown by our previous studies and in the present study , although KSHV induces the same key molecules during entry of HMVEC-d and HFF cells , entry into fibroblast cells and into endothelial cells involves two different endocytic pathways . This is potentially due to where EphA2 is distributed in these cells . EphA2 is detected strictly in the LRs of HMVEC-d cells [19] . However , in HFF , EphA2 is distributed mostly in the nonLR region . Our previous studies show that EphA2 interaction is vital for virus entry in HMVEC-d cells and KSHV induced c-Cbl promotes the rapid virus bound integrin receptors translocation into LRs of infected cells where KSHV interaction with EphA2 amplifies and couples integrin associated Src and PI-3K signaling with macropinocytosis [19] . In contrast , in our current study since EphA2 is predominantly distributed in the non-LR region of HFF cells , there is probably no need for receptor translocation and KSHV induced interaction of integrin with EphA2 enhances the FAK , Src , PI-3K , c-Cbl and clathrin activation in nonLR region followed by clathrin mediated endocytosis ( Fig . 12 ) . Ubiquitin ( Ub ) modification of membrane proteins appears to be one of the preferred sorting signals for clathrin mediated internalization where Ub represents a binding surface to clathrin effectors like Epsin or Epsin15 [33] , [58] . However , the mechanism of how ubiquitination of viral particles bound to cell surface receptors affects productive viral entry and infection remains to be clarified . c-Cbl E3 ubiquitin ligase not only plays important roles in signal transduction as negative regulators of receptor-based signaling by promoting ubiquitination but also acts as a part of a complex signaling interactome controlling a variety of functions , including internalization and endosomal sorting [30] , [59] and as an endocytic modulator during KSHV entry in endothelial cells [18] . In HFF cells , KSHV infection induces the association of EphA2 with c-Cbl very early during infection which directs the K63 specific rather than K48 type of polyubiquitination of EphA2 . This is relevant in the context of signaling and endocytosis as K63 linked polyubiquitination has been mostly implicated in a variety of non-proteolytic functions including the endocytosis pathway [37] , [60] whereas K48-linked chains have been implicated in protein degradation [36] . Many of the proteins that undergo Ub-dependent internalization are modified by K63-polyUb and some of the key cargo-modifying Ub ligases favor the formation of K63-linked chains probably because a single Ub appended to a model protein may serve as a rather poor internalization signal . For example , TrkA ( tropomyosin receptor kinase A , the nerve growth factor receptor ) , is modified by TRAF6 ( tumor necrosis factor receptor-associated factor ) , an enzyme that favors formation of K63 chains [61]; MHC-I , is modified by KSHV encoded K3-MARCH ( membrane-associated RING-CH ) ligase and the Ub conjugating enzymes UbcH5/Ubc13 , which favor K63-linked chains [62] . Importantly , perturbing the Ub system leading into the inhibition of the formation of K63 chains diminishes internalization . For instance , over expression of mutant UbK63R , which interferes with the formation of K63-polyUb , attenuates internalization of the prolactin receptor , TrkA , and MHC-I [61] , [62] . Therefore , the more Ub a cargo carries , the more effective its internalization–most probably because this promotes the interaction of cargo with Ub-sorting receptors , which typically have poor affinity for a single Ub . Formation of a polyUb chain may simply be the simplest mechanism for achieving good binding , especially when a limited number of lysine residues are available on relevant cargo . These events could also be effective in our study and c-Cbl could preferentially participate in K-63 liked polyubiquitination of EphA2 for efficient sorting by Ub-sorting proteins like Epsin 15 associated in clathrin mediated endocytosis of KSHV similar to EGFR endocytosis [63] . Non-association between EphA2-ΔKD and c-Cbl-TKB ( c-Cbl-G306E ) mutants ( Figure 9D ) together with the observed colocalization of p-EphA2 with p-c-Cbl ( Figure 4A ) clearly suggested the need for tyrosine phosphorylation in the tyrosine kinase domain of EphA2 to facilitate its association with the c-Cbl TKB domain . The question of how these domain interactions aid EphA2 polyubiquitination needs additional detailed analysis . Nevertheless , since the SAM domain of EphA2 is known to help dimerization of EphA2 and thus activates its tyrosine kinase activity necessary for EphA2 functionality , c-Cbl TKB domain in the presence of the RING domain could be contributing to the association with the tyrosine kinase domain of dimerized EphA2 leading into its polyubiquitination . The human EphA2 protein with 976 aa length has >30 lysine ( K ) residues in its cytosolic region ( aa 559–976 ) spreading over kinase domain , SAM domain and PDZ domain . The minimal ubiquitination of truncated EphA2 as found in figure 9D may probably be due to the presence of lysine residue/s other than that specific mutated region . To determine the specific lysine residue/s undergoing polyubiquitination , extensive mutational and functional analysis are required which could be more intricate for a separate study and is beyond the scope of the current study . However , since c-Cbl knockdown impairs EphA2 polyubiquitination ( Figures 9 A and B ) and EphA2 association with clathrin ( Figures 10 A and B ) and polyubiquitination appears to be one of the preferred sorting signals for clathrin mediated internalization [33] , [58] , our data suggest the role of c-Cbl induced polyubiquitination of EphA2 coordinated with EphA2 dependent modulation of KSHV-integrin interaction induced entry associated signaling as essential event for clathrin mediated KSHV entry . Our studies show that c-Cbl mediated K63 type but not degradative K48 type polyubiquitination of EphA2 acts as an internalization signal for clathrin mediated virus entry . Sorting of ubiquitinated cargo from early to the late endosome and lysosome relies on endosomal sorting machinery known as ESCRT ( endosomal sorting complex required for transport ) [64] . Moreover , Ub may also be used as a sorting signal that gives proteins access to parts of the endocytic system without necessarily being degraded in lysosomes . For example , Ub might label receptors for transport to ‘signaling endosomes’ , which would allow them to efficiently stimulate downstream signaling pathways . This could operate by allowing DUBs ( deubiquitination enzymes ) to intervene along the trafficking pathway to prevent efficient transport into lysosomes [65] . Therefore , even as cargoes are being sorted by ESCRTs , either last minute Ub conjugation with different E3 ligases or mutual regulation of Ub and DUBs ( deubiquitination enzymes ) for deubiquitination determines the fate of sorted molecules [66] . The sorting of specific cargoes can be influenced greatly by a particular polyUb chain , or that a polyUb chain can provide a set of powerful regulatory opportunities as evident from our study . Further studies are required to determine whether ESCRT and ESCRT-associated proteins play roles in KSHV infection . Overall , our studies reveal EphA2 as a critical non-LR region associated molecule in fibroblast cells which coordinates the signals induced during interaction of KSHV with its entry receptors and regulates the endocytic events leading to clathrin mediated entry and productive infection ( Figure 12 ) . These results together with a similar role of EphA2 in LR-dependent KSHV entry in dermal endothelial cells [19] , suggest a global or broad-spectrum effect of EphA2 in different cell types . These studies suggest that EphA2 and c-Cbl are promising therapeutic targets to control the initial stage of KSHV infection of endothelial , epithelial and fibroblast cells .
Primary human foreskin fibroblast ( HFF ) cells ( Clonetics , Walkersville , MD ) and human embryonic kidney epithelial ( HEK ) 293 cells were grown as described before [11] , [17] , [67] . Induction of the KSHV lytic cycle in BCBL-1 cells , supernatant collection , and virus purification procedures were described previously [11] . KSHV DNA was extracted and copies were quantitated by real-time DNA PCR using primers amplifying the KSHV ORF 73 gene as described previously [25] . Mouse anti-integrin α3β1 ( Mab 1992 ) , αVβ5 ( Mab 2019Z ) , and αVβ3 ( Mab 1976 ) , and mouse anti-phosphotyrosine ( p-tyr ) ( 4G10 clone ) antibodies were purchased from Chemicon International , Temecula , CA . Goat anti-β6 integrin antibody was purchased from Santa Cruz Biotechnology , CA . Mouse anti-c-Cbl , anti-phospho c-Cbl pY731 ( phosphotyrosine ) and clathrin antibodies were from BD Biosciences , San Diego , CA . Rabbit anti-caveolin-1 antibody , DAPI , Alexa 488 conjugated LysoTracker , Alexa 594 conjugated transferrin , Alexa 594 or 488 anti-rabbit and anti-mouse secondary antibodies were from Molecular Probes , Invitrogen , Carlsbad , CA . Protein A-Sepharose 6 MB and Protein G-Sepharose CL-4B were from Amersham Pharmacia Biotech , Piscataway , NJ . Anti-rabbit and anti-mouse antibodies linked to horse-radish peroxidase were from KPL Inc . , Gaithersburg , MD . CD-71 hybridoma cell line was from American Type Culture Collection ( ATCC ) , Manassas , VA . CD-71 Mab secreted in culture medium was purified by Protein-A-Sepharose affinity chromatography . Mouse monoclonal anti-KSHV gpK8 . 1A ( 4A4 ) antibody and rabbit monoclonal anti-KSHV gB were generated in our laboratory [68] . Rabbit anti-HA was from Zymed , Invitrogen , Carlsbad , CA and mouse anti-ubiquitin ( P4D1 ) was from Santa Cruz , CA . Mouse anti-polyubiquitin FK-1 , anti-K63 and -K48 monoclonal Abs were from Millipore , Temecula , CA . Rab5 , EphA2 , phospho-EphA2 , phospho-Src , phospho-myosin IIA , myosin IIA , phospho-PI3-K , Epsin15 and AP-2 antibodies were obtained from Cell Signaling Technology , Danvers , MA . Heparin , other fine chemicals and buffers were purchased from Sigma , St Louis , MO . HFF cells were transduced by lentiviruses encoding shEphA2 as described previously [19] . For validation of shRNA constructs , HEK293 cells were cotransfected with EphA2 expression plasmid ( target ) and shRNA lentiviral vector constructs . The construction and production of lentiviral gene transfer vectors were done as previously described [67] . Western blot analysis was performed to confirm the level of knockdown . Cells were lysed in RIPA buffer ( 15 mM NaCl , 1 mM MgCl2 , 1 mM MnCl2 , 2 mM CaCl2 , 2 mM phenylmethylsulfonyl fluoride and protease and phosphatase inhibitor mixture ) and incubated on a rocker at 4°C for 15 min . Lysates were normalized to equal amounts of protein , boiled in sample buffer , separated by 10% SDS-PAGE , transferred to nitrocellulose and probed with the indicated primary antibodies . Detection was by incubation with species-specific HRP-conjugated secondary antibodies . Immunoreactive bands were visualized by enhanced chemiluminescence ( Pierce , Rockford , IL ) . The bands were scanned and quantitated using FluorChemFC2 and Alpha-Imager ( Alpha Innotech Corporation , San Leonardo , CA ) . Clarified and pre-cleared two hundred ( 200 ) µg of cell lysates or one hundred and fifty ( 150 ) µg of Non-LR fractions were incubated for 2 h or overnight with immunoprecipitating antibody at 4°C , the resulting immune complexes were captured by Protein A or G-Sepharose and analyzed by Western blots using specific detection antibodies . HFF cells either mock infected or infected with KSHV and HEK 293 cells transfected with various plasmid combinations followed by virus infection were lysed in 2% SDS lysis buffer ( 2% SDS , 150 mM NaCl , 10 mM Tris-HCl , pH 7 . 5 , 2 mM EDTA , 10% glycerol , 1× protease inhibitor and 1× phosphatase inhibitor cocktail ) and boiled for 5 min followed by sonication . Lysates were diluted 1∶10 in dilution buffer ( 10 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 2 mM EDTA , 1% Triton X-100 ) , incubated at 4°C for 1 h with rotation and centrifuged at 20 , 000× g for 30 min . Equal amounts of protein were used for immunoprecipitation . Immunoprecipitated proteins were washed with washing buffer ( 10 mM Tris-HCl , pH 7 . 5 , 1M NaCl , 1 mM EDTA , 1% NP-40 ) , boiled in SDS sample buffer , and separated on SDS-PAGE followed by western immunoblotting for ubiquitin specific antibodies . Lipid raft extraction was performed as per the manufacturer's protocol for the caveolae/rafts isolation kit ( Sigma ) based on non-detergent density gradient extraction of lipid rafts [69] . Briefly , HFF cells were lysed in 0 . 5 M sodium bicarbonate solution in water ( 500 mM sodium carbonate pH 11 . 0 , 2 mM EDTA , 1 mM NaF , 1 mM orthovanadate , Sigma protease inhibitor cocktail ) . Cell lysates transferred into pre-cooled microfuge tubes were homogenized using a Dounce homogenizer ( 10 strokes ) and sonicated for 10 secs . A discontinuous density gradient made of 5 layers of OptiPrep with different concentrations was prepared as described previously [18] . Two ml of gradient layer ( 35% OptiPrep ) was placed at the bottom of the pre-cooled ultracentrifuge tube . Each OptiPrep layer was placed over the other using a Pasteur pipette . The tubes were ultracentrifuged at 45 , 000 rpm for 4 h using a Beckman SWI-55 rotor . One ml fractions were collected from the top of the centrifuge tube and pooled . Lipid raft containing fractions were characterized by the presence of caveolin-1 and non-lipid rafts were confirmed by the presence of CD-71 as described before [18] . HFF cells were infected with KSHV ( 30 DNA copies/cell ) at 37°C for 1 h . For measuring KSHV entry , cells were washed with HBSS and partially bound uninternalized virus was removed with 0 . 25% trypsin-EDTA for 5 min at 37°C . Internalized KSHV DNA was quantitated by amplification of the ORF73 gene by real-time DNA PCR [25] . The KSHV ORF73 gene cloned in the pGEM-T vector ( Promega ) was used for the external standard . The cycle threshold ( Ct ) values were used to generate the standard curve and to calculate the relative copy numbers of viral DNA in the samples . Percentage inhibition was calculated by considering the ORF73 copy numbers in control shRNA transduced cells as 100% . A paired t-test was used between control and sh-RNA treated cells to obtain p values . Total RNA was prepared from infected or uninfected cells using an RNeasy kit ( QIAGEN ) as described previously [25] . To quantitate viral gene expression , total RNA was subjected to ORF73 expression by real-time RT-PCR using gene specific primers and Taqman probes . The relative copy numbers of the transcripts were calculated from the standard curve using the Ct values of different dilutions of in vitro-transcribed transcripts . These values were normalized to GAPDH control reactions . HFF cells seeded on 8 well chamber slides ( Nalge Nunc International , Naperville , IL ) were used . Infected and uninfected cells were fixed with 4% paraformaldehyde for 15 min , permeabilized with 0 . 2% Triton X-100 , and blocked with Image-iTFX signal enhancer ( Invitrogen ) . The cells were reacted with primary antibodies against the specific proteins , followed by fluorescent dye-conjugated secondary antibodies . For colocalization with transferrin , cells were incubated with fluid-phase marker Alexa 594 transferrin ( 35 µg/ml ) at 37°C in the presence or absence of KSHV followed by immunostaining with the appropriate antibodies . Cells were imaged with a Nikon fluorescence microscope equipped with a Metamorph digital imaging system . Excitation and emission detection for each fluor was performed sequentially to avoid cross-talk . All experiments were performed at least three times . Cololalization of mean pixel intensities are analyzed for three different fields with a minimum of 10 cells each with the metamorph pixel intensity calculator . The mean colocalization pixel intensities ( for yellow color ) are measured in arbitrary unit ( a . u . ) that are represented as percentage colocalization with respect to uninfected control in a graph and a paired t test is used to obtain the p values . Wild Type ( pAM HA Cbl-Wt ) , RING mutant ( pAM-HAC3HC4C5 ) and TKB mutant ( pAM HA Cbl G306E ) constructs of c-Cbl were generously provided by Dr . Hamid Band [70] ( Eppley Institute for Cancer and Allied Diseases , University of Nebraska Medical Center ) . Wild Type EphA2 ( pMyc-EphA2 Wt ) , Kinase Dead mutant ( pMyc-EphA2 KD ) and SAM mutant ( pMyc-EphA2 SAM ) of EphA2 constructs were a kind gift from Dr . Horonori Katoh ( Laboratory of Molecular Neurolobiology , Graduate School of Biostudies , Kyoto University , Kyoto , Japan ) [45] . HEK293 cells were transiently cotransfected with each of the HA-tagged c-Cbl wild type or mutant plasmids and Myc-tagged EphA2 wild type or mutant plasmids . Transfection was performed using 5 µg of plasmid DNA , Lipofectamine 2000 ( Invitrogen ) and Opti-MEM ( Invitrogen ) according to the manufacturer's instructions . After 48 h , the cells were serum starved and either mock infected or infected with KSHV ( 30 DNA copies/cell ) for 5 min . Cell lysates were prepared for use in immunoprecipitation and immunoblotting studies . Transfection of primary HFF cells with siRNA was performed using the Neon transfection system ( Invitrogen ) according to the manufacturer's instructions . Briefly , subconfluent cells were harvested and washed once with 1×PBS and resuspended at a density of 1×107 cells/ml in resuspension buffer R ( provided by the company ) . 10 µl of this cell suspension was mixed with 100 pmol of siRNA and then microporated at room temperature using a single pulse of 1700 V for 20 ms . After microporation , cells were distributed into complete medium and placed at 37°C in a humidified 5% CO2 atmosphere . 72 hours post-transfection , cells were analyzed for knockdown efficiency by Western immunoblotting . All si-RNA oligonucleotides ( siGenome SMARTpool ) for c-Cbl and non-targeting siRNA pool no . 1 were purchased from Thermo Scientific ( Catalog no . M-003003-02-0010 and D-001206-13-20 , respectively ) . Data are expressed as means ± SD of at least three independent experiments ( n≥3 ) . In all tests , p<0 . 05 was considered statistically significant . Experiments in which p is <0 . 05 are marked with single asterisk and p<0 . 01 are marked with double asterisk . | KSHV is etiologically associated with Kaposi's sarcoma and primary effusion B-cell lymphoma . To initiate its in vitro infection of endothelial cells , KSHV interacts with cell surface heparan sulfate , integrins , and EphrinA2 ( EphA2 ) molecules in the lipid raft ( LR ) regions , which induces the integrin-c-Cbl associated signaling and macropinocytic entry . In contrast , KSHV enters human foreskin fibroblast ( HFF ) cells via LR-independent clathrin mediated endocytosis . The present studies conducted to define the key molecules regulating KSHV entry in HFF cells demonstrate that KSHV induces the association of integrins ( αVβ5 , αVβ3 and α3β1 ) with EphA2 in the non-LR regions of HFF cells and activates EphA2 , which in turn associates with c-Cbl , myosin IIA , FAK , Src , PI3-K , clathrin , AP2 and Epsin15 . Loss of EphA2 function reduces the induction of these signals , virus entry and infection . c-Cbl knockdown also abolishes the EphA2 polyubiquitination and clathrin association with EphA2 and KSHV . These results reveal for the first time the role of EphA2 in clathrin mediated endocytosis of a virus and c-Cbl directed polyubiquitination of EphA2 regulating KSHV infection by coordinating signal induction and underscores EphA2 and c-Cbl as potential targets to intervene in KSHV entry and infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"virology",
"biology",
"microbiology",
"viral",
"diseases",
"viral",
"replication"
] | 2013 | EphrinA2 Regulates Clathrin Mediated KSHV Endocytosis in Fibroblast Cells by Coordinating Integrin-Associated Signaling and c-Cbl Directed Polyubiquitination |
Paramyxovirus hemagglutinin-neuraminidase ( HN ) plays roles in viral entry and maturation , including binding to sialic acid receptors , activation of the F protein to drive membrane fusion , and enabling virion release during virus budding . HN can thereby directly influence virulence and in a subset of avirulent Newcastle disease virus ( NDV ) strains , such as NDV Ulster , HN must be proteolytically activated to remove a C-terminal extension not found in other NDV HN proteins . Ulster HN is 616 amino acids long and the 45 amino acid C-terminal extension present in its precursor ( HN0 ) form has to be cleaved to render HN biologically active . Here we show that Ulster HN contains an inter-subunit disulfide bond within the C-terminal extension at residue 596 , which regulates HN activities and neuraminidase ( NA ) domain dimerization . We determined the crystal structure of the dimerized NA domain containing the C-terminal extension , which extends along the outside of the sialidase β-propeller domain and inserts C-terminal residues into the NA domain active site . The C-terminal extension also engages a secondary sialic acid binding site present in NDV HN proteins , which is located at the NA domain dimer interface , that most likely blocks its attachment function . These results clarify how the Ulster HN C-terminal residues lead to an auto-inhibited state of HN , the requirement for proteolytic activation of HN0 and associated reduced virulence .
Newcastle disease virus ( NDV ) belongs to the large and diverse family of paramyxoviruses , which is responsible for many human and animal diseases [1] . The paramyxoviruses include other viruses such as mumps virus , measles virus , Sendai virus , respiratory syncytial virus ( RSV ) , metapneumovirus ( MPV ) , parainfluenza viruses ( PIV ) 1–5 , Nipah virus and Hendra virus . NDV infects birds and many different strains have been isolated worldwide that vary in pathogenicity and virulence . Highly virulent strains can cause a contagious disease with respiratory , neurological and digestive tract pathology , with the most severe infections leading to substantial economic losses in the poultry industry , despite aggressive vaccination programs . Highly virulent velogenic viscerotropic NDV strains , known as exotic NDV ( END ) strains , not endemic in the US , have caused severe disease outbreaks , such as the 1971 outbreak that required the killing of over 12 million chickens at a cost of $56 million with additional costs over a 4 year cleanup . A more recent END outbreak in 2002–2003 required the culling of over 3 million chickens in California at a cost of over $161 million . Continuing concerns about the severe economic impact of NDV outbreaks has led to the classification of NDV strains with an intracerebral pathogenicity index of >0 . 7 , or containing a fusion protein with a multi-basic cleavage site , as a U . S . Department of Agriculture Select Agent [2] , [3] . Recently it has been shown that NDV is able to selectively kill tumor cells , suggesting it could be useful as an oncolytic agent , and NDV is also being investigated as a potential vaccine vector ( reviewed in [4] ) . Paramyxoviruses are enveloped , negative-sense , single-stranded RNA viruses . Virions consist of a nucleocapsid , a matrix protein , and an envelope formed by a lipid membrane , typically with two glycoproteins displayed on the surface [1] . For virus penetration into target cells , the lipid envelope must fuse with a cell membrane . Membrane fusion , for nearly all paramyxoviruses , is triggered at the cell surface in a receptor-dependent , pH-independent manner , unlike the pH-dependent influenza virus hemagglutinin mechanism [1] , [5] , [6] . For most members of the virus family , two viral glycoproteins are required to mediate this entry process – the fusion ( F ) glycoprotein and an attachment protein referred to as hemagglutinin-neuraminidase ( HN ) , hemagglutinin ( H ) , or glycoprotein ( G ) , depending on the virus [1] , [5] , [6] . Activation of the F protein requires virus-specific ( homotypic ) interactions with the attachment glycoprotein for viral entry , except for RSV and MPV [7] , [8] . The HN attachment proteins are found in a subset of the paramyxoviruses , including NDV , mumps virus , parainfluenza virus 5 ( PIV5 ) , Sendai virus , and human parainfluenza viruses 1–4 ( hPIV1–4 ) [1] . HN protein binds to the receptor ( sialic acid ) for virus attachment to cells and plays additional roles in the virus life cycle including F activation , and receptor-destroying ( neuraminidase ) activity to facilitate virus budding . HNs are type II membrane proteins , with N-terminal transmembrane domains followed by a stalk region and a large C-terminal globular head domain . The active form is thought to be a tetramer [1] . The four-helix bundle ( 4HB ) stalk structures of the NDV Australia-Victoria ( AV ) HN [9] and the PIV5 HN [10] have been determined and revealed a surface site in the HN stalk that is thought to interact directly with F . Mutations at this HN stalk site affect both F protein binding and fusion activation ( reviewed in [11] ) . The severity of NDV-associated disease depends on the virus strain and the host species . NDV strains are grouped into three main pathotypes of high ( velogenic ) , intermediate ( mesogenic ) , or low ( lentogenic ) virulence . Both F and HN proteins are important determinants of the virus virulence [12] , [13] , [14] , [15] , [16] , [17] , [18] . In general , the HN proteins from different paramyxoviruses exhibit high sequence similarity , particularly in the core neuraminidase ( NA ) domain . The sequence of HN of NDV is overall highly conserved among isolates but one striking example of a difference occurs in lentogenic ( low virulence ) strains , such as the Ulster , D26 and Queensland strains that are 616 amino acids in length [19] , [20] , [21] . This is longer than most other NDV HN proteins ( 571 , 577 amino acids ) , due to a C-terminal extension of 45 amino acids . Proteolytic cleavage of the C-terminal extension in HN0 removes 42 residues [19] , leading to increases in neuraminidase and hemadsorption activities that are necessary for the viral life cycle [14] , [15] . To understand better how the Ulster HN C-terminal extension down-regulates HN activities and may reduce virus pathogenicity , we determined the crystal structure of the NDV Ulster HN “head” or “NA” domain precursor . The structure reveals that the 45 amino acid extension crosses the outside of the NA domain and inserts C-terminal residues into the active site , interacting with residues involved in sialic acid receptor binding . The C-terminal extension also occludes a secondary sialic acid binding site implicated in virus attachment , effectively blocking both catalytic and binding functions of NDV HN . The C-termini are stabilized by an interchain disulfide bond formed by C596 and mutagenesis of this residue releases the inhibition of neuraminidase and hemadsorption activities . The results clarify the auto-inhibitory functions of the C-terminal extension present in Ulster HN and many other low virulence NDV strains .
In the NDV Ulster strain , both HN and F proteins are initially produced as precusor proteins , F0 and HN0 , which must be proteolytically cleaved to be fully active and support viral replication [14] , [15] . The Ulster HN C-terminal extension lengthens the protein to 616 residues and contains many charged as well as aromatic amino acids ( Figure 1A ) . To confirm prior observations obtained with virally expressed proteins , recombinant Ulster F and HN were expressed from cDNA by transfection of HeLa-CD4-LTR-β-gal ( HeLa ) cells with plasmids expressing the F and HN genes . The F and HN proteins were immunoprecipitated from [35S]-labeled lysates using rabbit polyclonal antibodies specific for F or HN and the precipitated proteins were analyzed by SDS-PAGE . The recombinant F and HN proteins were not cleaved efficiently but were detected as the precursors F0 and HN0 ( Figure 1B ) . Both precursors could be cleaved extracellularly by the addition of exogenous trypsin , to produce F1 , F2 and HN , as shown previously for the F and HN proteins synthesized in virus-infected cells [14] , [15] . The receptor binding activity of the recombinant HN protein was examined using a hemadsorption assay as described in Materials and Methods . Without trypsin treatment , HeLa cells expressing Ulster HN protein exhibited a low level of chicken red blood cell ( RBC ) binding ( Figure 1C , bottom left ) . Trypsin treatment of the HN expressing HeLa cells , prior to the addition of RBCs , greatly increased RBC binding ( Figure 1C , bottom right ) . The results of a quantitative hemadsorption assay are shown in Table 1 , further confirming these observations . The low level of binding activity detected in the absence of trypsin treatment is presumed to be due to the small amount of HN that is cleaved intracellularly as observed previously for virus-infected cells [14] , [15] . The ectodomain ( residues 49–616 ) and NA domain ( residues 124–616 ) of Ulster HN were expressed as soluble , secreted proteins in monolayers of 293T cells and purified as described in Materials and Methods . The proteins were >90% pure after elution from a Ni-NTA agarose column ( Figure 2A ) . The construct of the NDV Ulster HN NA domain begins at residue 124 , 1 residue after a conserved cysteine ( C123 ) that forms an interchain disulfide bond . Previous studies have demonstrated that secreted paramyxovirus HN proteins form dimers and tetramers , but that isolated NA domains are primarily monomeric in solution [22] , [23] , [24] , [25] , [26] , [27] . However , both the full length Ulster HN ectodomain , containing C123 , and the shorter NA domain , lacking C123 , form disulfide-linked dimers . Both the ectodomain and NA domain proteins contain disulfide bonds that can be reduced by dithiothreitol treatment ( Figure 2A ) . The purified NA domain protein eluted with an apparent molecular weight of 107 kDa in gel filtration chromatography ( Figure 2B ) , consistent with covalent dimer formation . The C-terminal extension contains an additional cysteine ( C596 ) not present in HN proteins containing 571 or 577 amino acids . The additional cysteine residue could account for the NA domain dimer formation ( Figure 1A ) . To investigate the role ofC596 in forming a disulfide bond , we generated a mutant in which C596 was replaced by serine ( C596S ) . C596S migrates as a monomer when analyzed by SDS PAGE under non-reducing conditions ( Figure 3A ) . Further evidence that C596 forms an interchain disulfide bond was provided by sucrose gradient sedimentation and also electron microscopy of WT HN and C596S NA domains . Fractionation of sucrose gradients showed that the WT HN protein was found mainly in fractions 7–8 , whereas C596S was mostly detected in fractions 9–11 ( Figure 3B ) , consistent with their migration as dimers and monomers , respectively , on SDS-PAGE . Comparison of the sedimentation of HN and HN C596S with tetrameric , dimeric and monomeric forms of a soluble version of influenza virus neuraminidase ( NAF ) [28] further indicated that the WT HN NA domain sedimented as a dimer and the HN C596S NA domain sedimented as a monomer . In electron micrographs of negatively stained protein , the WT NA domain is observed mostly as dimers ( Figure 3C , left ) whereas the HN C596S mutant is observed mostly as monomers ( Figure 3C , right ) , in agreement with the biochemical data . We tested the neuraminidase activity of WT HN and HN C596S NA domains as described in Materials and Methods . Mutation of C596S results in a >3-fold increase in neuraminidase activity ( Table 2 ) . The C596S mutant was incorporated into the full length Ulster HN for receptor binding studies . As shown in Table 1 , the mutation of C596S increased hemadsorption activity to levels comparable to trypsinized WT HN , with C596S binding activity being independent of trypsin . The interchain disulfide bond formed by C596 is therefore critical for maintaining the auto-inhibited state of the Ulster HN . The purified NA domain protein crystallized in space group P3121 and diffracted X-rays to 3 . 5 Å resolution ( Table 3 ) . The structure was solved by molecular replacement , using models of the NDV Kansas NA domain [25] , [26] , [29] . Four copies of the NA domains are located in the crystallographic asymmetric unit , arranged as an apparent dimer-of-dimers , reminiscent of other HN structures ( Figures 4A , 5 ) . The NA domain dimers are highly similar to those observed in structures of NDV Kansas HN [26] , [29] , hPIV3 HN [30] , and PIV5 HN [22] . The apparent tetramer arrangement of the dimer-of-dimers differs from the NDV AV HN structure [9] , which also contains its stalk region , but is similar to that observed for PIV5 HN [22] . Although two dimers pack into this tetramer arrangement , the Ulster HN ectodomain , like AV HN [9] , forms dimers in solution , in contrast to the stable tetramers observed for PIV5 HN [22] , [23] . Electron density maps clearly show the presence of a majority of the C-terminus with a short main chain break at residue 571 ( Figure 4A , B ) . The C-terminal extension wraps around the external surfaces of the HN dimer-of-dimers . The cleavage site for the activation of the precursor HN0 is located within the break in the electron density at residues 571–578 ( Figures 1A , 4B ) , consistent with this region being flexible and accessible to proteolytic digestion [12] , [14] , [15] , [19] . The extra C-terminal residues adopt a mostly extended conformation beginning near the bottom of the β-propeller structure , traversing the NA dimer interface , rising above the NA domain and inserting C-terminal residues in the HN catalytic site ( Figures 4 , 5A ) . A short helix ( C-helix ) is formed by residues 604–611 ( Figure 4B ) . The four C-terminal extensions adopt very similar conformations , with RMSD values after superposition ranging from 0 . 27–0 . 65 Å . Two C-terminal extensions from adjacent subunits meet at C596 , approaching from opposite directions . Electron density for the C596 interchain disulfide bond was observed , consistent with the biochemical and mutational data indicating the presence of the disulfide linkage . The Ulster HN C-terminal extension inactivates the NA domain catalytic site through three potential mechanisms ( Figures 5A , 6A ) . First , the extreme C-terminal residues occupy the active site , forming specific interactions with residues involved in sialic acid binding . Second , an active site variable loop ( the D198 loop ) shifts into an inactive conformation , further inhibiting NA binding and catalytic activities . Third , residues in the C-terminal extension positioned outside and above the active site ( the 590–611 ‘turret’ ) provide additional steric hindrance to the binding of larger oligosaccharide structures . The interactions of the C-terminal residues that block the NA active site bury ∼350 Å2 of surface area and involve both polar and non-polar contacts ( Figure 6B ) . The side chain of W615 is buried deepest within the active site , forming a hydrogen bond with Y526 and hydrophobic contacts with multiple residues , including R498 , R416 , D198 , I175 , and R174 . S614 forms a hydrophobic contact with R498 , while the main chain of A613 forms a hydrogen bond with R498 . A612 , which is positioned at the edge of the active site , forms a hydrophobic contact with A497 . Y611 , at the end of the C-helix , anchors the entry of the remaining residues into the active site . Many of these active site residues , R174 , R498 , R416 , and E258 , are directly involved in the interaction with sialic acid or a sialic acid inhibitor ( 2-deoxy-2 , 3-didehydro-N-acetylneuraminic acid or DANA ) , as revealed in the structures of NDV Kansas [29] ( Figure 6C ) and PIV5 HN [22] . Therefore , the insertion of the C-terminal residues into the active site directly disrupts receptor binding and NA activity . Cleavage of HN0 removes an 8kD glycopeptide corresponding to the C-terminal extension , which is not found in the mature protein [12] , [14] , [15] , [19] . We have shown that mutation of the C596 disulfide bond also releases the auto-inhibited state . Both observations indicate that the interactions of residues 612–615 on their own are likely weak and insufficient to mediate inhibition in the absence of the remainder of the supporting C-terminal extension structure . Although the NDV Ulster HN structure is overall very similar to the NDV AV [9] and Kansas HN structures [26] , [29] , one variable active site loop , the D198 loop comprising R197-S200 , shows significant structural differences ( Figure 6A ) . D198 , which is equivalent to D151 in influenza virus NA , is thought to be directly involved in the hydrolysis of the glycosidic bond via a water molecule . In the Ulster HN structure , the D198 loop switches conformation , pointing deeper into the active site , and forms a hydrophobic contact with W615 ( Figure 6B ) . In this conformation , the side chain of D198 would clash with the sialic acid substrate . In a low pH , ligand-free structure of NDV Kansas HN , the D198 loop also points into the active site [26] , but not as deeply as observed in the Ulster HN structure ( Figure 6A ) . The Ulster HN crystals were grown at pH 7 . 5 , therefore the conformation of the D198 loop is not linked to an acidic pH condition and is more likely a result of the insertion of the C-terminus , and W615 , into the active site . Together with the C-terminal residues ( A612-P616: electron density for P616 was weak ) , the conformational change in the D198 loop would occupy the active site , fully blocking receptor engagement and catalytic activity . The I590-Y611 region of the C-terminal extension forms a projection , or ‘turret’ , to one side of the NA domain and above the active site ( Figure 6A ) . Residues 590–611 could provide additional steric interference that would limit the ability of HN to bind larger , branched oligosaccharide structures present at the cell surface . The disulfide bond formed by C596 is critical for the Ulster HN auto-inhibition and is located at the center of a ‘turret’ supported by an extensive hydrophobic interface between C-terminal extensions of adjacent HN subunits ( Figures 5B , 7A ) . The C-terminal extension forms symmetric intersubunit interactions through an S-curve structure as it rises above the β-propeller domain through residues 590–599 . Residue 596 defines the center of the top of this turret structure , and the polypeptide chain then descends into the NA domain active site , forming the C-helix through residues 604–611 . Residues 590–611 from the adjacent subunits form a set of interactions that support the turret and C-terminal helix , guiding the polypeptide chain into the active site ( Figure 7A ) . Residues 590–595 from one subunit insert underneath the α-helix from the second subunit , interacting with its hydrophobic face . Aromatic residues F595 , Y604 and Y611 are central to this inter-subunit hydrophobic core ( Figure 7A , B ) . I594 and F595 from one subunit pack against Y604 and Y611 from the second subunit . Additional hydrophobic core residues include I590 and L608 , while S592 appears to form a hydrogen bond with Y604 . Y611 packs at the lip of the active site , and is likely critical for stabilizing the position of the C-terminus and inhibition of receptor binding ( Figure 7A , B ) . The C596 disulfide bond is centrally located in the crossover junction between the subunits , consistent with its critical role in auto-inhibition ( Figure 7B ) . Mutation of C596 to serine likely destabilizes this structure , leading to displacement of the C-terminus from the active site and restoration of neuraminidase and hemadsorption activities . A second sialic acid binding site at the NA domain dimer interface ( Figures 5C , 8A ) was revealed in the X-ray structure of NDV Kansas HN in complex with thiosialoside [24] , [29] . This second site is not conserved across the paramyxovirus HN proteins , as shown by its absence in the atomic structure of PIV5 HN [22] . For hPIV3 an N-linked glycan masks a second receptor-binding site [30] , [31] . However , for NDV , functional studies indicate that this second site does play a role in NDV attachment and entry [24] , [29] , [32] , [33] . In the Ulster HN structure , we observe that residues E584-K587 in the C-terminal extension directly engage the second sialic acid binding site , indicating that this would be sterically blocked in HN0 ( Figures 5C , 8A ) . Within the second site , interactions with sialic acid come from polar contacts with mainchain atoms , and through a conserved hydrophobic pocket , including residues G169 , L552 , F553 from one subunit , and F156 , V517 , and L561 from the second subunit ( Figure 8B ) . In the thialoside complex , sialic acid forms hydrogen bonds with S519 , R516 , F156 and G169 . In the Ulster HN structure , the mainchain oxygen of K587 forms a hydrogen bond with S519 and W586 interacts with R516 through hydrophobic contacts , which would block thialoside interactions ( Figure 8C ) . In addition , the side chain of K587 occupies the hydrophobic pocket , whose residues are conserved in Ulster and Kansas HN proteins . The second receptor binding site would thereby be completely blocked by the HN0 C-terminus .
In this report , we demonstrate that the longer HN precursor , HN0 , found exclusively in avirulent NDV strains such as NDV Ulster [19] , [20] , [21] , is folded to an auto-inhibited state . The C-terminal extensions of the Ulster HN block the neuraminidase active sites and second sialic acid binding sites , both of which engage receptors on target cells , explaining the reduced hemadsorption and catalytic activities of the HN0 precursor . We further demonstrate that this auto-inhibition is dependent upon a disulfide bridge formed by C596 in the extension and that C596 is centrally located at an extensive interdomain interaction between the C-termini , which rises like a turret above the HN NA domain dimer . Both proteolysis and mutation of C596 lead to HN activation , indicating that disruption of the C-terminal extension structure releases HN from the auto-inhibited state . Other NDV strains , such as D26 and Queensland [19] , [20] , [21] , also express a 616 amino acid long HN precursor protein . Sequence features important to the structure of the extension observed here in Ulster HN are also conserved in these related HN proteins , indicating that they fold similarly . Of 45 amino acids in the C-terminal extension structure , 12 appear absolutely conserved , including C-terminal residues that engage the active site ( S614 , W615 ) , residues that interact with the second site ( G585 , W586 and D588 ) and residues that form the turret structure and C-helix ( V591 , P593 , F595 , N600 , T602 , R605 , E609 ) . Other key residues are highly conserved , such as Y611 , although C596 is not absolutely conserved and is replaced by arginine or serine in a subset ( <10% ) of available database sequences . Compensatory changes may stabilize the HN turret structure in the absence of C596 , as this appears critically important for placing inhibitory C-terminal residues into the HN active site and N-terminal residues in the second sialic acid binding site . The observation that the C-terminal extension engages both of these receptor-binding sites further indicates that both sites are functionally important in vivo , as other studies have suggested [24] , [29] , [32] , [33] . Although the longer HN0 is only found in avirulent NDV strains , one study with recombinant engineered virus suggested that expression of HN0 does not reduce virulence , as followed by intracerebral inoculation of 1 day old chickens [18] . However this study did not address the natural route of infection or spread to various internal tissues . NDV virulence has been repeatedly linked to the efficiency of F protein cleavage , which is associated with the presence of multiple basic ( Arg/Lys ) residues at its cleavage/activation site [16] , [18] , [34] , [35] . Despite these observations , a subset of naturally occurring NDV strains have evolved to express the precursor HN0 and HN0 clearly has inhibited receptor binding and neuraminidase activities , which would be expected to influence infectivity , viral spread and virulence [14] , [15] , [19] , [36] . Precursor HN0 proteins likely confer some overall selective advantage in a natural infection setting , perhaps reducing virulence or pathogenicity in the avian host in a manner that may beneficially promote virus survival in a population . The evolution of an auto-inhibited attachment protein , requiring proteolytic activation , appears to be unique to NDV within the larger paramyxovirus family , although other paramyxovirus attachment glycoproteins exhibit length variations which might have functional consequences . Given the dramatic range of virulence of NDV strains , which encompass symptoms from mild respiratory infections to 100% mortality , there may have been strong selective pressure to moderate NDV pathogenicity , resulting in this unique mechanism for the inhibition of HN functions that are essential to the virus life cycle .
293T cells were maintained in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) . HeLa CD4-LTR-βgal cells were maintained in DMEM supplemented with 10% FBS , 200 µg/ml G418 ( InvivoGen , San Diego , CA ) , and 100 µg/ml hygromycin B ( InvivoGen ) . The growth of NDV Ulster and whole cell RNA eztraction was performed by Drs . Daniel J . King and Claudio Afonso at the USDA Poultry Rserach Laboratory , Athens , Georgia . DF-1 cells were infected with NDV Ulster at a multiplicity of infection of 5 plaque forming untits/cell and total cell RNA extracted . The cDNAs encoding NDV Ulster F and HN were synthesized by reverse transcription-PCR using the whole cell RNA . The cDNAs were cloned into the eukaryotic expression vector pCAGGS [37] and their nucleotide sequence obtained using an Applied Biosystems 3100-Avant automated DNA sequencer ( Life Technologies Corp . , Carlsbad , CA ) . DNA encoding the Ulster HN ectodomain ( residues 49–616 ) and head ( NA ) domain ( residues 124–616 ) were cloned into the vector pHLsec [38] . HN sequences were preceded by an N-terminal 6 His tag and a thrombin cleavage site . Mutation of Ulster HN cysteine 596 to serine was carried out by 4 primer PCR . HeLa CD4-LTR-βgal cells were transfected with pCAGGS Ulster HN or F using Lipofectamine Plus ( Life Technologies Corp . ) according to the manufacturer's protocol . 24 h post-transfection ( p . t . ) the cells were washed with DMEM without methionine and cysteine ( DMEM met− cys− ) , incubated in DMEM met −cys− for 30 min and labeled for 30 min with 100 µCi/ml EXPRE35S35S Protein Labeling Mix ( PerkinElmer , Waltham , MA ) . Following the labeling period the cells were incubated for 2 h in DMEM supplemented with 2 mM methionine , 2 mM cysteine . For half the samples the medium was removed after 90 min and replaced with medium containing 5 µg/ml TPCK-treated trypsin ( Worthington Biochemical Corp , Lakewood , NJ ) . Cell lysates were prepared and immunoprecipitation was carried out as described previously [39] using a rabbit polyclonal antibody ( R9722 ) raised against NDV Australia Victoria HN ectodomain and a rabbit polyclonal antibody raised against residues 409–423 of Australia-Victoria F protein ( R1680 ) . The samples were analyzed by SDS-PAGE and the radioactivity visualized using a Fuji FLA-5100 imaging system ( Fuji Medical Systems , Stamford , CT ) . The NDV Ulster “head” and full ectodomain constructs were expressed by transient expression in 293T cells . Briefly , 293T cells in 10 cm dishes were transfected using 23 µg pHLsec HN head or pHLsec ectodomain and 34 µg 25 kDa branched polyethyleneimine ( Sigma-Aldrich; St . Louis , MO ) per plate , the supernatant media was collected at 3 days p . t . , clarified by centrifugation ( 650×g ) in an Allegra 6 centrifuge ( Beckman Coulter ) and filtered through a 0 . 2 µm filter . The pH was adjusted to pH 8 . 0 and the proteins were purified by affinity chromatography using nickel-nitrilotriacetic acid ( Ni-NTA ) agarose ( Qiagen; Valencia , CA ) . The Ni-NTA columns were washed with buffer containing 10 mM and 50 mM imidazole and the proteins were eluted using buffer containing 250 mM imidazole . The proteins were further purified by size exclusion chromatography , using a Superdex S200 column , with a running buffer of 50 mM Tris , pH 8 . 0 , 200 mM NaCl . The protein buffer was subsequently exchanged to 25 mM Tris , pH 7 . 5 , 50 mM NaCl . The eluted proteins were >90% pure by SDS-PAGE analysis and Coomassie brilliant blue or silver staining . Ulster HN head NA domain or HN C596S NA mutant protein were layered on top of a 5–25% ( W/V ) sucrose gradient formed over a 60% sucrose cushion . The gradients were centrifuged in a SW41 swinging bucket rotor at 38 , 000 rpm for 16 . 75 h at 20°C using an Optima L80 XP ultracentrifuge ( Beckman Coulter ) . Sixteen 0 . 75 ml fractions were collected and analyzed by SDS-PAGE under non-reducing conditions . HeLa CD4-LTR-βgal cells were transfected with pCAGGS Ulster HN or empty vector using Lipofectamine Plus ( Life Technologies Corp . ) . 18–20 h p . t . the media was removed , the cells were washed twice in phosphate buffered saline containing calcium and magnesium ( PBS+ ) and incubated at 37°C for 20 min in PBS+ or PBS+ containing 5 µg/ml TPCK-treated trypsin ( Worthington Biochemical Corp ) . The cells were then placed on ice , washed several times with PBS+ and incubated for 2 h at 4°C with 1 ml of a 1% suspension of chicken red blood cells ( RBCs ) in PBS+ . The cells were thoroughly washed with PBS+ and photographed using an inverted phase-contrast microscope ( Diaphot; Nikon , Melville , NY ) connected to a digital camera ( DCS 760; Kodak , Rochester , NY ) . For quantification of hemadsorption , after washing to remove unbound RBC's , 0 . 5 ml of water was added and the cells were incubated for 2 h at 4°C . The absorption of the clarified supernatant was read at 540 nm using a Beckman Coulter DU 730 Life Sciences UV/Vis spectrophotometer ( Beckman Coulter , Brea , CA ) . 10 µl 12 . 5% dimethyl sulfoxide was added to the wells of a 96 well plate followed by 10 µl HN protein diluted in enzyme buffer ( 162 . 5 mM 2- ( n-morpholino ) ethanesulfonic acid ( MES ) pH 6 . 5 , 5 mM CaCl2 ) . 30 µl substrate ( 54 . 2 mM MES , 5 mM CaCl2 , 0 . 167 mM 4-methylumbelliferyl-N-acetyl-α-D-neuraminic acid ) was added and the plate was incubated at 37°C for 15 min with shaking . 150 µl of stop solution ( 0 . 014 M NaOH in 83% ethanol ) was added and the fluorescence of the cleaved substrate was measured at excitation and emission wavelengths of 360 and 440 nm respectively using a Spectramax M5 plate reader ( Molecular Devices , Sunnyvale , CA ) . Solutions of NDV HN ( Ulster ) at a concentration of approximately 5 mg/ml were absorbed onto 300 mesh copper grids covered with a carbon film that had been freshly glow discharged . Grids were stained with a 1% aqueous solution of uranyl formate , freshly prepared and filtered immediately prior to use . Grids were observed in a JEOL 1230 electron microscope operated at 100 kV and images were acquired with a Gatan 831 CCD camera , at the Biological Imaging Facility , Northwestern University , Evanston , IL . The NDV HN ( Ulster ) neuraminidase “head” domain protein was concentrated to ∼8 mg/ml in 25 mM Tris , pH 7 . 5 , 50 mM NaCl . The crystallization condition was identified by screening with the Wizard crystallization kits ( Emerald BioSystems , WA ) set up with a Phoenix robot . For further optimization of the initial hit condition , the protein was crystallized at room temperature by the hanging drop vapor diffusion method . The drops contain an equal ratio of protein and precipitant consisting of 17% PEG 8000 , 200 mM MgCl2 , 100 mM Tris , pH 8 . 5 . Crystals appeared with a rod shape with dimension of ∼80×400 µm . The harvesting and freezing buffer contained 20% PEG 8000 , 200 mM MgCl2 , 100 mM Tris , pH 8 . 5 , and 15% glycerol . The diffraction dataset was collected at the Argonne Advanced Photon Source NE-CAT , station 24IDE and processed to 3 . 5 Å using HKL2000 [40] . The NDV HN Kansas [26] monomer structure ( PDB ID: 1E8T ) was used as the search model to determine the initial phases . Model building and structure refinement were performed with Coot [41] , Refmac [42] in the CCP4 package [43] , and Phenix [44] . The model after molecular replacement was first refined by rigid body refinement with Refmac [42] , followed by two rounds of restrained refinement without and with NCS restraints , where residues 124–569 in Chains A , B , C and D were restrained by NCS . The C-terminus was built both by Buccaneer [43] and manual modeling . Further refinement was carried out using Phenix [44] , with NCS restraints at the beginning and TLS refinement throughout to the final model . TLS groups were generated automatically in Phenix , with each chain ( A , B , C , and D ) divided into groups of residues as follows: 124∶192 , 193∶229 , 230∶300 , 301∶459 , 460∶555 , 556∶557 , and 578∶615 . The data collection and final refinement statistics are collected in Table 3 . Coordinates and structure factors have been deposited in the RCSB under PDB ID code 4FZH . Protein alignments were performed using MultAlin [45] . | Newcastle disease virus ( NDV ) can cause severe disease in birds , with the most virulent strains causing sudden death , even in vaccinated populations in the poultry industry . Highly virulent exotic NDV ( END ) strains have caused largescale outbreaks in the US in 1971 and 2003 , requiring the culling of 12 million and 3 million chickens , respectively . Additional economic costs were associated with containment and cleanup . NDV strains vary greatly in their virulence and ability to cause such outbreaks . Two proteins at the surface of the virus , the hemagglutinin-neuraminidase ( HN ) and the fusion ( F ) protein , activate NDV entry into cells and variations in both proteins are linked to differences in strain-specific virulence . Certain avirulent strains of NDV , such as NDV Ulster , express a longer HN protein with a C-terminal segment that must be proteolytically cleaved to fully activate the protein . Here we demonstrate that the extra C-terminal 45 amino acids of NDV Ulster HN adopt a well-defined structure , not present in the shorter HN proteins from virulent strains , that blocks two key receptor binding regions necessary for attachment to cells and virus entry . The results clarify how this unique evolutionary adaptation suppresses HN functions in avirulent NDV strains , consistent with an important role for this region in modulating NDV pathogenicity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"viral",
"enzymes",
"microbiology",
"host-pathogen",
"interaction",
"viral",
"structure",
"protein",
"folding",
"virulence",
"factors",
"and",
"mechanisms",
"proteins",
"biology",
"pathogenesis",
"biophysics",
"biochemistry",
"protein",
"chemistry",
"viral",
"envelope",
"virology"
] | 2012 | Structure of the Ulster Strain Newcastle Disease Virus Hemagglutinin-Neuraminidase Reveals Auto-Inhibitory Interactions Associated with Low Virulence |
Through their domestication and subsequent selection , sheep have been adapted to thrive in a diverse range of environments . To characterise the genetic consequence of both domestication and selection , we genotyped 49 , 034 SNP in 2 , 819 animals from a diverse collection of 74 sheep breeds . We find the majority of sheep populations contain high SNP diversity and have retained an effective population size much higher than most cattle or dog breeds , suggesting domestication occurred from a broad genetic base . Extensive haplotype sharing and generally low divergence time between breeds reveal frequent genetic exchange has occurred during the development of modern breeds . A scan of the genome for selection signals revealed 31 regions containing genes for coat pigmentation , skeletal morphology , body size , growth , and reproduction . We demonstrate the strongest selection signal has occurred in response to breeding for the absence of horns . The high density map of genetic variability provides an in-depth view of the genetic history for this important livestock species .
Man's earliest agricultural systems were based on the captive management of sheep and goats . The transition from hunting to animal husbandry involved human control over the reproduction , diet , and protection of animals . The process of domestication was initiated approximately 11 , 000 years ago in the Fertile Crescent [1] . The impact was a profound redirection of human society , as domesticated livestock and plants increased the stability of human subsistence and fuelled population growth and expansion . Domestication also reshaped the morphology , behaviour , and genetics of the animals involved , with the first consequences likely to have included changes to coat pigmentation and horn morphology . Sheep were first reared for access to meat before human mediated specialisation for wool and milk commenced ca 4 , 000–5 , 000 years ago [2] . Phenotypic radiation under selection is ongoing , resulting in a spectrum of modern breeds adapted to a diverse range of environments and exhibiting the specialised production of meat , milk , and fine wool . The last few hundred years has seen the pace of genetic gain increase dramatically through the division of animals into breeds , the implementation of quantitative genetics methodology , and the use of artificial insemination to prioritise genetically superior rams . Patterns of genetic variation have long proven insightful for the study of domestication , breed formation , population structure , and the consequences of selection . Variation within the mitochondrial genome has documented the global dispersal of two major haplogroups in modern sheep [3] , [4] . Analysis of endogenous retroviruses suggests the development of breeds has occurred in multiple waves , where primitive breeds have been displaced by populations which display improved production traits [2] . Investigations into the genetic relationship between populations have primarily relied on a modest collections of autosomal microsatellites [5]–[7] , Y chromosomal markers [8] , or SNP [9] . To date , the majority of populations tested have been European-derived breeds . This prompted assembly of the global sheep diversity panel , which contains animals from 74 diverse breeds sampled from Asia , Africa , South-West Asia ( the Middle East ) , the Caribbean , North and South America , Europe , and Australasia . Our goal in assembling this animal resource was 2-fold . Firstly , we sought to examine levels and gradients of genetic diversity linking global sheep populations to better understand the genetic composition and history of sheep . We therefore genotyped all of the animals in the global diversity panel using the ovine SNP50 Beadchip , an array consisting of approximately 50 , 000 evenly spaced SNP . We present the relationship between breeds in terms of divergence time , estimated from the extent of haplotype sharing . Secondly , we sought to characterise the genetic legacy that selection and adaptation have imparted on the sheep genome . By performing a genome-wide scan for the signatures of selection , 31 genomic regions were identified that contain genes for coat pigmentation , skeletal morphology , body size , growth , and reproduction . By combining the collection of a global sample of ovine breeds with the ability to interrogate 50 , 000 genetic loci , the results provide unprecedented insight into the phylogeographic structure of sheep populations and the results of centuries of breeding practices .
Analysis of genetic variation was performed for 2 , 819 animals in the global sheep diversity panel . Breeds were sampled from each continent across the species range ( Figure 1 ) , including six breeds from both Africa and America , seven from South-West Asia ( the Middle East ) , eight from Asia , and the rest from northern , north-western , central , and southern and south-western Europe ( Table S1 lists the breed and their geographic origin ) . All animals were genotyped using the ovine SNP50 Beadchip , an array consisting of SNP derived from three separate sequencing experiments ( Roche 454 , Illumina GA and Sanger sequencing; Table S2 ) . A series of quality control filters were applied to identify 49 , 034 SNP used in subsequent analysis ( Table S3 ) . Levels of SNP polymorphism were generally high , with greater than 90% of loci displaying polymorphism within the majority of breeds ( Table S4 ) . The distribution of minor allele frequency ( MAF ) differed between population groups chosen to reflect the geographic origin of breed development . African and Asian breeds had an excess of low MAF SNP ( <0 . 1 ) compared to European-derived populations . This partly reflects ascertainment bias in SNP discovery , as the same analysis conducted using SNP discovered without use of African or Asian sheep ( 454 SNP; Figure S1 , Table S2 ) shows a more pronounced excess compared with SNP discovered using a broad genetic base ( Illumina GA SNP ) . To examine diversity on a global scale , we calculated observed heterozygosity ( He ) within breeds and between regions ( Table S4 ) . Allele frequency-dependent diversity estimates such as He are sensitive to ascertainment bias , prompting the removal of SNP in high LD , which acts to counter the effect of the bias and generate meaningful comparisons between populations [10] . Applied here , breed rankings based on He were generally stable following LD-based pruning and when calculated using SNP sets ascertained using different methods ( Figure S2 ) . Following LD-based correction , animals from Southern and Mediterranean Europe displayed the highest heterozygosity ( Figure S2 ) . This likely reflects the first migrations of Neolithic communities and their animals , following the Mediterranean as a sea route into Europe [11]–[13] . Relative levels of genetic diversity are expected to decrease with increasing distance from the domestication centre . For sheep , breed heterozygosity revealed only a weak association with increasing physical distance ( Figure 1B , r = −0 . 40 ) . This appears much less pronounced in sheep compared with human migration out of Africa [14] . One likely explanation is the widespread use of Merino sires across Europe that commenced after the Middle Ages . The result is extensive haplotype sharing between Merinos and other breeds ( Figure 1C ) . Generally high SNP diversity in sheep was accompanied by many breeds displaying high current effective population size ( Ne , Table S4 ) . Compared with domestic cattle where the majority of breeds have a current Ne of 150 or less [15] , estimates here revealed 25 breeds have Ne exceeding 500 and only two sheep populations showed evidence of a comparatively narrow genetic base ( Ne<150 ) . Global patterns of genetic structure were inferred by principal components analysis ( PCA , Figure 2 ) . The analysis ignores breed membership but revealed clear structure as animals from the same breed clustered together . As demonstrated in human and other livestock species such as cattle [15]–[17] , the combination of PC1 , PC2 , and PC3 separated individuals according to their geographic origin . The largest PC ( 2 . 98% of total variation ) positioned European sheep apart from African , Asian , and South-West Asian animals . The second PC ( 1 . 44% ) separated European-derived animals from those developed in Africa and Asia animals . PC3 ( 1 . 19% ) identified admixed populations such as the African Dorper and breeds developed in South America and the Caribbean were positioned away from other clusters . It also resolved two primitive and geographically isolated Scottish breeds ( Soay and Boreray ) as outliers from all other animals [2] , [6] , [9] . PC4 ( 1 . 09% ) separated British Dorset types ( DSH , APD , and ASU ) from other European derived breeds and PC7 identified the Valais breeds as genetically distinct . Additional PCs reveal the divergence of single or a few related breeds ( refer to the heatmap in Figure 2 ) . To explore in detail the relatedness between European animals , analysis was performed separately for Mediterranean and northern-derived breeds ( Figure S3 ) . Even closely related populations such as Irish and Australia Suffolk had non-overlapping clusters , confirming the dataset provides an extremely high resolution view of population divergence . This power of resolution results from the large number of markers used , as a pilot study using only 1 , 315 SNP failed to distinguish closely related European-derived breeds [9] . Model-based clustering partitioned the genome of each animal into a predefined number of components ( K ) [18] . For unsupervised clustering assuming two ancestral populations ( K = 2 ) , a clear division was observed between Northern European and Asian breeds ( Figure S4 ) , corresponding to PC1 . Clusters were reproducible up to K = 9 and grouped individuals according to their geographic origin in the same way as for PCA ( Figure 2 ) . The 20 largest PCs accounted for only 16% of the total variation ( Figure S5 ) , consistent with reports suggesting sheep have a weak population structure [3] , [9] . To evaluate if this was accompanied by high levels of haplotype sharing between breeds , the extent of LD was characterised by the signed r statistic between SNP pairs at different lengths ( e . g . , [19] ) . For SNP pairs separated by 10 kb or less , a high degree of conservation of LD phase was observed between all breeds ( Figure S7 ) . Given that LD at short haplotype lengths reflects population history many generations ago [20] , [21] , this also supports a common ancestral origin of all domestic breeds of sheep . The result is in contrast to cattle , where two distinct groups emerge from a similar analysis , even at haplotype lengths of 0–10 kb , reflecting the Bos taurus taurus and Bos taurus indicus sub-species and their separate domestication events [15] . To determine if our LD-based estimates of haplotype sharing and effective population size were influenced by strong admixture , simulation was performed using a mutation drift model [22] and populations designed to mimic HapMap sheep breeds . This revealed admixture did affect inferred Ne , however the impact was minimal outside of the period in which the admixture took place ( Figure S6 ) . The relationship between breeds was examined using two distance metrics . Firstly , the divergence time separating all breed pairs was estimated from LD and haplotype sharing using the methods of ( Figures S7 , S8 , S9 , S10 ) [19] . Divergence time ( in generations ) revealed a strong correspondence with known population history for recently separated breed pairs . For example , breeds established within the last 100 years ( e . g . , Poll Dorset and Poll Merino ) had the shortest divergence time ( <80 generations ) . Breeds with longer history , such as American Rambouillet , had divergence from Merino estimated at 160–240 , which matches with their export from Spain to America starting in the late 1800s . The deepest divergence was estimated at only 800 generations , which appears to be an underestimate likely reflecting the influence of admixture . Divergence times between all breeds were explored as a NeighborNet graph that had branches of approximately equal length , suggesting the approach is robust to differences in genetic drift and effective population size between populations ( Figure 3 ) . NeighborNet graphs allow for reticulation as a consequence of relatedness and mixed breed origin , and the topology of the graph reproduced both the geographic groups and relationships obtained by PCA . Reticulations were observed toward the extremity of the graph for breed pairs that clustered together in PCA ( e . g . , Dorset Horn and Australia Poll Dorset ) . Conversely breeds identified as outliers by PCA such as the Soay had branches that originated from the centre of the graph . The second distance metric , Reynold's distance , relies on allele frequency differences , and branch lengths were highly variable ( Figure 4 ) . To test for the impact of ascertainment bias in SNP selection , we compared graphs generated using different SNP sets . In each case , the graphs had highly similar topology , which argues against a major influence of bias during SNP discovery ( Figure S11 ) . Short branches were observed for Spanish , Italian , and Iranian breeds with a high heterozygosity , while long branches were found for isolated populations containing small effective population size . Omitting the crossbred populations resulted in a remarkable demarcation of the geographic clusters . The topology of the graph suggests a major migration route along an axis that runs from South-West Asia to the Mediterranean region and via central Europe to Britain and the Nordic regions . Testing of additional breeds will be required to assess if migration was strongly influenced by a Danubian colonisation route . Animal husbandry and directed mating have been used to successfully adapt sheep to a diverse range of environments and to the specialised production . Selection is predicted to alter allele frequencies within the target population for both functional mutation ( s ) and their neighbouring SNP . Global FST was calculated , which measures differentiation within each breed versus all other breeds and detects both positive and balancing selection . The genome-wide distribution of global FST for 49 , 034 SNP revealed the highest selection signal was detected on Chromosome 10 ( Figure 5 ) . The highest ranked SNP ( OAR10_29511510; FST = 0 . 682 ) was located at Mb position 29 . 54 near the Relaxin/insulin-like family peptide receptor 2 ( RXFP2 ) , which was recently linked with the absence of horns ( poll ) in sheep [23] and displayed strong evidence for selection in cattle [24] . This prompted calculation of pairwise FST between breeds defined as either being horned or polled . This recapitulated a single strong and striking selection signal at RXFP2 ( Figure 6 ) . Importantly , the FST signal was absent when polled breeds or horned were compared with each other . A total of 31 genomic regions contained the top 0 . 1% of markers ranked using global FST ( 47 SNP , Table 1 ) . This implicated 17 . 85 Mb of sequence containing 181 genes as being under selection . The exact target of selection was difficult to identify as six genes , on average , were present within each genomic region . Gene ontology ( GO ) terms associated with the 181 genes were evaluated for evidence of functional enrichment against a background set of 11 , 098 genes physically tagged by the ovine SNP50 Beadchip ( Table S5 ) . This revealed enrichment for GO terms associated with regulation of bone remodelling ( p = 5 . 5×10−5 ) and bone resorption ( p = 4 . 0×10−5 ) . Given it is unlikely all 181 genes have undergone selection but each contributed to the GO analysis , caution is required during interpretation . Nonetheless , the content of the differentiated regions strongly suggests enrichment for genes under selection given their roles in pigmentation , body size , reproduction , animal production , and domestication . Selection for specialised coat pigmentation represents breed-defining characteristics across domestic animals including sheep . Selection signals were detected spanning KIT , ASIP , and MITF ( regions 8 , 19 , and 26 on OAR 6 , 13 , and 19 , respectively , Table 1 ) . KIT and MITF interact during melanocyte development and account for pigmentation phenotypes in pigs and cattle [25] , [26] , while duplication of ASIP in sheep controls a series of alleles for black and white coat colour [27] . Global FST peaks spanned NPR2 , HMGA2 , and BMP2 , which are each involved in skeletal morphology and body size ( regions 1 , 5 , and 18 on OAR 1 , 5 , and 18 , respectively , Table 1 ) . HMGA2 is of particular interest as it was recently shown to be under selection in dogs with divergent stature [28] , [29] . Positive selection was detected surrounding two genes known to regulate growth and reproduction ( PRLR on OAR6l; TSHR on OAR 7; Table 1 ) . Prolactin receptor ( PRLP ) is a key regulator of mammalian reproduction that is critical for the onset of lactation and is associated with milk traits in dairy cattle [30] . In addition , a very strong selection sweep surrounds the thyroid stimulating hormone receptor ( TSHR ) in chicken , which given its pivotal role in metabolic regulation and the control of reproduction , was postulated to be a domestication gene [31] . Finally , an FST peak on Chromosome 6 spanned the FGF5 gene , recently shown to contain mutations in dog responsible for variation in hair type [32] . Each putative gene target for selection is recorded in Table 1 , however this does not include examples where the 31 regions intersect with previous findings arising from QTL that have not been resolved to identify individual genes . One example is Mb position 6 . 8–7 . 2 on OAR 25 , which contains QTL for wool production and quality in a number of breeds [33] , [34] . The location of all 31 regions were compared to selection signals identified within the cattle genome [15] , [24] , [35]–[40] . Eleven of the 31 genomic regions identified here appear to be under selection in cattle , suggesting genes such as KIT , FGF5 , MITF , and RXFP2 are targets for selection across multiple mammalian lineages ( Table S6 ) . To search for selection observed across multiple breeds , the number of populations that displayed divergence was plotted across the genome . This revealed peaks where selection was shared across breeds , and troughs where signals were absent or unique to only a small number of breeds . Four regions were detected with positive selection shared across 30 or more breeds , while five different regions were observed with shared balancing selection . The strongest balancing selection signal was observed for the MHC region on sheep Chromosome 20 ( Figure 7 ) , a result previously observed in other species including cattle [37] . Conversely , some selection signals were breed specific . The global sheep diversity panel contained three geographically separate samples of the Texel , a meat sheep known for its growth and muscling ( Table S1 ) . When Texels were grouped and compared against all other animals , a strong peak was detected on Chromosome 2 ( Figure 7 ) . The peak spans GDF8 , a gene known to carry a mutation in Texel responsible for muscle hypertrophy [41] .
Access to patterns of SNP diversity within a global sample of domestic sheep was used to examine the population history of a species amongst the first to be domesticated by man . Our analysis revealed this domestication process must have involved a genetically broad sampling of wild stock . Approximately 75% of modern sheep breeds have retained an effective population size in excess of 300 , higher than cattle and much higher than most breeds of dog . This suggests a highly heterogeneous pre-domestication population was recruited , and the genetic bottleneck which took place was not as severe during the development of sheep as for some other animal domesticates . It is also possible that cross-breeding with wild populations persisted following the initial domestication events to generate the diversity observed . Surveys of ovine mtDNA variability support a broad genetic base during domestication , with at least five lineages identified within modern breeds that diverged well before domestication approximately 11 , 000 years ago [4] , [11] , [42] , [43] . Three aspects of the SNP diversity documented in this study indicated high levels of gene flow have occurred between populations following domestication . First , a high degree of conservation in LD phase and haplotype sharing across short chromosomal distances was detected amongst almost all breeds independent of geographic origin . Secondly , we did not detect a strong association between genetic diversity and physical distance from the domestication centre , and thirdly , the proportion of variation explained by principal component analysis suggests a weak global population structure . High gene flow and introgression between breeds has been postulated previously , based on the phylogeographic distribution of mtDNA lineages [3] , [4] . In addition , human-mediated transportation of sheep is well documented including the export of wool sheep from Italy during the Roman period and use of British sires on the European continent from the early Middle Ages onwards [44]–[46] . What remained unclear until now , however , was the extent of admixture that accompanied these sheep transportations and the high diversity this has left within many breeds . Inspection of a much larger number of SNP than in previous studies [9] allowed PCA and model-based clustering to successfully detect a clear phylogeographic pattern within the breeds genotyped . At a global scale , clear genetic divisions were detected separating European , Asian , and Africa sheep . This division likely reflects variation between the populations that participated in the earliest migrations outwards from the domestication centre . At the breed level , isolated populations were identified as outliers in PCA with low Ne ( e . g . , Soay , Wiltshire Horn , and Macarthur Merino ) . Conversely , sheep from the Americas ( Brazil and the Caribbean ) had high Ne and clustered separately from European , African , or Asian populations . Decomposing the genome into two or more components ( K<2; Figure S4 ) revealed a genetic origin for Caribbean breeds in common with African animals mixed with those of Mediterranean Europe . Similar results have been observed for New World Creole cattle [24] . This likely reflects the transportation of animals during the migration of enslaved West Africans bought to the Caribbean as slave labourers starting in the 1500s and the introduction of sheep by European colonialists . The observed patterns of genetic variation used to make inferences about population history can be explained by neutral fluctuations and the action of genetic drift . Not all loci tested in this experiment , however , appeared neutral as clear evidence was obtained for accelerated divergence in response to selection . A genome-wide scan for differentiation using global FST revealed 31 chromosomal regions with evidence for selection . It is important to recognise that genome scans such as this , even when conducted using a meaningfully large number of loci and animals , have several limitations . Foremost amongst these is that the identification of SNP displaying outlier behaviour is not , in itself , proof that selection has taken place . Where convincing signals are detected , it can be difficult to clearly identify the target of selection within a region , and even more difficult to establish the link between selection and its morphological consequence . In this study the strongest selection signal was identified immediately adjacent to RXFP2 , a gene involved in reduced bone mass and sexual maturation [47] , [48] . Strong evidence supports that RXFP2 was targeted by breeding for the removal of horns , likely to be one of the oldest morphological modifications that accompanied domestication [49] , [50] . The gene underpins QTL for horn morphology [23] and the selection signal was reconstituted only when comparing horned with polled populations . Taken together the results represent a rare example where selection has been detected and demonstrated to have occurred in response to a clearly identified human-mediated breeding objective . Given the long-standing nature of the selection , it was surprising it gave rise to the strongest selection signal . Our interpretation is that this reflects the widespread frequency of polled animals across a large number of breeds , as this assists in generating extreme FST when calculated across all breeds . Conversely , strong selection at a locus that is private to only one or two breeds is not reflected using the global FST metric . Selection surrounding Myostatin in Texel illustrates this clearly , as a strong signal is revealed when Texels are compared with all other breeds , but it is absent from the 31 regions identified using the full dataset ( Figure 7 and Table 1 ) . Analysis was performed to search for selection signatures common to more than one domesticated species . It seems reasonable to expect common signals may exist , given some breeding goals are constant across livestock species . One example is man's desire to breed animals that display consistent pigmentation type within breeds . It follows that key pigmentation genes may show evidence for selection in more than one species , and indeed that is what was detected here for genes such as MITF and KIT ( Table S6 ) . In summary , the phenotypic variability and population history of domestic animals make them an appealing model to study the consequences of selection . This promise is being realised through the recent availability of meaningfully large collections of SNP . Applied here , patterns of diversity were examined to systematically identify genomic regions in sheep that have undergone accelerated change in response to selection . Identification of the adaptive alleles within each genomic region remains a challenge . If resolved , the outcome will be knowledge describing the functional variants that characterise differences between breeds . The analysis of genomic polymorphism conducted here carries practical consequences . With the division of animals into breeds during the last few hundred years , animal breeding has witnessed a dramatic change . Most recently , the identification of superior rams and their disproportionate genetic contribution via artificial insemination has lifted the pace of genetic gain for production traits . The population-level consequence is a dramatic reduction in effective population size , which is best illustrated for cattle where the sharp decline in Ne already threatens breed viability [15] . The finding here that the majority of breeds have retained a high genetic diversity and effective population size implies that selection response for wool , meat , adaptation , and welfare traits may be expected to continue .
The number of animals per population and geographic origin of breed development is given in Table S1 . Individuals were collected from multiple flocks to capture a representative sample of within-breed genetic diversity . Beadchip array manufacture and genotyping was performed by Illumina ( San Diego , CA ) before raw signal intensities were converted into genotype calls using the Genome Studio software . SNP that failed any of the five following criteria were removed: ( 1 ) markers with <0 . 99 call rate; ( 2 ) markers identified during clustering as having atypical X-clustering , evidence for a nearby polymorphism , compression , intensity values only , or evidence of a deletion; ( 3 ) SNP with minor allele frequency equal to zero; ( 4 ) SNP with discordant genotypes identified by comparison of 10 animals genotyped independently at Illumina ( San Diego , CA ) and GeneSeek ( Lincoln , NE ) ; and ( 5 ) SNP showing Mendelian inconsistencies within 44 trios ( dam , sire , and offspring ) and the International Mapping Flock [51] . A total of 5 , 207 were removed ( Table S3 ) , leaving 49 , 034 SNP . Genotypes are available formatted for analysis in PLINK [52] from the ISGC website [53] . Five metrics were used to estimate levels of within-breed genetic diversity ( Table S4 ) . The proportion of polymorphic SNP ( Pn ) gives the fraction of total SNP that displayed both alleles within each population . Expected heterozygosity ( He ) and the inbreeding coefficient ( F ) were estimated using PLINK [52] , while allelic richness ( Ar ) and private allele richness ( pAr ) were estimated by ADZE [54] . Analysis of allele frequency distributions , plotted separately for SNP identified by Roche 454 and Illumina GA sequencing , indicated the presence of ascertainment bias ( Figure S1 ) . To determine its effect on estimates of genetic relatedness between populations , Reynold's distance was calculated between breeds using five different subsets of SNP ( Figure S11 ) . The SNP sets were as ( i ) all 49 , 034 SNP , ( ii ) 33 , 115 SNP identified using Roche 454 , ( iii ) 15 , 427 SNP identified using Illumina GA , and ( iv ) 22 , 678 SNP identified by application of LD pruning using PLINK–indep ( 50 5 0 . 05 ) . This calculated LD between SNP in windows containing 50 markers before removing one SNP from each pair where LD exceeded 0 . 05 and ( v ) 20 , 279 SNP polymorphic in non-domestic sheep that were SNP pruned using LD as described for ( iv ) . The resulting five NeighborNet trees were almost identical , indicating ascertainment bias did not have a large impact on the interpretations based on genetic distance . The removal of SNP in high LD has been shown to counter the effect of ascertainment bias and generate meaningful comparisons between populations [10] . LD-based pruning as described above preferentially reduced mean SNP heterozygosity within European populations used heavily during SNP discovery . In order to understand the relationship within and between breeds across each major geographic group , Principal Components Analysis ( PCA ) was performed using EigenStrat [55] . Initial PCA using all 2 , 819 animals revealed six breeds containing in excess of 100 animals skewed the clustering . This prompted a reduction in the number of animals used , where 1 , 612 animals were randomly selected to ensure 26 or fewer animals were included per breed ( Figures 2 and S3 ) . To ensure uncorrected LD did not distort the PCA [55] , SNP pruning was used to identify two SNP sets . First , all 49 , 034 markers were subjected to LD-based pruning ( >0 . 05 ) using PLINK to identify 22 , 678 SNP . Secondly , 32 , 847 SNP that retain polymorphism within wild feral sheep were subjected to the same LD-based SNP pruning ( >0 . 05 ) to identify 20 , 279 SNP . The PCA results obtained did not differ significantly dependent on the SNP set used . Model-based clustering was performed using the admixture model , correlated allele frequencies , and 15 , 000 burnin and 35 , 000 simulation cycles in STRUCTURE version 2 . 3 [18] . Convergence was checked using two runs for each value of K ( number of subpopulations ) . For supervised clustering , prior population information was introduced from six meta-populations consisting of regional pool of breeds considered to represent ancestral populations . The same meta-populations were used for updating the allele frequencies during the simulations . NeighborNet graphs were constructed from a matrix of Reynolds' distances using Splitstree [56] . To estimate historic effective population size for each breed , the degree of linkage disequilibrium ( LD ) was calculated as r2 between all SNP pairs where MAF for each SNP in the pair was >0 . 10 . r2 values were grouped into bins based on the distance between SNP from the physical map . Nt was then calculated as ( 1−r2 ) / ( 4cr2 ) , where c is the distance between the SNPs in Morgans ( we assumed 100 Mb = 1 Morgan ) and Nt is the effective population size t generations ago , where t = 1/2c . The most recent estimate of effective size was taken as Nt when c = 1 Mb . We performed simulations to assess the sensitivity of the estimates of effective population size over generations based on LD , in populations with and without admixture events ( Figure S6 ) . A mutation-drift model was used in the simulations following [22] . The population consisted of individuals made up of a chromosome segment 50 Mb long with 6 , 901 SNP . A population of individuals was simulated with an initially very large population size 10 , 000 generations ago , declining to a small effective population size in recent generations . In the final 420 generations , the population was split into two “breeds . ” In the non-admixed population , there was complete divergence between the breeds for the 420 generations . In the first admixed population , there was an admixture event , with crossing between the breeds ( matings chosen at random across the two breeds ) 220 generations ago . The admixing lasted 20 generations , after which the breeds diverged for a further 200 generations , with no more admixture events . LD ( r2 ) was calculated between all marker papers and Ne estimated at different times in the past as described for the real data . Five replicate simulations were performed for each scenario . The extent of haplotype sharing among populations was characterised with the r statistic , where r is a signed r2 [19] . A high correlation between r values for all locus pairs separated by the same physical distance among two breeds requires that the same haplotypes are found within both breeds . This means the sign of the r statistic is preserved across breeds only if the phase relationship among alleles is the same in both populations ( leading to a high value for r if this is the case ) . The correlation of r between breeds was calculated for SNP separated by <10 kb , 10–25 kb , 25–50 kb , 50–100 kb , and 100–250 kb ( Figures S7 , S8 , S9 ) . There will be some error in calculating the correlation of r between two breeds due to finite sampling of haplotypes within a breed ( e . g . , limited sample size ) . To determine the extent of this error , we calculated the correlation of the r values at these different lengths of haplotypes for the Merino and Industry Merino samples , which are samples from the same breed . This gave a correlation between the r values for each bin size of 0 . 6 . All correlations of r values for all breed comparisons were then divided by 0 . 6 to correct for sampling . Only corrected values are presented . As detailed in [19] the change in correlation of r between two breeds with increasing marker distance can be used to estimate generations since divergence from a common ancestral population . From [19] , the expectation for r after T generations of divergence is E ( rT ) = e−2cT . The natural logarithm of the expected correlation of r then follows a linear decrease as a function of distance with slope −2T , and this was used to calculate divergence time between all breeds ( Figure S10 ) . Global FST was calculated as described by [57] . Raw values were ranked and used to identify regions under position selection . Centred on the top SNP ( 0 . 1% ) , neighbouring markers were included until consecutive markers were encountered ranking outside of the top 5% . The second marker was excluded and the Mb position of each region was determined using sheep genome assembly version 1 . SNP-specific FST values were smoothed using a local variable bandwidth estimator as described in [35] and plotted as a line in Figures 6 and 7 . To identify genomic regions with shared selection signals across breeds , raw FST within each population was smoothed into 500 genomic divisions ( 98 SNP per region ) . The number of breeds with smoothed FST in excess of one standard deviation of the mean was plotted for values at each tail of the distribution . Analysis was performed to identify gene ontology ( GO ) terms that were significantly overrepresented in 181 genes residing within the 31 regions under selection ( Table 1 ) . The terms associated with the 181 genes were compared against a background set of 11 , 098 genes . Each of the 11 , 098 genes contain a SNP present on the SNP50 Beadchip , or a SNP within 2 . 5 Kb . Comparison of the two gene lists ( target and background ) was performed using the software GOrilla , which implements a hypergeometric distribution and mHG p value approach to determine significance [58] . | During the process of domestication , mankind recruited animals from the wild into a captive environment , changing their morphology , behaviour , and genetics . In the case of sheep , domestication and subsequent selection by their animal handlers over thousands of years has produced a spectrum of breeds specialised for the production of wool , milk , and meat . We sought to use this population history to search for the genes that directly underpin phenotypic variation . We collected DNA from 2 , 819 sheep , belonging to 74 breeds sampled from around the world , and assessed the genotype of each animal at nearly 50 , 000 locations across the genome . Our results show that sheep breeds have maintained high levels of genetic diversity , in contrast to other domestic animals such as dogs . We also show that particular regions of the genome contain strong evidence for accelerated change in response to artificial selection . The most prominent example was identified in response to breeding for the absence of horns , a trait now common across many modern breeds . Furthermore , we demonstrate that other genomic regions under selection in sheep contain genes controlling pigmentation , reproduction , and body size . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"analysis",
"tools",
"genomics",
"genetics",
"population",
"genetics",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] | 2012 | Genome-Wide Analysis of the World's Sheep Breeds Reveals High Levels of Historic Mixture and Strong Recent Selection |
Increased chronic immune activation and inflammation are hallmarks of HIV/SIV infection and are highly correlated with progression to AIDS and development of non-AIDS comorbidities , such as hypercoagulability and cardiovascular disease . Intestinal dysfunction resulting in microbial translocation has been proposed as a lead cause of systemic immune activation and hypercoagulability in HIV/SIV infection . Our goal was to assess the biological and clinical impact of a therapeutic strategy designed to reduce microbial translocation through reduction of the microbial content of the intestine ( Rifaximin-RFX ) and of gut inflammation ( Sulfasalazine-SFZ ) . RFX is an intraluminal antibiotic that was successfully used in patients with hepatic encephalopathy . SFZ is an antiinflammatory drug successfully used in patients with mild to moderate inflammatory bowel disease . Both these clinical conditions are associated with increased microbial translocation , similar to HIV-infected patients . Treatment was administered for 90 days to five acutely SIV-infected pigtailed macaques ( PTMs ) starting at the time of infection; seven untreated SIVsab-infected PTMs were used as controls . RFX+SFZ were also administered for 90 days to three chronically SIVsab-infected PTMs . RFX+SFZ administration during acute SIVsab infection of PTMs resulted in: significantly lower microbial translocation , lower systemic immune activation , lower viral replication , better preservation of mucosal CD4+ T cells and significantly lower levels of hypercoagulation biomarkers . This effect was clear during the first 40 days of treatment and was lost during the last stages of treatment . Administration of RFX+SFZ to chronically SIVsab–infected PTMs had no discernible effect on infection . Our data thus indicate that early RFX+SFZ administration transiently improves the natural history of acute and postacute SIV infection , but has no effect during chronic infection .
The current paradigm of HIV/SIV pathogenesis is that chronic systemic immune activation is a major determinant of progression to AIDS , independent of the levels of chronic viral replication or CD4+ T cell depletion [1] . Several lines of evidence support this paradigm . In both HIV-infected patients and SIV-infected macaques on antiretroviral therapy ( ART ) , poor prognosis and high incidence of non-AIDS comorbidities are associated with residual increased levels of immune activation , and not with viral replication , which is controlled with antiretroviral ( ARV ) drugs [2–4] . Additionally , comparative studies between progressive and nonprogressive animal models of HIV infection [5–7] revealed that the lack of disease progression in natural hosts of SIVs is due to their ability to control immune activation rather than contain virus replication during chronic infection [8–15] . In stark contrast , pathogenic SIV infection of macaques is associated with high levels of immune activation and inflammation , as illustrated by increases in both adaptive and innate immune effectors [16–18] . The consequences of persistent immune activation during chronic HIV/SIV infection are plentiful but not completely elucidated [19] . In HIV-infected patients and SIV-infected macaques receiving ART , residual increased levels of systemic and tissue T cell immune activation strongly correlate with incomplete immune restoration [20] , non-AIDS comorbidities and premature aging [21] . Therefore , it is critical to identify therapies that resolve immune activation , with the goal to improve the response to ART , increase the quality of life and ultimately prolong survival in HIV-infected patients [21] . The causes of the persistent immune activation in the clinical setting of controlled virus replication in patients and macaques on ART are not yet fully elucidated , which is a major obstacle to the achievement of complete therapeutic success . Systemic immune activation is correlated with increased levels of systemic circulating microbial products translocated from the intestinal lumen , as a consequence of the severe mucosal damage occurring during HIV infection [22–25] . These microbial products translocate relatively early during HIV/SIV infection and overstimulate numerous innate and adaptive immune effectors , resulting in excessive inflammation [22–25] . Strong support for a significant role of microbial translocation in triggering chronic immune activation in HIV-infected patients also comes from studies of SIV-infected NHPs . In macaques , progression to AIDS is correlated with increased levels of microbial translocation and circulating microbial products [24] , while the lack of disease progression in the natural African NHP hosts of SIVs is associated with maintenance of the mucosal barrier and no significant change in microbial translocation [11 , 26–28] . Furthermore , experimental modeling of microbial translocation in natural hosts resulted in increased levels of systemic immune activation and CD4+ T cell depletion [29 , 30] . Altogether , these NHP studies point to microbial translocation as a key determinant in driving excessive chronic immune activation during HIV/SIV infection . Early HIV/SIV-induced intestinal damage results in increased gut inflammation , which in turn fuels virus replication that deepens the intestinal dysfunction favoring increased microbial translocation , further increasing systemic immune activation , non-AIDS comorbidities and ultimately progression to AIDS [1] . With the goal of breaking this vicious cycle of AIDS pathogenesis , we designed a therapeutic strategy based on a combination of a luminal antibiotic ( Rifaximin-RFX ) and a gut-targeting antiinflammatory drug ( Sulfasalazine-SFZ ) . RFX is an intraluminal antibiotic with broad activity against Gramneg , Grampos , Mycobacteria and anaerobes that is used in patients with hepatic encephalopathy , and can reduce overall intestinal microbial burden [31] . SFZ is used in patients with inflammatory bowel disease and can reduce mucosal inflammation [32] . Both clinical categories are associated with elevated microbial translocation levels , similar to those observed in HIV-infected patients [22 , 25] . Our therapeutic approach was thus designed to reduce both the levels of intestinal inflammation and microbial burden in the intestine and ultimately reduce microbial translocation-induced systemic immune activation in SIVsab-infected PTMs . We report that early administration of RFX+SFZ to acutely SIV-infected PTMs significantly reduced levels of microbial translocation and systemic immune activation , limited mucosal CD4+ T cell loss , and reduced the levels of viral replication . We also report that administration of both drugs to ART-naïve chronically SIVsab-infected PTMs failed to modify these key parameters of infection .
Fifteen PTMs were challenged intravenously with plasma equivalent to 300 TCID50 of SIVsabBH66 . At the time of virus inoculation , five PTMs ( treated group; red in graphs ) received oral therapy with RFX ( 400 mg/day ) and SFZ ( dosed at 75 mg/kg for the first month of treatment , then adjusted to 25 mg/kg for two months ) . These doses were in the range of those administered to human subjects [31–33] . The rationale for the adjustment of the SFZ dose at the end of the acute infection is that we reasoned that a higher dose of SFZ would better control the higher levels of inflammation characteristic to acute SIV infection , while at the end of acute infection , with the partial resolution of inflammation , it would be appropriate to bring the SFZ dose to a maintenance dose similar to that used in patients with inflammatory bowel disease . Total duration of therapy was 3 months . Seven PTMs were used as untreated SIVsab-infected controls ( control group; black in graphs ) in which infection followed its natural course . We chose to initiate the RFX+SFZ treatment during the acute stage of SIV infection , despite the majority of HIV-infected patients receiving ART and presenting with residual immune activation are in more advanced stages of chronic infection , because the mucosal barrier damage inflicted by HIV/SIV occurs during the first few weeks of infection [23 , 34] . This early therapeutic approach thus permitted us to assess the efficacy of the RFX+SFZ treatment by measuring the major parameters of SIV infection before and after mucosal injury , without any interference of ART . To this end , we collected peripheral blood , duodenal and lymph node ( LN ) biopsies at multiple critical time points of infection , as shown in S1 Fig . Due to the relatively small size of the groups and to variations in the different parameters between different animals , we used baseline levels of the tested parameters ( average of three different time points ) to normalize the results as fold increase from the baseline , except for viral load and peripheral CD4+ T cells . Thus , throughout the paper , graphs illustrate the arithmetic mean of the fold increase from baseline in both treated ( in red ) and control groups ( in black ) . Detailed results are presented as supplementary information ( S2–S6 Figs ) . We assessed treatment efficacy using mixed-effects models in two distinct periods: the acute phase , between 1 and 21 days postinfection ( dpi ) , and the chronic stage of infection , >40 days post infection ( see Methods for details ) . The ultimate goal of the administration of RFX+SFZ was to provide proof of concept data showing that lowering microbial translocation and impacting systemic inflammation and immune activation breaks the vicious circle of SIV pathogenesis . Therefore , rather than assessing the independent effects of each drug , we first assessed the impact of the RFX+SFZ therapy , by comparing levels of circulating microbial products between treated and untreated PTMs . We used two different methods to assess the impact of RFX+SFZ therapy on systemic levels of microbial translocation . First , we measured LPS levels in plasma and tissues ( Figs 1 , 2 and S2 ) . LPS levels in the bloodstream were shown in numerous studies to be an excellent surrogate of the breakdown of the gut-mucosal immune barrier in HIV/SIV infected humans and NHPs [24 , 34 , 35] . Using this method , we determined that administration of RFX+SFZ significantly reduced the levels of microbial products translocated into the bloodstream of treated PTMs compared to untreated controls throughout the follow-up ( Fig 1a ) ( p = 0 . 0004 for the difference in the acute phase and p = 0 . 0003 for the early chronic stage of infection ) . However , LPS levels increased in treated PTMS during the early chronic stage of infection , indicating a loss of treatment efficacy . Similar to LPS testing , the sCD14 assay showed that administration of RFX+SFZ significantly lowered the levels of macrophage activation throughout the follow-up ( Figs 1b and S2 ) . sCD14 levels were significantly lower in treated vs . untreated PTMs during acute infection ( p = 0 . 01 ) and also in early chronic infection ( p = 0 . 007 ) ( Fig 1b ) . Since plasma LPS levels are a reflection of both alterations in intestinal barrier integrity and of LPS clearance by the liver [22 , 25] , we next assessed the levels of microbial translocation in SIVsab-infected PTMs using immunohistochemical staining for LPS [16]performed on peripheral LN samples collected at critical time points after SIVsab infection from treated and untreated PTMs . IHC revealed that the number of LPS positive macrophages was significantly lower in LNs of treated animals during the first 42 dpi . However , like for the plasma LPS levels , by the end of treatment we could detect a significant increase in the LPS levels in peripheral LNs in PTMs receiving RFX+SFZ therapy ( Fig 2 ) . Altogether , these results demonstrate that administration of both an antibiotic and an antiinflammatory drug targeting both intraluminal microbial burden and local luminal inflammation transiently alleviates the levels of microbial translocation in acutely SIV-infected NHPs . We next assessed the impact of RFX+SFZ treatment on systemic levels of T cell immune activation by comparing the expression of HLA-DR and CD38 on CD4+ and CD8+ T cells between treated and nontreated SIVsab-infected PTMs . HLA-DR and CD38 expression on both CD4+ T and CD8+ T cells was lower in the PTMs receiving the RFX+SFZ therapy compared to untreated controls ( Figs 3a , 3b and S3 ) . The differences between the two groups were significant throughout the acute infection for both CD4+ ( p = 0 . 008 ) , and in CD8+ T cells ( p = 0 . 001 ) ( Fig 3a and 3b ) . During chronic infection , the differences in HLA-DR expression were only significant for CD8+ T cells ( p = 0 . 04 ) ( Fig 3b ) , due to large variability in the CD4+ T cells of the control animals ( Figs 3a and S3 ) . We then assessed T cell proliferation by measuring the expression of Ki-67 on CD4+ and CD8+ T cells . While no significant difference could be established with regard to Ki-67 expression between the two groups , a trend toward a lower increase in the levels of Ki-67-expressing T cells was observed in animals receiving RFX+SFZ during the acute infection ( Figs 3c , 3d and S3 ) . Altogether , these results indicate that therapy with RFX+SFZ moderately and transiently reduced the levels of systemic immune activation in acutely SIV-infected PTMs . Microbial components are known as potent triggers of inflammatory responses by numerous lymphoid and nonlymphoid cells . We therefore assessed the effect of RFX+SFZ therapy on systemic proinflammatory responses in SIVsab-infected PTMs by comparing the levels of a variety of cytokines/chemokines in plasma collected from treated and untreated PTMs . RFX/SFZ therapy resulted in a reduction of proinflammatory cytokine and chemokine responses ( Figs 4 and S4 ) , as illustrated here for TNF-α ( Fig 4a ) , and I-TAC ( Fig 4b ) . We also monitored C-reactive protein ( CRP ) , a marker of acute inflammation found to predict mortality in HIV-infected patients by the INSIGHT/SMART trial [36] . CRP levels were lower in PTMs receiving RFX+SFZ , but this did not reach significance due to high variability among animals ( Figs 4c and S4 ) . Our results indicate that RFX+SFZ therapy transiently reduced inflammatory responses in SIVsab-infected PTMs . We next assessed the clinical impact of our therapeutic approach on early stages of SIVsab infection . The levels of viral replication were significantly impacted by administration of RFX+SFZ during acute infection , when the treated animals received a higher dose of SFZ , and acute VLs in the treated group were significantly lower than those in the control group ( p<0 . 0001 ) ( Figs 5a and S5a ) . During the chronic infection , this effect waned , as the levels of viral replication were not statistically different between the two groups ( Figs 5a and S5a ) . Since a causal relationship between systemic immune activation and CD4+ T cell loss was previously reported [37–40] , we compared absolute counts and percentages of CD4+ T cells in blood and intestine between treated and untreated PTMs at key times postinfection . We found somewhat lower levels of peripheral CD4+ T cells in untreated vs . treated PTMs ( p = 0 . 036 ) ( Figs 5b and S5b ) . Importantly , this was also the case if we normalized each animal’s peripheral CD4+ T cells by its initial values ( p = 0 . 011 for the fold change ) . This difference in the CD4+ T cell counts was confirmed by analyses of mucosal sites , where there was a tendency for better preservation of CD4+ T cells in PTMs receiving RFX+SLZ therapy compared to controls ( Figs 5c and S5c ) ( p = 0 . 003 and p = 0 . 0001 , for acute and chronic infection , respectively ) . Our results suggest that even a partial reduction of inflammation helps to better preserve mucosal CD4+ T cells . Note however , that the nature of this pilot project in which the follow-up was relatively short prevented us from assessing the long-term impact of RFX+SFZ therapy on mucosal CD4+ T cell restoration , which occurs later during SIVsab infection in PTMs [17] . Recent studies have suggested a strong association between hypercoagulability and inflammation during both HIV and SIV infections [29 , 36 , 41] . Therefore , we compared the levels of coagulation marker D-dimer ( 2-DD ) in treated vs . untreated PTMs and report that administration of RFX+SFZ resulted in a significant reduction of 2-DD levels throughout the treatment ( Figs 6a and S6 ) ( p = 0 . 002 for both acute and chronic infection ) . It has been proposed that the mechanism responsible for the hypercoagulable status observed in HIV patients involves tissue factor ( TF ) overexpression in response to microbial products [42 , 43] . As such , we measured plasma soluble TF ( sTF ) levels in RFX+SLZ-treated PTMs and compared them to those observed in untreated PTMs . We report that , while a 3–4 fold increase in sTF expression occurred during acute SIVsab infection in untreated PTMs relative to the control group , plasma levels of sTF remained virtually unchanged from the baseline in PTMs that received RFX+SFZ therapy ( Figs 6b and S6 ) ( p = 0 . 0001 for the change in TF levels in acute infection , and p = 0 . 055 , for the difference of those levels in chronic infection ) . Our results thus reinforce the role of microbial translocation-induced immune activation in development of cardiovascular comorbidities associated with HIV/SIV infection . Our results showed that therapy with RFX and SFZ administered early during SIVsab infection of PTMs had a higher efficacy during the acute stage of infection ( Figs 1–6 ) . The vast majority of HIV infected patients are , however , in the chronic stage , when the mucosal lesions are extensive and thus microbial translocation from the lumen fuels high levels of immune activation . It is therefore likely that in chronically-infected patients the vicious circle of HIV pathogenesis is more difficult to break , as suggested by previous experience with ART , which cannot completely control immune activation , in spite of providing long-term control of virus replication [19] . We therefore assessed the efficacy of the RFX+SFZ therapy in a small group of chronically SIVsab-infected PTMs . Three animals were included in this study . They received the same combination of RFX+SFZ for 90 days ( starting from 170 dpi ) . Despite of using the high SFZ dose , this treatment had no discernible effect on key parameters of SIV infection . Thus , RFX+SFZ therapy did not have any impact on microbial translocation , as monitored by plasma levels of LPS ( Fig 7a ) and sCD14 ( Fig 7b ) . No effect could be observed on peripheral ( Fig 7c ) or mucosal ( Fig 7d ) CD4+ T cells , viral loads ( Fig 7e ) , or on the levels of systemic inflammation as illustrated by the virtually unchanged levels of CRP ( Fig 7f ) . Finally , administration of RFX+SFZ therapy to chronically-infected PTMs had no impact on the coagulation markers , neither on 2-DD ( Fig 7g ) nor on TF ( Fig 7h ) . Our results thus show that breaking the negative chain of events characteristic to SIV/HIV pathogenesis is more difficult during chronic infection than during the early stages that precede the onset of mucosal injuries . Note however , that , unlike HIV-infected patients , the majority of which are on ART , our RFX+SFZ study used ART-naïve NHPs . As in our original study we did not collect fecal samples to assess the impact of RFX treatment on the gut microbiome , we treated 4 SIV-naive PTMs with the same dose of RFX and comprehensively profiled fecal bacterial communities in serial fecal samples using high-throughput 16S rRNA sequencing . RFX was administered for two weeks . The fecal samples were collected before treatment initiation , at 7 and 14 dpt , as well as at 14 days after RFX treatment cessation . Our results demonstrated that RFX administration significantly shifted the composition of the fecal microbiota at treatment days 7 ( p = 0 . 006 PERMANOVA test ) and 14 ( p<0 . 001 PERMANOVA test ) , relative to baseline ( Fig 8a ) . These differences were most evident in >2 fold expansion in relative abundance of Prevotellaceae in the treatment groups at 7 dpt ( p = 0 . 003; median values of 14 . 9% and 31 . 0% at the baseline and 7 dpt , respectively ) and 14 dpt ( p = 0 . 003; median values of 14 . 9% and 37 . 4% at the baseline and 14 dpt , respectively ) . Treatment also significantly reduced fecal biodiversity , as demonstrated by the significantly altered richness and complexity indices at 7 and 14 dpt , compared with baseline ( Fig 8b ) . Although samples collected at 14 days posttreatment interruption also differed from baseline in microbiota composition ( p<0 . 053; Fig 8a ) the differences appeared to be less drastic than those observed between baseline and 7 and 14 dpt samples . For example , the median Prevotellaceae abundance at 24 dpt was no longer significantly different from baseline ( p = 0 . 15; median values of 14 . 9% vs . 25 . 6% at the baseline and 24 dpt , respectively ) . Nevertheless , both richness and complexity indices remained significantly reduced , compared to the baseline . To determine whether RFX treatment altered the fecal bacterial loads , we assayed all specimens using a pan-bacterial 16S rRNA QPCR assay and normalized results to the wet weights of the fecal aliquots that were extracted . Although average bacterial loads were reduced to 38–47% of the baseline values in all three postbaseline timepoints ( overall p = 0 . 04 ) , only the differences between baseline and 7 dpt ( p = 0 . 08 ) and 24 dpt ( p = 0 . 1 ) approached significance ( Fig 8c ) .
In this study we showed that administration of an intraluminal antibiotic ( RFX ) and a gut-focused anti-inflammatory drug ( SFZ ) transiently improved the natural history of the early stages of SIVsab infection in PTMs . In NHPs receiving this treatment , a partial control of microbial translocation , immune activation and inflammation could be documented , resulting in a transient amelioration of their deleterious consequences , such as CD4+ T cell depletion or the procoagulant state . Therefore , our results support therapies aimed at controlling immune activation as effective approaches to improve the outcome of HIV/SIV infection . The rationale for such a therapeutic approach for HIV/SIV infections is based on the current paradigm of AIDS pathogenesis [1] . In this paradigm , upon infection , activation of target cells and massive acute viral replication induce inflammation in the gut . At this site , massive immune and epithelial cell death occur due to virus depletion of target cells and activation-induced bystander apoptosis . Cell death and inflammation alter the mucosal barrier and result in translocation of microbial elements from the intestinal lumen to the general circulation [1 , 22 , 23 , 25 , 34] . The main consequence of microbial translocation is systemic immune activation , which further increases target cells for the virus , thus fueling virus replication and additional immune activation . This vicious circle in which each process fuels the others ultimately results in a complete exhaustion of the immune system and death . In support of this paradigm are not only studies demonstrating that high viral replication and mucosal CD4+ T cell depletion that triggers this deleterious chain of events occur very early in the infection [44–47] , but also the results of clinical studies ( i . e . , SMART/INSIGHT ) showing that biomarkers of inflammation and microbial translocation are associated with increased risk of death [36 , 41] . While most of the results in the field converge and support this paradigm , the vast majority of available data is correlative . Some authors contest a central role of microbial translocation in the pathogenesis of AIDS , arguing that it is not possible to clearly establish whether or not microbial translocation is a cause or a consequence of generalized immune activation [48–52] . Here , rather than directly testing the primordiality of microbial translocation over inflammation ( or vice versa ) in driving disease progression , we designed a therapeutic approach consisting of drugs targeting both mucosal inflammation ( SFZ ) and intestinal microbial burden ( RFX ) . We reasoned that this approach would result in a reduction of microbial translocation by either limiting mucosal breaches or by limiting availability of microbial products , thus reducing immune activation and comorbidities . Rather than targeting a specific mechanism of generalized immune activation , our intervention focused on breaking the vicious circle of SIV pathogenesis that leads to persistent systemic immune activation , comorbidities , and ultimately , disease progression . We employed our SIVsab-infected PTM model of HIV infection [17] . The reason for choosing this model over the classical SIVmac239-infected rhesus macaque model is that PTMs are more prone to mucosal dysfunction and consequently may be a better system to model interventions aimed at controlling mucosal inflammation and microbial translocation [16 , 17 , 29 , 53 , 54] . Moreover , PTMs better reproduce HIV-associated comorbidities than rhesus macaques [29 , 55] . To reduce bacterial load in the intestinal lumen and thus diminish microbial translocation into general circulation we employed RFX , a semisynthetic derivate of rifamycin with broad activity against Gramneg , Grampos , Mycobacteria and anaerobes which significantly decreases gastrointestinal flora [56] . The rationale for choosing RFX to control intestinal flora is that it has a very limited absorption into blood stream after oral administration [31 , 56] . The lack of RFX absorption explains both drug efficacy and safety in treating enteric infections . Plasma levels of RFX being negligible , bacteria are not exposed to selective pressures outside the gastrointestinal tract , and thus bacterial resistance to RFX ( one of the major concerns with antibiotics administration ) is low [31] . RFX is effective in patients with microbial translocation-associated hepatic encephalopathy [31 , 56] and is safe for long-term administration . To reduce intestinal inflammation we administered SFZ , an antiinflammatory sulfa drug derivate of mesalazine . SFZ decreases the synthesis and/or release of leukotrienes , platelet activating factor , IL-1 and IL-2 , and inhibits TNF-α signaling and HLA-DR expression in colonic epithelial cells and the levels of reactive oxygen species in neutrophils [57] . The drug was previously used in patients with inflammatory bowel disease , including ulcerative colitis and Crohn’s disease , both of which are associated with elevated levels of microbial translocation [32 , 33 , 58] . To assess the effects of RFZ+SFZ therapy on SIV pathogenesis without additional confounding factors , we did not administer ART to SIVsab-infected PTMs receiving RFX+SFZ . Meanwhile , to maximize the clinical effects , we initiated RFX+SFZ therapy at the time of PTM challenge with SIVsab . Two rationales stand behind this approach . First , this design allowed administration of the RFX+SFZ therapy prior to mucosal damage inflicted by high levels of viral replication and initial depletion of mucosal CD4+ T cells . Second , during the acute and early chronic stages of SIV infection , there are multiple and dramatic changes in biomarkers associated with immune activation , and thus therapeutic impact can more readily be detected . In a second step , we administered the same RFX+SFZ therapy to a subset of chronically SIVsab-infected , ART naïve PTMs to compare therapeutic efficacy before and after occurrence of the mucosal damage inflicted by SIVsab . One may argue that administration of the RFX+SFZ therapy should have been performed in the chronic stage of infection , under ART and that given the fact that we administered a combination of drugs; we cannot clearly conclude which of the two drugs had a more prominent effect on microbial translocation . However , we reasoned that should we document a benefit of this therapeutic approach in this system , we could then embark on additional animal studies designed to more realistically reproduce the vast majority of HIV-infected patients , who receive ART and are in more advanced stages of infection . Administration of the RFX+SFZ therapy to acutely SIV-infected PTMs clearly impacted microbial translocation , as shown by plasma levels of both LPS and sCD14 in treated PTMs compared to untreated controls . IHC assessment of LPS levels in peripheral LNs confirmed that the number of LPS positive macrophages was significantly lower in treated PTMs during the first 42 dpi . A key result of the study was that , paralleling the lower levels of microbial translocation , the levels of immune activation and inflammation biomarkers were also significantly reduced in treated PTMs . Probably as a result of a better control of immune activation , a trend toward better preservation of mucosal CD4+ T cells was observed in PTMs treated with RFX+SFZ . One may argue that the overall impact of the RFX+SFZ therapy on CD4+ T cells was very limited . This result might be due to the fact that our therapeutic approach only marginally reduced acute viral replication , which is responsible for the massive acute depletion of mucosal memory CD4+ T cells characteristic of HIV/SIV infection [26 , 27 , 44–47] . Also , the follow-up was relatively short , preventing us from assessing a putative positive impact on CD4+ T cell restoration . Nevertheless , our results point to a clear therapeutic benefit of controlling microbial translocation in SIV-infected NHPs . Finally , RFX+SFZ therapy resulted in a significant reduction in levels of sTF and of coagulation biomarker 2-DD in PTMs receiving RFX+SFZ compared to controls . Strong correlations between microbial translocation and coagulation markers were previously reported to occur in both progressive and nonprogressive HIV/SIV infections [29 , 36] . This is not surprising , as microbial products can trigger increased TF expression on immune cells [42] . TF ( thromboplastin ) being a major activator of the coagulation cascade , its increase during progressive HIV/SIV infections may explain the generation of the prothrombotic status described in chronic HIV/SIV infection . In a clinical trial targeting microbial translocation with sevelamer , an effect on sTF was reported [43] . Our study provides further evidence that interventions targeting microbial translocation reduce TF and positively impact coagulation . It was previously reported that high persistent levels of systemic inflammation are related to the development of comorbidities and premature aging in HIV-infected patients and SIV-infected NHPs [29 , 36 , 59] . Also , previous studies showed that 2-DD is strongly and independently linked to cardiac events and death [29 , 36 , 59] . We also report that RFX+SFZ therapy impacted the levels of 2-DD and TF , and we concluded that therapies aimed at reducing microbial translocation may have a potential impact on HIV/SIV-associated noninfectious comorbidities . Since we have employed more than a single drug , we cannot conclude at this stage whether the observed effects are being mediated by the presumed reduction in enteric bacterial burden mediated by RFX or by the anti-inflammatory effects of SFZ , or by both . Brenchley et al . , [24] demonstrated that treatment of SIV infected macaques with antibiotics could reduce plasma LPS . Furthermore , prolonged antibiotic treatment with co-trimixazole successfully reduced the markers of microbial translocation [60] . It thus seems reasonable that the reduction in plasma LPS and systemic activation could be a result of antibiotic treatment . On the other hand , eicosanoids , which are modulated by sulfasalazine , have been reported to have direct anti HIV effects [61 , 62] , as well as potential secondary effects through effects on immune cell activation [63] . Administration of mesalamine , a related anti-inflammatory drug to immunocompetent patients reduced diarrhea and mucosal inflammation [64] . Thus , both drugs could have contributed to the observed effects . A disappointing outcome of our study is that the therapeutic effect was only transient , which will likely limit its applicability as a therapy in HIV-infected patients . While we do not have a definitive explanation for the waning of therapeutic efficacy , there are several potential causes for the observed results . First , a limited efficacy of RFX , due to: ( i ) its inability to decrease the overall bacterial load; ( ii ) emergence of dysbiotic bacteria under RFX treatment; and/or ( iii ) emergence of microbial resistance to RFX . Due to the fact that we did not analyze the microbiome of the RFX-treated PTMs , we cannot definitively opt for any of these scenarios . It was reported that RFX does not induce significant changes in microbial abundance [65] . Based on the reported data and to determine whether or not RFX has any impact on the intestinal microbiome , we designed a study to comprehensively profile fecal bacterial communities in serial samples from 2-week RFX-treated , uninfected PTMs . We report that administration of RFX significantly altered richness and complexity of the intestinal flora in uninfected NHPs , thus directly documenting a clear therapeutic effect of RFX . Fecal bacterial loads were reduced ~2 fold even after Rifaximin treatment ended . Although treatment did not dramatically alter bacterial load , the composition and diversity of the fecal microbiota differed significantly during and after Rifaximin treatment . Our experiment was , however , short and cannot confirm that the therapeutic effect can be maintained at long term . During previous administration of antibiotic therapy to control microbial translocation in chronically SIV-infected RMs , a loss of the therapeutic effect was reported after few weeks of therapy [24] , so we cannot exclude this scenario for our current study . The offset of the antibiotic efficacy may be a consequence of the fact that the antibiotic therapy can only partially reduce the microbial flora of the gut , but cannot prevent the breaches in the intestinal wall from occurring during HIV infection . Once the lesions produced , disrupting the vicious circle of HIV/SIV pathogenesis becomes a much more difficult task to achieve and , as such , based on both our results as well as reported data from human and macaque trials [24 , 66] , it is to be expected that antibiotic therapy , especially a short term one , will have only a transient and limited impact on the levels of microbial translocation observed in HIV infection . Corroborated with the need to keep microbial translocation at bay indefinitely and with the long term drug toxicity , this observation points to the relatively limited applicability of this therapeutic strategy in HIV-infected patients . A second reason for the loss of therapeutic effect might have been the reduction in the SFZ dose that was operated after 30 days of treatment , at the end of the acute infection . The rationale behind the dose reduction was that the high levels of acute inflammation are partially resolved at the end of acute infection , which might have allowed us to pass to a maintenance SFZ dose , similar to the strategy used in patients with inflammatory bowel disease . Finally , therapeutic efficacy may have been impacted by a more extensive gut damage occurring at later stages of infection . This might have limited the ability of the two drugs to control the mucosal lesions and the levels of translocated microbial products . To assess whether the therapeutic effect is different during different stages of SIV infection , we administered RFX+SFZ ( high dose ) to three late chronically-infected PTMs , and report that this approach did not have a discernible effect on the natural history of chronic SIV infection . While these results have to be interpreted with extreme caution due to the small sample size of the study group , the differences in treatment efficacy between acutely-infected and chronically-infected PTMs suggest that the degree of gut damage at the initiation of treatment is a key determinant of the therapeutic success . Furthermore , since in this experiment we employed the high dose of SFZ , we can conclude that the dose of SFZ is likely not responsible for the loss of therapeutic efficacy observed at the passage from acute to chronic infection . These results corroborate the results of previous studies that reported a modest therapeutic response after a short course treatment with RFX in HIV-1 infected patients on ART [66] . The results of the two studies are not , however comparable , because the human trial did not utilize an anti-inflammatory drug , while our NHP study did not achieve virus control with ART . The only common point of the two studies was that they both employed severely immune suppressed subjects ( immune nonresponder HIV patients vs . late chronically-infected ART naïve PTMs ) . While a clear effect was documented after administration of RFX+SFZ , it is unlikely that such a therapeutic approach can be employed in HIV-infected patients . First , treatment efficacy was transient even in this study design in which RFX+SFZ were initiated very early postinfection , prior to the occurrence of the gut damage . Second , clinical trials , as well as our results here , showed that RFX administration during chronic HIV/SIV infection has only marginal effects [66] . Third , long-term antibiotic treatment may induce dysbiosis that can be detrimental for HIV-infected patients , for which improvement of the gut flora with probiotics is beneficial [67] . Finally , prolonged treatment with anti-inflammatory drugs may boost cardiovascular disease that is already problematic in HIV patients [36 , 68] . As such , one may downplay the therapeutic endpoint of this study and of this strategy to control microbial translocation in general . However , our results are relevant for understanding the pathogenesis of HIV/SIV infection as we demonstrate that limiting the intestinal dysfunction and control of microbial translocation even for a short period of time contribute to a significant reduction of the immune activation . Thus , our study confirm our previous results with sevelamer administration to SIVsab-infected PTMs [16] . It also complements other approaches in the field aimed at directly modulating inflammatory responses during acute SIV infection [69] . While some of these approaches may fail to improve the outcome of infection , they all point to a critical need to better understand the nature and the sources of inflammation during HIV/SIV infection in order to guide anti-inflammatory therapies which otherwise may be proven harmful . To achieve these goals , the use of NHP models is critical and warranted by the possibility to perform interventions at well-defined time points of infection , in a refined system and without any interference of multiple behavioral risk factors for comorbidities .
Fifteen male PTMs were included in this study . All animals were housed and maintained at the RIDC Park animal facility of the University of Pittsburgh according to the standards of the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) , and experiments were approved by the University of Pittsburgh Institutional Animal Care and Use Committee ( IACUC ) ( IACUC protocol: #09039 , approved in 2009 ) . The animals were fed and housed according to regulations set forth by the Guide for the Care and Use of Laboratory Animals and the Animal Welfare Act [70] . All PTMs included in this study were socially housed ( paired ) indoors in stainless steel cages , had 12/12 light cycle , were fed twice daily , and water was provided ad libitum . A variety of environmental enrichment strategies were employed including housing of animals in pairs , providing toys to manipulate and playing entertainment videos in the animal rooms . In addition , the animals were observed twice daily and any signs of disease or discomfort were reported to the veterinary staff for evaluation . For sample collection , animals were anesthetized with 10 mg/kg ketamine HCl ( Park-Davis , Morris Plains , NJ , USA ) or 0 . 7mg/kg tiletamine HCl and zolazepan ( Telazol , Fort Dodge Animal Health , Fort Dodge , IA ) injected intramuscularly . The animals were sacrificed by intravenous administration of barbiturates prior to the onset of any clinical signs of disease . All animals were intravenously infected with plasma equivalent to 300 tissue culture infectious doses ( TCID50 ) of SIVsabBH66 ( and containing 107 vRNA copies/ml ) . At the time of virus inoculation , five PTMs received therapy with Rifaximin ( 400 mg/day orally ) and Sulfasalazine ( dosed at 75 mg/kg orally for the first month of treatment , then adjusted to 25 mg/kg for the following 2 months ) . Total duration of therapy was 3 months . The remaining PTMs ( n = 7 ) were grouped as untreated SIVsab-infected controls in which infection followed its natural course . Blood was collected from all PTMs prior to infection ( day -15 and day 0 ) , during acute SIVsab infection ( 3 , 8 , 10 , 14 , 21 , and 28 dpi ) , around the viral set-point ( 35 and 42 dpi ) , and during chronic infection ( 60 , 72 , and 90 dpi ) ( S1 Fig ) . Intestinal biopsies were collected prior to infection , during acute infection ( 14 and 28 dpi ) and during chronic infection ( 42 , 72 , and 90 dpi ) , as previously described [30 , 71–73] ( S1 Fig ) . Within one hour after blood collection , plasma was harvested and peripheral blood mononuclear cells ( PBMCs ) were separated from the blood using Ficoll density gradient centrifugation . Lymphocytes from the intestine were isolated and stained for flow cytometry , as previously described [30 , 71–73] . Intestinal biopsies were processed as described previously to obtain an enriched mononuclear cell suspension . Briefly , intestinal samples were minced mechanically , washed with EDTA and subjected to collagenase digestion , followed by Percoll density gradient centrifugation [30 , 71–73] . SIVsabBH66 viral RNA ( vRNA ) loads were quantified by real-time PCR , as described previously [72–74] . Whole blood and mononuclear cells isolated from intestinal biopsies were stained for flow cytometry using a six-color technique as described previously to assess changes in the levels of major T cell populations and their immune activation status . The mAb combination used was: CD3-Pacific Blue , CD4-allophycocyanin , CD8-Texas Red , HLA-DR-allophycocyanin-Cy7 , CD38-PE ( BD Biosciences ) and Ki-67–FITC ( BD Pharmingen ) . All Abs were validated and titrated using PBMCs from PTMs [17 , 75] . Samples were stained for Ki-67 using the Ki-67/FITC–conjugated mouse anti–human mAb set ( BD Pharmingen ) as per the manufacturer’s instructions . Stained cells were analyzed with an LSRII flow cytometer ( BD Biosciences ) and FlowJo Version 7 . 6 software ( TreeStar ) . CD4+ and CD8+ T cell percentages were obtained by first gating on lymphocytes , then on CD3+ T cells . Activation markers were determined by gating on lymphocytes , then on CD3+ T cells , and finally on CD4+CD3+ or CD8+CD3+ T cells . Plasma levels of LPS were measured as previously described [24] . Several factors present in plasma have been shown to interfere with LPS measurements ( LBP , EndoCAb , HDL , plasma turbidity , proteins and triglycerides ) . Therefore , to minimize any possible interference , plasma samples were diluted 5 fold with endotoxin-free water and then heated to 85°C for 15 min to inactivate plasma proteins . Plasma LPS was quantified with a commercially available Limulus amebocyte lysate assay ( Cambrex ) , according to manufacturer’s protocol . Each sample was run in duplicate . Plasma-soluble CD14 ( sCD14 ) levels were measured as a surrogate marker of microbial translocation [24] . CD14 is a transmembrane protein which also exists in soluble form ( sCD14; both as a shed membrane form and an alternatively spliced form ) , as a part of the complex that presents endotoxin ( lipopolysaccharide-LPS ) to TLR4 on monocytes . When monocytes are activated , ectodomain shedding results in increased sCD14 levels . sCD14 is therefore surrogate for direct measurement of endotoxin or Gram negative bacteria which translocate from the intestinal lumen to the general circulation [24 , 29 , 30] . sCD14 levels were measured using a quantitative sandwich enzyme immunoassay technique ( Quantikine Human sCD14 Immunoassay , R&D Systems , Minneapolis , MN ) . The detection limit of this kit is 200 ng/mL and can range up to 5000 ng/mL at a dilution factor of 1:200 , with an interassay coefficient of variability of 7 . 19% to 10 . 9% . Immunohistochemical ( IHC ) analysis of LPS was performed on formalin-fixed , paraffin-embedded tissue samples , as described [16] . Four μm-thick sections were deparaffinized , rehydrated , and rinsed . For antigen retrieval , the sections were microwaved in Vector Unmasking Solution ( Vector Laboratories Burlingame , CA , USA ) and treated with 3% hydrogen peroxide . Sections were incubated with LPS ( Hycult Biotech , USA ) monoclonal primary antibody at a 1:100 dilution . Secondary antibodies and Avidin/Biotin complex were from the Vector Vectastain ABC Elite Kit . For visualization , sections were treated with DAB ( Dako Carpinteria , CA . USA ) , counterstained with hematoxylin , dehydrated , and mounted in a xylene-based mounting media . Quantification was performed using open source FIJI image software using 10 images , per section , per time point , per animal . The positive signal was isolated via color threshold; percent area positive was measured and averaged . Cytokine testing in plasma was done using a sandwich immunoassay-based protein array system , the Cytokine Monkey Magnetic 28-Plex Panel ( Invitrogen , Camarillo , CA ) , as instructed by the manufacturer . Results were read by the Bio-Plex array reader ( Bio-Rad Laboratories , Hercules , CA ) , which uses Luminex fluorescent-bead-based technology ( Luminex Corporation , Austin , TX ) . The analysis was focused on proinflammatory cytokines . CRP is an acute-phase protein which rises in the plasma in response to inflammation . It was first identified in the serum of patients with acute inflammation that reacted with the C-polysaccharide of Pneumococcus . CRP binds to phosphocholine expressed on the surface of dead cells and some types of bacteria , in order to activate the complement system via the C1Q complex . The SMART trial identified CRP as one of the biomarkers associated with death in HIV-infected patients [36] . CRP was measured using a monkey CRP ELISA kit ( Life Diagnostics , PA ) as per manufacturer recommendations . Coagulation status was estimated by measuring plasma levels of 2-DD . 2-DD is a terminal product of plasmin acting on a fibrin clot that increases during coagulation , disseminated intravascular coagulation , and deep vein thrombosis . 2-DD was reported to independently correlate with lentiviral disease progression and death in HIV-infected patients [36] and SIV-infected macaques [29] . 2-DD was measured using a STAR automated coagulation analyzer ( Diagnostica Stago ) and an immunoturbidimetric assay ( Liatest D-DI;Diagnostica Stago ) . The analytical coefficient of variation ranged from 5%-14% . TF is a transmembrane cell-surface glycoprotein known for its role in initiating coagulation . Once TF complexes with factor VII , it can initiate both intrinsic and extrinsic pathways of coagulation . TF increase in plasma indicates a procoagulant environment and is found in patients diagnosed with malignant solid tumors [76 , 77] . In HIV-infected patients , monocyte expression of TF is correlated with HIV levels in plasma , immune activation , and plasma levels of sCD14 [42] . TF levels also correlate with plasma levels of 2-DD , reflective of in vivo clot formation and fibrinolysis [42] . TF levels in plasma were tested by IMUBIND Tissue Factor ELISA ( Sekisui Diagnostic , Lexington , MA ) based on the manufacturer’s instructions . DNA extractions from ~100 mg of stools were performed using the UltraClean Fecal DNA kit ( MoBio Laboratories , Inc . , Carlsbad , CA ) per the manufacturer’s protocols . Preparation , high-throughput sequencing , and taxonomic classification of broad-range bacterial 16S rRNA amplicon libraries followed our previous work [78–80] . In brief , PCR amplicons were generated using primers that target approximately 300 base pairs of the V1V2 variable region of the 16S rRNA gene using primers 27F-YM ( 5’ AGAGTTTGATYMTGGCTCAG ) [81] and 338R ( 5’ TGCTGCCTCCCGTAGGAGT ) [82] . Illumina paired-end sequencing was performed on the Miseq platform with version v2 . 3 . 0 . 8 of the Miseq Control Software and version v2 . 3 . 32 of MiSeq Reporter , using a 600-cycle version 3 reagent kit . Paired-end reads were assembled then aligned and classified with SINA ( 1 . 3 . 0-r23838 ) [83] using the 479 , 726 sequences in Silva 115NR99 [84] as reference configured to yield the Silva taxonomy . A total of 4 , 475 , 358 high-quality 16S sequences were generated , with a median of 192 , 833 sequences/sample ( range: 141 , 630–365 , 560 ) . Operational taxonomic units ( OTUs ) were produced by clustering sequences with identical taxonomic assignments . Relative abundances of OTUs were calculated for each subject by dividing the sequence counts observed for each OTU by the total number of high-quality bacterial 16S rRNA sequences generated for the subject . All sequence libraries had Goods coverage scores ≥ 99 . 9% at the rarefaction point of 141 , 630 sequences , indicating that sequence coverage was excellent . All DNA sequence data were deposited in the NCBI short read archive under Project PRJNA296587 . The Explicit sequence analysis software package ( v . 2 . 7 ) [85] and R-statistical package ( v . 2 . 15 . 2 ) were used for all microbiome statistical analyses . Ecological indices of richness ( Sobs , Schao1 ) , diversity ( Shannon’s diversity [Ho] ) , evenness ( [Ho/Hmax] ) , and coverage ( e . g . , Good’s index ) were computed through bootstrap resampling ( 1000 replicates ) and rarefaction of the OTU distributions obtained from each specimen . Differences in microbiome composition ( i . e . , OTU distributions taken as a whole ) between infant subsets were quantified by the Bray-Curtis beta-diversity index using the adonis function of the vegan R package , which performs a non-parametric multivariate analysis of variance ( PERMANOVA with 50 , 000 replicate resamplings ) [86] . Individual OTUs that differed in abundance between groups were identified by Kruskal-Wallis non-parametric analysis of variance tests . Because of the exploratory nature of the microbiome analysis and the relatively small sample size , we did not correct P-values for multiple comparisons . Approximately 25 mg aliquots of stool were weighed , and then DNA purified using the UltraClean Fecal DNA kit ( MoBio Laboratories , Inc . , Carlsbad , CA ) . 16S rRNA gene copy numbers were measured in duplicate for each sample using a panbacterial quantitative PCR TaqMan assay [87] . 16S copy numbers were estimated by reference to a dilution series standard curve of a plasmid carrying a 16S rRNA gene [88] . Results were averaged for technical replicates and normalized by the weight of each stool aliquot . Log10-transformed data were analyzed across all four timepoints by ANOVA and pairs of timepoints by Tukey Honest Significant Difference tests . We compared microbial translocation , cytokine and other immune parameters between RFX+SFZ-treated and control animals at two time periods separately: acute phase and post-acute phase . To improve the power of these analyses , we used linear mixed-effects models [89] . In this approach , we use all the measurements available together , and use macaque as the grouping ( or random ) factor to account for the repeated measurements made in each animal . We tested multiple models with fixed effects for time and treatment , with or without interactions . In this way , we are analyzing not only differences in the levels of the variable between treated and control monkeys , but also whether there is a difference in the variability of those levels over time ( corresponding to the interaction term ) . During the acute phase the behavior of the different parameters assayed over time is variable , so to allow for a general pattern of dependency of the variable on time , we considered the number of days since infection ( between day 1 and day 21 , as the acute phase ) as a categorical factor ( akin to a repeated ANOVA analysis ) [89] . For this , we need to have measurements on the same days for controls and treated animals . Therefore , we did not consider time points when there was only data for one of the groups . During the post-acute phase , from day 42 p . i . onwards , the values of all parameters tended to be more constant or changed monotonically , therefore , we considered the number of days since infection as a continuous variable ( akin to an ANCOVA analysis ) . Assumptions on the distribution of residuals and appropriateness of the fitted values were checked by visual inspection of residual and fitted plots . The best model for the data ( with or without the interaction term ) was chosen comparing the log likelihood . For these analyses we used the lme function of the nlme package [89] of R ( http://cran . r-project . org/ ) . Finally , to compare the percentage of CD4+ T cells in the intestine , for which there was only one measurement in the acute phase for both groups ( at day 14 ) , precluding the use of mixed-effects models , we used a Mann-Whitney test . With the exception of viral loads and peripheral CD4+ T-cells , we analyzed fold-changes from baseline for all variables studied . P-values<0 . 05 were considered to be significant . | We report that administration of the intraluminal antibiotic Rifaximin and the gut-focused anti-inflammatory drug Sulfasalazine to acutely SIV-infected pigtailed macaques is associated with a transient disruption of the vicious circle of inflammation-microbial translocation-immune activation which is pathognomonic to pathogenic HIV/SIV infection and drives HIV disease progression and non-AIDS comorbidities in HIV-infected patients . This therapeutic approach resulted in transient lower microbial translocation , lower systemic immune activation , lower viral replication , better preservation of mucosal CD4+ T cells and lower levels of hypercoagulation biomarkers throughout acute SIV infection . Our results thus support the use of therapeutic approaches to reduce microbial translocation , improve the clinical outcome of HIV-infected patients receiving antiretroviral therapy and prevent non-AIDS comorbidities . Our results also reinforce the importance of early therapeutic management of HIV infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | Antibiotic and Antiinflammatory Therapy Transiently Reduces Inflammation and Hypercoagulation in Acutely SIV-Infected Pigtailed Macaques |
Signaling pathways that control the activities in non-photosynthetic plastids , important sites of plant metabolism , are largely unknown . Previously , we demonstrated that WRKY2 and WRKY34 transcription factors play an essential role in pollen development downstream of mitogen-activated protein kinase 3 ( MPK3 ) and MPK6 in Arabidopsis . Here , we report that GLUCOSE-6-PHOSPHATE/PHOSPHATE TRANSLOCATOR 1 ( GPT1 ) is a key target gene of WRKY2/WRKY34 . GPT1 transports glucose-6-phosphate ( Glc6P ) into plastids for starch and/or fatty acid biosynthesis depending on the plant species . Loss of function of WRKY2/WRKY34 results in reduced GPT1 expression , and concomitantly , reduced accumulation of lipid bodies in mature pollen , which leads to compromised pollen viability , germination , pollen tube growth , and male transmission in Arabidopsis . Pollen-specific overexpression of GPT1 rescues the pollen defects of wrky2 wrky34 double mutant . Furthermore , gain-of-function activation of MPK3/MPK6 enhances GPT1 expression; whereas GPT1 expression is reduced in mkk4 mkk5 double mutant . Together , this study revealed a cytoplasmic/nuclear signaling pathway capable of coordinating the metabolic activities in plastids . High-level expression of GPT1 at late stages of pollen development drives Glc6P from cytosol into plastids , where Glc6P is used for fatty acid biosynthesis , an important step of lipid body biogenesis . The accumulation of lipid bodies during pollen maturation is essential to pollen fitness and successful reproduction .
Plastids including those that are non-photosynthetic are important sites of metabolism in plants . How the metabolic pathways in plastids and those outsides are coordinated is not well understood . Pollen , the male gametophyte , is critical to reproductive success of all flowering plants [1 , 2] . Development of the heterotrophic pollen requires energy and carbon inputs throughout the whole process [3 , 4] . At early stages , microspore is immersed in locular fluid containing nutrients from the sporophytic tapetal cells . Later , pollen maturation requires the accumulation of carbohydrates in the forms of starch and/or lipids [5 , 6] . Nutrient filling during pollen maturation is important to successful fertilization because pollen germination and pollen tube growth ( at least at the early stage ) are dependent on the storage compounds for carbon/energy sources [7–9] . Furthermore , stress-induced male sterility is frequently associated with the lack of storage compounds [10 , 11] . As a result , understanding the regulation of nutrient accumulation during pollen maturation is important to agriculture production . In Arabidopsis , mature pollen contains mostly lipid bodies , although starch is present in the vegetative cell at the early stages of pollen development [12 , 13] . Lipid body biogenesis in pollen is analogous to the formation of storage oil bodies in oil seeds [14 , 15] , which involves two important steps that occur in different organelles . The first step is the de novo biosynthesis of fatty acids in plastids , which produces acyl-CoA using carbon from source materials . The second step occurs in specialized endoplasmic reticulum ( ER ) where acyl-CoA is added to glycerol-3-phosphate ( G3P ) to form triglycerides [16–20] . In pollen , fatty acid biosynthesis in the non-photosynthetic plastids relies on the import of carbon sources carried out by a number of sugar transporters including glucose-6-phosphate/phosphate translocator ( GPT ) isoforms that mediate the import of glucose-6-phosphate ( Glc6P ) into plastids [21–26] . At present , it is unclear how these different steps are coordinated in a spatiotemporal-specific manner . It is also unclear which step is the rate-limiting step in lipid body biogenesis during pollen development . Mitogen-activated protein kinase ( MAPK , or MPK ) cascades are highly conserved signaling modules in eukaryotes [27–33] . MPK3/MPK6 , two MAPKs among the 20 MAPKs in Arabidopsis , are involved in a number of growth and developmental processes by receiving signals from different receptors/sensors [31] . The multi-functionality of MPK3/MPK6 can also be attributed to the spatiotemporal-specific phosphorylation of MAPK substrates . For instance , MPK3/MPK6 are able to phosphorylate multiple WRKY transcription factors . Depending on the cell/tissue-specific expression of these WRKYs , MPK3/MPK6 carry out unique functions in different cells/tissues/organs . In vegetative tissues/organs such as leaves , phosphorylation of WRKY33 by MPK3/MPK6 regulates phytoalexin biosynthesis in plant immunity [34] . In developing pollen , MPK3/MPK6 phosphorylate WRKY34 , and possibly WRKY2 , in a spatiotemporal-specific manner to regulate pollen development . Mutation of both WRKY2 and WRKY34 resulted in defective pollen development , germination , and pollen tube growth [35] . In this report , we demonstrate that the reduced viability and transmission of wrky2 wrky34 double mutant pollen is a result of the lack of or reduced levels of lipids , the main storage compounds in Arabidopsis pollen . WRKY2 and WRKY34 regulate the temporal expression of GPT1 , which is essential to the lipid body accumulation during pollen development . This study revealed an important cytoplasmic/nuclear signaling pathway capable of coordinating the metabolic activities in plastids and other parts of the cells . In addition , it demonstrated that Glc6P is the key source carbohydrate for lipid body biogenesis in pollen of Arabidopsis . High-level expression of GPT1 at the late stages of pollen development drives Glc6P from cytosol into plastids , where Glc6P is used for fatty acid biosynthesis , an important step of lipid body biogenesis . The accumulation of lipid bodies during pollen maturation is essential to pollen fitness and successful reproduction .
We previously reported that mutation of both WRKY2 and WRKY34 greatly reduces pollen viability , which is associated with decreased pollen germination , pollen tube growth , and male transmission [35] . A more careful examination of the transmission electron microscopic ( TEM ) images ( Fig 5L and 5M in [35] ) revealed a reduced number of oil bodies and more void spaces in wrky2 wrky34 double mutant pollen ( Note: Oil bodies were mislabeled as plastids in the images . Plastids are double membrane-bound organelles with simple membrane structures inside . In contrast , lipid bodies have a homogenous neutral lipid core bound by a phospholipid monolayer with oleosins . ) , suggesting that WRKY2/WRKY34 might be involved in regulating lipid biosynthesis during pollen maturation . As a result , we compared the expression of genes encoding important enzymes/proteins related to lipid body biogenesis in pollen grains of wrky2 wrky34 double mutant and wild type using quantitative RT-PCR . We found that the expression of GPT1 ( encoding glucose-6-phosphate/phosphate translocator 1 ) , HAD2 ( encoding β-hydroxyacyl-ACP dehydrase 2 ) , LPAAT2 ( encoding lysophosphatidic acid acyltransferase 2 ) , DGAT1 ( encoding diacylglycerol acyltransferase 1 ) , PDAT1 ( encoding phospholipid:diacylglycerol acyltransferase 1 ) , OLE5 ( encoding oleosin 5 ) , and CLO4 ( encoding caleosin 4 ) was reduced by at least 50% in wrky2 wrky34 mutant pollen . In contrast , the expression of other genes was less affected ( Fig 1 ) . A literature search revealed that , similar to wrky2 wrky34 double mutant , mutation of GPT1 also results in pollen defects including reduced lipid bodies [36] . GPTs carry out the transportation of Glc6P into plastids where Glc6P can be used for starch biosynthesis , fatty acid biosynthesis , or NADPH generation via the oxidative pentose phosphate pathway ( OPPP ) [21 , 24 , 36 , 37] . Of the two GPT isoforms in Arabidopsis , GPT1 gene expression was severely reduced in the pollen of wrky2 wrky34 double mutant plants . The expression of GPT2 was very low in pollen , but detectable using RT-qPCR . Its expression was not altered in wrky2 wrky34 mutant ( Fig 1 , inset ) . For these reasons , we set out to test whether GPT1 is a target gene of WRKY2/WRKY34 transcription factors during pollen development . As the first step , we compared the pollen viability and germination of wrky2 wrky34 double mutant and gpt1+/- mutant side by side . Because gpt1 homozygous mutant is embryo lethal , we used pollen from gpt1+/- heterozygous mutant plants . The flowers and anthers of gpt1+/- heterozygous mutant plants developed normally ( S1 Fig ) . Propidium iodide ( PI ) staining , which gives dead pollen red fluorescence under the microscope [38] , revealed a large proportion of dead pollen grains from gpt1+/- heterozygous mutant plants . In contrast , pollen grains from wild-type Ws-2 control plants were mostly non-fluorescent , i . e . viable ( Fig 2A ) . Quantitative analysis revealed 32 ± 6% ( n = 5 ) of the pollen grains from gpt1+/- heterozygous mutant plants were dead under our experimental conditions ( Fig 2C ) . Using fluorescein diacetate ( FDA ) , which stains viable pollen fluorescent green [35] , we observed a similar percentage of non-viable pollen from gpt1+/- plants ( S2 Fig ) . Because FDA staining is not compatible with GPT1-eYFP fusion-rescued gpt1 pollen ( both have green fluorescence ) , PI staining was used throughout this study . We observed a higher percentage of dead pollen grains from gpt1+/- plants than the previous report [36] , probably due to different experimental conditions . Pollen grains from gpt1+/- plants showed a significant reduction in in vitro germination rate , only 36 ± 8% ( n = 3 ) relative to 82 ± 6% ( n = 3 ) in Ws-2 wild type ( Fig 2B and Fig 2D ) . Pollen grains from wrky2 wrky34 double mutant plants have similar phenotypes , with only a 37% viable rate and 28% germination rate [35] . However , these numbers are not directly comparable with those from gpt1+/- plants because gpt1+/- plants produce 50% wild-type pollen grains . Phenotypes of gpt1 mutant pollen were attributed to defective fatty acid biosynthesis based on reduced lipid bodies in TEM images [36] . To determine whether the reduced viability of wrky2 wrky34 pollen is associated with a decrease in lipid contents , we performed TEM analysis . As shown in S3 Fig , lipid bodies in gpt1 pollen grains , similar to wrky2 wrky34 pollen grains [35] , were greatly reduced in comparison to the wild type . We also stained pollen with Nile blue A to detect the lipids . Pollen grains with sufficient lipids are stained blue . As shown in Fig 2E and 2F , approximately 58 ± 5% ( n = 6 ) pollen grains from gpt1+/- heterozygous plants and 18 ± 3% ( n = 3 ) pollen grains from wrky2 wrky34 double mutant plants were stained blue , which are much lower in comparison to their respectively wild-type controls , 89 ± 1% ( n = 3 ) in Ws-2 and 95 ± 3% ( n = 3 ) in Col-0 . Because it is difficult to quantify the amount of fatty acid based on the intensity of Nile blue A staining , we used the percentage of positively stained pollen grains as a measure of the lipid accumulation in pollen as a population . Furthermore , we stained the pollen with BODIPY 505/515 to visualize the lipid bodies . As shown in Fig 2G , the number of lipid bodies in gpt1 and wrky2 wrky34 pollen was greatly reduced . Quantitation of fluorescence intensity indicated about 50% reduction in lipid body accumulation in the mutant pollen ( Fig 2H ) . Based on these results , we can conclude that mutant pollen grains from both gpt1+/- heterozygous mutant and wrky2 wrky34 double mutant plants have a reduced level of storage lipids , which could lead to reduced pollen viability , germination , and transmission . Since pollen grains from gpt1+/- heterozygous plants are a mixture of wild-type pollen and gpt1 mutant pollen , it is difficult to 1 ) attribute a phenotype to pollen of a specific genotype , and 2 ) perform genetic analysis . To overcome these difficulties , we attempted to generate a rescued gpt1 homozygous mutant system using fluorescent tagged GPT1 , a strategy used in one of our previous studies [39] . A GPT1 promoter-driven GPT1-eYFP fusion construct ( PGPT1:GPT1-eYFP ) was transformed into gpt1+/- heterozygous plants . In T2 progenies , we successfully obtained gpt1 homozygous plants with PGPT1:GPT1-eYFP single-insertion transgenes in homozygous state ( genotype: PGPT1:GPT1-eYFP gpt1 ) . They were then crossed with gpt1+/- plants to obtain PGPT1:GPT1-eYFP+/- gpt1 plants . Successful rescue of gpt1 mutant by PGPT1:GPT1-eYFP transgenes demonstrated that the transgene product is fully functional , which allowed us to 1 ) use the fusion protein to examine the spatiotemporal expression patterns of GPT1 , and 2 ) identify the gpt1 mutant pollen , which is non-fluorescent , from PGPT1:GPT1-eYFP+/- gpt1 plants . In addition , using a plastid marker construct pt-rk CD3-999 [40] , we demonstrated that GPT1-eYFP fusion co-localizes with a mCherry plastid marker ( S4A Fig ) . This conclusion is consistent with the previous conclusion based on biochemical evidence [21] . Pollen viability assay revealed that PGPT1:GPT1-eYFP gpt1 plants had wild-type phenotype ( Fig 3A , upper panels ) . All pollen grains from PGPT1:GPT1-eYFP gpt1 plants had eYFP signal , showed normal morphology , and were PI-unstainable . Half of the pollen grains from PGPT1:GPT1-eYFP+/- gpt1 plants did not have eYFP signal ( Fig 3A , lower panels ) . They were gpt1 mutant pollen grains . A large percentage of them were PI-stainable nonviable pollen ( Fig 3A , lower panels ) . In contrast , pollen grains with eYFP signal ( ~50% ) from PGPT1:GPT1-eYFP+/- gpt1 plants , which were complemented gpt1 pollen ( genotype: gpt1 PGPT1:GPT1-eYFP ) , showed wild-type phenotype . Fig 3B showed that only 47 ± 8% ( n = 4 ) of gpt1 pollen grains was viable , which was significantly lower than the value of PGPT1:GPT1-eYFP gpt1 pollen , 94 ± 4% ( n = 4 ) . Such a decrease was consistent with the reduction of pollen viability from ~96% in the wild-type plants to ~68% in the gpt1+/- heterozygous plants ( Fig 2C ) . The viable gpt1 mutant pollen grains also suggest the presence of other transporter ( s ) , such as GPT2 and TPT , and/or pathway ( s ) that can compensate the loss of GPT1 and synthesize sufficient amount of lipids . We also noticed that all non-fluorescent pollen grains ( genotype: gpt1 ) were smaller and more rounded in shape , while the fluorescent GPT1-eYFP-rescued pollen grains ( genotype: PGPT1:GPT1-eYFP gpt1 ) were bigger and oval shaped ( Fig 3A , lower panels ) . The length of the gpt1 pollen grains was 20 . 8 ± 1 . 6 μm ( n = 156 ) , significantly smaller than the value of 26 . 6 ± 1 . 2 μm ( n = 185 ) for the rescued fluorescent pollen grains ( Fig 3C ) . Although this phenotype could be observed in the mixed pollen grains from gpt1+/- plants ( Fig 2A ) , the genotype of these smaller pollen grains was not clear . In vitro pollen germination assay showed a major reduction in gpt1 pollen germination rate , from 86 ± 2% ( n = 5 ) for PGPT1:GPT1-eYFP gpt1 pollen to 21 ± 7% ( n = 5 ) for gpt1 pollen ( Fig 3D and 3E ) . Pollen from PGPT1:GPT1-eYFP gpt1 plants had a germination rate ( 82 ± 1% , n = 3 ) similar to wild type ( 82 ± 6% , n = 3 ) ( Fig 2B and Fig 3D , upper panels ) . Nile blue A staining revealed that pollen from PGPT1:GPT1-eYFP gpt1 plants all was stained blue and had green fluorescence ( Fig 3F , upper panels ) . Among the pollen grains from PGPT1:GPT1-eYFP+/- gpt1 plants , only 31 ± 15% ( n = 5 ) of the non-fluorescent pollen grains ( gpt1 mutant pollen ) could be stained blue , while 90 ± 9% ( n = 5 ) of the fluorescent pollen grains ( PGPT1:GPT1-eYFP gpt1 rescued pollen ) were stained blue ( Fig 3F , lower panels , and Fig 3G ) . In summary , using a rescued gpt1 homozygous mutant system with GPT1-eYFP fusion transgene , we were able to quantitatively define the viability , size , germination rate , and lipid accumulation in gpt1 pollen , which would otherwise be impossible to identify specifically . The gpt1 pollen grains are smaller with reduced lipid accumulation , and have reduced viability and germination rate , which is consistent with its reduced transmission rate of 20% ( n = 526 ) based on backcross using pollen from gpt1+/- plants . Using the fully complemented PGPT1:GPT1-eYFP gpt1 plants , we analyzed the expression pattern of GPT1 reporter during pollen development . As shown in Fig 4 , GPT1-eYFP fluorescence was not visible in uninucleate microspores ( UNM ) and early bicellular pollen ( BCP ) . It became detectable at late BCP stage and early tricellular pollen ( TCP ) , peaked at TCP stage , and stayed high in mature pollen ( MP ) . GPT1-eYFP fusion protein appeared as small speckles in pollen , consistent with its localization on plastids ( S4 Fig ) . The accumulation of GPT1-eYFP is preceded by the appearance of WRKY2 and WRKY34 proteins in the vegetative nucleus [35] . As shown in the Fig 4B to 4Q of Guan et al paper [35] , both WRKY2 and WRKY34 proteins reached their peak levels at BCP stage , which is consistent with a role of these two WRKYs in regulating GPT1 expression . To visualize the accumulation of lipid bodies during pollen development , we used BODIPY 505/515 staining . As shown in Fig 4 , no or very few lipid bodies were visible in pollen at UNM and BCP stages . At later stages , there was an accumulation of lipid bodies , concurrently with the increase in GPT1-eYFP protein . Together with the loss-of-function genetic evidence , we can conclude that GPT1 plays a key role in lipid body accumulation during pollen maturation . To determine whether GPT1 expression during pollen development is regulated by WRKY2 and WRKY34 , we transformed wrky2 wrky34 double mutant plants with PGPT1:GPT1-eYFP construct . T2 homozygous plants with single transgene insertion were crossed with Col-0 to generate PGPT1:GPT1-eYFP , PGPT1:GPT1-eYFP wrky2 , PGPT1:GPT1-eYFP wrky34 , and PGPT1:GPT1-eYFP wrky2 wrky34 plants . We then compared the GPT1-eYFP signal in pollen grains from these four genotypes with the same transgene allele . As shown in Fig 5A , GPT1-eYFP signal in single wrky2 and single wrky34 mutant background was the same as that in the wild-type background based on quantification of fluorescence intensity ( Fig 5B ) . However , the GPT1-eYFP signal in the wrky2 wrky34 double mutant background was only about 30% of the wild type ( Fig 5A and 5B ) . This result is consistent with the lower expression of native GPT1 in the anthers of wrky2 wrky34 double mutant plants ( Fig 1 ) , providing further support that GPT1-eYFP expression is dependent on the functional WRKY2 and WRKY34 . We next stained pollen grains from these four genotypes with Nile blue A to determine whether reduced expression of GPT1-eYFP in wrky2 wrky34 double mutant pollen would result in a reduction in lipid accumulation . As shown in Fig 5A ( lower panels ) , pollen grains in wrky34 or wrky2 single mutant background accumulated lipids at a level similar to the wild type . In contrast , the majority of the pollen grains in wrky2 wrky34 double mutant background could not be stained . Quantitative analyses indicated that only 30 ± 6% ( n = 3 ) of the pollen grains from PGPT1:GPT1-eYFP wrky2 wrky34 plants was stained blue using Nile blue A assay , which was significantly lower than those from PGPT1:GPT1-eYFP , PGPT1:GPT1-eYFP wrky34 , and PGPT1:GPT1-eYFP wrky2 plants with percentages of 92 ± 2% ( n = 3 ) , 95 ± 3% ( n = 3 ) , and 92 ± 3% ( n = 3 ) , respectively ( Fig 5C ) . BODIPY 505/515 staining further confirmed the compromised accumulation of lipid bodies in wrky2 wrky34 double mutant pollen ( S5 Fig ) . Loss of function of GPT1 or WRKY2/WRKY34 leads to compromised accumulation of lipid bodies at the late stages ( TCP and MP ) of pollen development and reduced pollen viability . To determine when the cell death occurred in gpt1 and wrky2 wrky34 pollen , we first performed DAPI staining of pollen nuclei at different pollen developmental stages . As shown in S6 Fig , gpt1 mutant pollen was not distinguishable morphologically at the early stages ( UNM to early TCP ) , despite their smaller sized at maturity ( Figs 2 and 3 ) . In addition , all pollen grains progressed to TCP stage , suggesting that the loss of GPT1 has minimal effects on pollen development before maturation , and pollen collapse/death is a late event during maturation process . Consistent with this , PI viability staining revealed pollen death only at TCP and MP stages ( S8 Fig ) . Similar results were observed in wrky2 wrky34 pollen ( S7 and S8 Figs ) . In Arabidopsis , it is known that mature pollen contains mostly lipid bodies , although starch is present in the vegetative cell at the early stages of pollen development [12 , 13] . To determine whether gpt1 or wrky2 wrky34 mutation has an impact on starch accumulation in pollen , we performed Lugol’s iodine staining . As shown in S9 and S10 Figs , no change in starch accumulation was observed in gpt1 or wrky2 wrky34 mutant pollen . Based on the biochemical function of GPT1 protein in transporting Glc6P into plastids , it is more likely that the mutation of GPT1 results in the loss/reduction of carbon source and/or NADPH needed for fatty acid biosynthesis in plastids , and subsequently lipid body accumulation during pollen maturation , which then leads to the reduction of pollen viability . To genetically test whether GPT1 functions downstream of WRKY2 and WRKY34 in regulating pollen storage lipid accumulation , we performed epistatic analysis by overexpressing GPT1 gene in wrky2 wrky34 background . We transformed wrky2 wrky34 double mutant with GPT1-eYFP fusion driven by LAT52 , a strong pollen-specific promoter [41] . T3 homozygous lines ( genotype: PLAT52:GPT1-eYFP wrky2 wrky34 ) were selected for experiments . Pollen viability assay revealed that , while the majority of wrky2 wrky34 pollen grains showed PI fluorescence , a much smaller percentage of PLAT52:GPT1-eYFP wrky2 wrky34 pollen had red fluorescence ( Fig 6A ) . Quantitative analyses showed that PLAT52:GPT1-eYFP wrky2 wrky34 plants had a higher percentage of live pollen ( 57 ± 11% , n = 3 ) than wrky2 wrky34 double mutant plants ( 20 ± 3% , n = 3 ) ( Fig 6B ) . In pollen germination assay , 37 ± 2% ( n = 3 ) of the pollen grains from PLAT52:GPT1-eYFP wrky2 wrky34 plants germinated , a percentage 4-times as high as the germination rate of wrky2 wrky34 pollen ( 9 ± 2% , n = 3 ) under the same conditions ( Fig 6A and 6C ) . Furthermore , we stained the pollen from wrky2 wrky34 and PLAT52:GPT1-eYFP wrky2 wrky34 with Nile blue A . As shown in Fig 6A and 6D , approximately 42 ± 2% ( n = 5 ) of the pollen grains from PLAT52:GPT1-eYFP wrky2 wrky34 plants could be stained blue , i . e . with normal accumulation of storage lipids . In contrast , only 16 ± 7% ( n = 5 ) of the pollen grains from wrky2 wrky34 double mutant plants were stained blue . BODIPY 505/515 staining further demonstrated the restoration of lipid body accumulation in wrky2 wrky34 pollen with pollen-specific overexpression of GPT1 ( S11 Fig ) . These data are consistent with the higher pollen viability and pollen germination of PLAT52:GPT1-eYFP wrky2 wrky34 plants . The successful rescue of wrky2 wrky34 double mutant pollen phenotypes by pollen-specific overexpression of GPT1 demonstrates that GPT1 is a major target gene of WRKY2 and WRKY34 transcription factors in the regulation of lipid body biogenesis during pollen maturation . W-box is the cis-element known to be the binding site of WRKY transcription factors [42] . GPT1 promoter contains four copies of W-boxes within the 1 . 2 kb region ( Fig 7A ) . To test whether these four W-boxes in the GPT1 promoter are important to GPT1 expression , we mutated the core sequence ( TGAC ) of all four W-boxes to TGAA ( named mutated W-box , or mW ) and compared its activity with the wild-type GPT1 promoter . GPT1-eYFP fusion driven by wild-type promoter ( PGPT1:GPT1-eYFP ) and mutated promoter ( PGPT1-mW:GPT1-eYFP ) were introduced into wild-type plants . Experiments in Fig 7 compared two representative lines of wild-type promoter and mW promoter transgenes . Quantitative RT-PCR revealed an approximately two-fold decrease in GPT1-eYFP transcripts when the W boxes in the GPT1 promoter were mutated ( Fig 7C ) . The decrease in transcript was accompanied by a reduced GPT1-eYFP fluorescent signal ( Fig 7D ) and reduced protein level ( Fig 7E ) in the PGPT1-mW:GPT1-eYFP lines based on microscopy observation and western blot analysis , respectively . Yeast one-hybrid assay further confirmed the binding of WRKY34 transcription factor to GPT1 promoter ( S12 Fig ) . Furthermore , mutation of the W-boxes in GPT1 promoter abolished this interaction . Compromised promoter activity after the mutation of W-boxes in GPT1 promoter provides another line of evidence that the W-boxes in the GPT1 promoter are important to the activation of GPT1 expression by WRKY2 and WRKY34 during pollen development . We then examined the involvement of MPK3/MPK6 , and their upstream MKK4/MKK5 , in GPT1 expression and lipid accumulation . Previously , we demonstrated that MPK3/MPK6 phosphorylate WRKY34 , and possibly WRKY2 , and are involved in pollen development [35] . We first transformed the native promoter-driven GPT1-eYFP construct into the conditional gain-of-function GVG-NtMEK2DD ( abbreviated as DD ) background to generate PGPT1:GPT1-eYFP DD plants , and then examined the GPT1-eYFP signal in pollen after dexamethasone ( DEX ) treatment . Induction of the constitutively active DD protein leads to the activation of downstream endogenous MPK3/MPK6 [34 , 43 , 44] . As shown in Fig 8A , GPT1-eYFP signal in pollen from PGPT1: GPT1-eYFP DD plants treated with DEX was more than three times stronger than that in pollen from the same plants treated with ethanol solvent control , demonstrating that gain-of-function activation of MPK3/MPK6 is sufficient to promote the ectopic expression of GPT1 . In this experiment , we examined pollen grains at the late UNM stage when the endogenous GPT1 signal has not been turned on yet and GPT1 expression is very low ( Fig 4 ) . At later pollen development stages , the gain-of-function phenotype was not very obvious because of the high GPT1 gene activation by the endogenous signaling pathway . In a loss of function experiment , we examined GPT1 expression and pollen lipid accumulation in the newly generated mkk4 mkk5 double TILLING mutant [45] . MKK4 and MKK5 play redundant function upstream of MPK3 and MPK6 in various processes [31 , 46] . We first transformed PGPT1:GPT1-eYFP construct into mkk4 mkk5 double mutant background , selected lines with single transgene insertions , and then crossed the transgene alleles into the wild-type background . We found that pollen grains from PGPT1:GPT1-eYFP mkk4 mkk5 plants ( Fig 8B ) had much weaker ( ~50% ) GPT1-eYFP signal than that from PGPT1:GPT1-eYFP plants . Furthermore , Nile blue A staining revealed that less pollen grains were stained blue in mkk4 mkk5 double mutant background , 62 ± 8% ( n = 6 ) in comparison to 95 ± 2% ( n = 6 ) in Col-0 wild type ( Fig 8C and 8E ) . BODIPY 505/515 staining revealed that the accumulation of lipid bodies was significantly reduced in mkk4 mkk5 double mutant pollen , but not mkk4 or mkk5 single mutant pollen , demonstrating a redundant function of MKK4 and MKK5 in the process ( S13 Fig ) . We further examined storage lipid accumulation in loss-of-function mpk3 mpk6 pollen . A chemical-genetically rescued mpk3 mpk6 double mutant system [47] was utilized . Because homozygous mpk3 mpk6 double mutant is embryo lethal , we used a chemical-sensitized version of MPK6 , MPK6YG , to rescue the double mpk3 mpk6 mutant , and the resulting plants were named MPK6SR plants ( genotype: mpk3 mpk6 PMPK6:MPK6YG ) . The kinase activity of MPK6YG can be specifically inhibited by 4-amino-1-tert-butyl-3- ( 1’-naphthyl ) pyrazolo[3 , 4-d]pyrimidine ( NA-PP1 ) , a derivative of PP1 kinase inhibitor with a bulky side chain that cannot enter the ATP binding pocket of other kinases [48] . As shown in Fig 8D and 8F , pollen grains from MPK6SR plants treated with NA-PP1 had reduced fatty acid . Only 61 ± 9% ( n = 5 ) of the pollen grains were stained blue in Nile blue A assay , while this value was 93 ± 3% ( n = 5 ) for DMSO-solvent control treated MPK6SR plants . Furthermore , BODIPY 505/515 staining revealed compromised lipid body accumulation in MPK6SR plants treated with NA-PP1 , but not DMSO control ( S14 Fig ) . Together , these experiments provide loss-of-function evidence to support the role of MPK3/MPK6 in storage lipid accumulation during pollen maturation .
Lipid bodies and/or starch granules stored in the vegetative cytoplasm of the mature pollen provide carbon source material and energy to support the rapid pollen tube growth [4 , 49 , 50] . The lack of storage compounds as a result of either developmental defect or environmental stress greatly limits plant reproduction . It is known that accumulation of storage compounds happens at the late stage of the pollen development in all plants . However , the signaling pathway that controls this process was unclear . The identification of a MAPK signaling pathway , its downstream WRKY transcription factors , and GPT1 , a key target gene of WRKY2/WRKY34 transcription factors , greatly advances our understanding of this process . GPT1 is directly involved in the lipid biosynthesis by transporting Glc6P into the plastids of heterotrophic pollen where Glc6P can be converted to acetyl-CoA and used to generate reducing equivalent for fatty acid biosynthesis . Using a fully functional eYFP fusion reporter , we demonstrated that GPT1 protein starts to accumulate in BCP/TCP ( Fig 4 ) , which is consistent with the findings that lipid bodies accumulate after the first mitosis and rapidly fill up the cytoplasm of the vegetative cell [4 , 8 , 12] . The identification of a cytoplasmic/nuclear signaling pathway that regulates the metabolic activities in plastids ( Fig 8G ) greatly advanced our understanding of the coordination/regulation of plant metabolism in different cellular compartments . We speculate that the regulation of pollen storage compounds involves developmental signal ( s ) sensed by pollen surface receptor ( s ) , which then activate the MPK3/MPK6 cascade . The phosphorylation of WRKY transcription factors by MPK3/MPK6 leads to the activation of GPT1 gene expression ( Fig 8G ) . This , together with other metabolic enzymes , gives the undifferentiated plastids the capacity to synthesize fatty acids , therefore , specifies the function of plastids in pollen at late development stages . Mature pollen of Arabidopsis , an oleaginous plant , contains a large number of storage lipid bodies , which are spherical organelles with a size ranging from 0 . 1 to 2 . 5 μm and contain a TAG matrix , enclosed by a phospholipid monolayer ( PL ) with unique embedded proteins including oleosins [51 , 52] . The formation of these lipid bodies in pollen is thought to be similar to that in oil seeds [17 , 18 , 53] . As the first step , potential carbon sources need to be transported into the plastids for the synthesis of acetyl-CoA and then fatty acids ( reviewed in [24 , 26 , 37 , 54] ) In the non-photosynthetic pollen , transportation of reduced carbons into plastids could be a key step in the control of fatty acid biosynthesis . This study , and previous report [36] , demonstrated the importance of GPT1 in fatty acid biosynthesis . GPT1 imports Glc6P into plastids in heterotrophic cells/tissues . Pollen grains of gpt1 genotype accumulate little or no lipid bodies , suggesting that Glc6P is a major carbon source transported into plastids for the generation of acetyl-CoA and/or reducing equivalent NADPH , essential components of fatty acid biosynthesis . GPT1 is highly expressed in pollen at late developmental stages . In contrast , GPT2 , the only other GPT in Arabidopsis , expresses at a very low level in pollen ( Fig 1 ) . In addition , the expression of TPT and PPT in pollen is also relatively low ( Fig 1 ) , suggesting that limited amounts of triose phosphate and/or PEP are imported into the non-photosynthetic plastids for fatty acid biosynthesis in pollen . Consistently , mutation of TPT gene alone does not result in pollen phenotype and the plants are pretty much normal [55] . It is known that feeding of Glc6P to isolated plastids supports a high rate of fatty acid biosynthesis [56–58] , again supporting our conclusion that GPT1 plays an important role in lipid body biogenesis during pollen maturation . It was suggested that the activities and properties of transporters are important in determining the metabolic routes by which carbon is imported into the plastid and utilized for fatty acid synthesis [26] . In the case of Arabidopsis pollen , GPT1 appears to be the key transporter involved . Mutation of both WRKY2 and WRKY34 leads to defective pollen development , reduced pollen viability , and reduced pollen germination , pollen tube growth and transmission [35] . Similar to gpt1 mutant pollen , wrky2 wrky34 double mutant pollen also shows a lack of or reduced number of lipid bodies based on Nile blue A staining ( Fig 2E and 2F ) and BODIPY 505/55 staining ( Fig 2G and 2H , S5 Fig ) , suggesting that the defective pollen development of wrky2 wrky34 double mutant is related to GPT1 activation . We analyzed the expression pattern of GPT1 in pollen development in details , and compared it with the temporal expression of WRKY2 and WRKY34 . As shown in the Fig 4B to 4Q of Guan et al paper [35] , both WRKY2 and WRKY34 proteins reached their peak levels at BCP stage , preceding the accumulation of GPT1-eYFP . At the MP stage , WRKY34 protein disappears , while WRKY2 protein is still present [35] . This is consistent with the conclusion that WRKY34 was an early pollen gene enriched in UNMs and BCPs based on expression profiling analysis [59] . Expression profiling revealed that a large number of genes including many transcription factors show spatiotemporal-specific expression [60 , 61] . However , the signaling pathway is mostly unclear . In addition , few precedents exist about the direct control of target gene expression by those transcription factors during pollen development . Based on cellular , molecular , and genetic analyses , we demonstrated that GPT1 functions downstream of WRKY2/WRKY34 in controlling pollen development . GPT1 expression in wrky2 wrky34 double mutant background was compromised ( Fig 1 and Fig 5 ) . More importantly , pollen-specific overexpression of GPT1 could partially rescue the defective pollen phenotypes of wrky2 wrky34 double mutant ( Fig 6 ) . Furthermore , both GPT1-eYFP transcript and protein levels were reduced when the W-boxes in the GPT1 promoter were mutated ( Fig 7 ) . Taken together , we conclude that spatiotemporal-specifically expressed WRKY2 and WRKY34 transcription factors target directly the GPT1 promoter and control its spatiotemporal-specific expression , which specifies the function of undifferentiated proplastids by promoting the storage lipid biosynthesis during pollen maturation . The partial rescue of wrky2 wrky34 phenotype by pollen-specific expression of GPT1 also indicates that these two WRKY transcription factors might be involved in regulating other downstream genes . Besides GPT1 , WRKY2 and WRKY34 may control the expression of additional genes involved in lipid body biogenesis . As shown in Fig 1 , the expression of enzymes in TAG biosynthesis such as LPAAT2 , DGAT1 , and PDAT1 were all reduced . In addition , expression of genes encoding the proteins embedded in the phospholipid monolayer that surrounds oil bodies including OLE and CLO was also reduced in wrky2 wrky34 double mutant . At this stage , it is unknown whether all these genes are co-regulated by these two WRKY transcription factors or , alternatively , their expression reduction is a secondary response caused by the lack/reduction of fatty acid biosynthesis . It is interesting to note that DGAT1 and PDAT1 were shown to have overlapping functions in Arabidopsis triacylglycerol biosynthesis and they are essential for normal pollen and seed development [15] . Double dgat1 pdat1 mutation results in sterile pollen that lacked visible oil bodies , a phenotype similar to that of gpt1 or wrky2 wrky34 . The potential regulation of DGAT1 and PDAT1 expression by MPK3/MPK6-WRKY2/WRKY34 pathway remains to be examined . Expression profiling revealed dynamic changes of gene expression during pollen development [61–63] . Genetic screens have also uncovered a large number of genes encoding transcription factors , receptor-like kinases , and putative peptide ligands involved in various aspects of anther/pollen development ( reviewed in [64 , 65] ) . These findings suggest possible signaling pathway ( s ) from the sensing of extracellular ligands by cell-surface receptors , to the activation of transcription activators/suppressors , to the gene expression reprogramming during pollen development . MAPK cascades are key signaling modules downstream of receptors in plant growth and development [31] . Besides regulation at transcriptional level , WRKY34 is also regulated at the post-translational level , and is phosphorylated by MPK3/MPK6 at the late BCP stage and early TCP stages , and becomes dephosphorylated at the late TCP stage . In addition , genetic analysis demonstrated that the phosphorylation of WRKY34 is important for its biological function in pollen development [35] . It is speculated that WRKY2 is likely subjected to the same post-translational regulation by MPK3/MPK6 based on 1 ) high homology between WRKY2 and WRKY34 , 2 ) conserved phosphorylation sites , and 3 ) functional redundancy with WRKY34 . However , direct experimental evidence is still lacking . Based on our understanding of the regulation of WRKY33 by MPK3/MPK6 [34] and the high homology of WRKY2/WRKY34 to WRKY33 , we speculated that phosphorylation of WRKY2/WRKY34 also changes the transactivation activity of WRKY2/WRKY34 [31] . This is consistent with the fact that the MPK3/MPK6-phosphorylation sites of WRKY2/WRKY34 are within their transactivation domains . The spatiotemporal phosphorylation of WRKY34 and the accumulation of WRKY2/WRKY34 protein in the vegetative nucleus of BCP stage pollen are consistent with the activation of GPT1 expression in the vegetative cell at the late stages of pollen development . Furthermore , fluorescent signal from the fully functional GPT1-eYFP fusion became stronger when MPK3 and MPK6 were activated in the gain-of-function system ( Fig 8A ) , and weaker when MKK4 and MKK5 were mutated ( Fig 8B ) . In addition , lipid bodies in mpk3 mpk6 and mkk4 mkk5 double mutant pollen were significantly reduced ( Fig 8C–8F and S13 and S14 Figs ) . In summary , our data suggest that MKK4/MKK5-MPK3/MPK6 module functions upstream of WRKY2/WRKY34 in regulating the spatiotemporal expression of plastid-localized GPT1 , an important transporter that translocates Glc6P into pollen plastids for storage lipid biosynthesis during Arabidopsis pollen development . Loss of components in this pathway will reduce the accumulation of storage compounds during pollen maturation process , which negatively impacts pollen viability , pollen germination , and pollen transmission in plant sexual reproduction . The roles of plastids in heterotrophic cells such as pollen grains are less well understood in comparison to their counterpart , chloroplasts , in photosynthetic cells . Demonstration of an important role of plastidic GPT1 in storage lipid body biogenesis under the control of a MAPK-WRKY signaling pathway highlights the regulation of metabolic activities in plastids by a cytoplasmic/nuclear signaling pathway . The upstream ligand ( s ) and receptor ( s ) that activate MPK3/MPK6 are unclear at present , and further research is needed to define the whole signaling pathway . Starch granule and lipid body accumulation during pollen development are critical to pollen functions including pollen germination , pollen tube growth , and successful fertilization . In crop production , reduced yield under environmental stresses is frequently associated with the reduction of storage starch/lipid accumulation in pollen [10 , 11] . MPK3/MPK6 cascade is involved in plant response to almost all stresses from both biotic and abiotic sources [32 , 33 , 46] . As a result , this MAPK cascade may also function as a key integration point where environment factors impinge on the program of pollen development and fitness .
Arabidopsis thaliana mutant and transgenic plants related to MKK4/MKK5 , MPK3/MPK6 , and WRKY2/WRKY34 were all in Col-0 ecotype background . Mutant and transgenic lines related to gpt1+/- were in Ws-2 ecotype . Wild-type plants of Col-0 or Ws-2 ecotype were used as controls depending on the mutants or transgenic plants with which they were compared . Steroid-inducible promoter-driven tobacco MEK2DD transgenic Arabidopsis line ( DD ) [43] , chemical genetically rescued mpk3 mpk6 double mutant ( MPK6SR ) [47] , and wrky2 wrky34 double mutant [35] were described previously . Heterozygous gpt1+/- mutant in Ws-2 background [36] was kindly supplied by Dr . Anja Schneider ( Department of Biology I , Ludwig-Maximilian-University ) . Tilling mkk4 and mkk5 single mutants [66] were kindly provided by Dr . Wolfgang Lukowitz ( Virginia Tech ) . Double mkk4 mkk5 mutant was generated by crossing after removing the er105 mutant allele . Arabidopsis seeds were surface sterilized . After being imbibed at 4 oC for 3 days , the seeds were plated on half-strength Murashige and Skoog medium with 0 . 45% Phytagar . Plates were incubated in a tissue culture chamber at 22 oC under continuous light ( 70 μE m-2 s-1 ) for 7 days . Seedlings were then transplanted to soil and grown in a growth chamber with a 14-h-light/10-h-dark cycle . Dexamethasone ( DEX ) and 4-amino-1-tert-butyl-3- ( 1’-naphthyl ) pyrazolo [3 , 4-d]pyrimidine ( NA-PP1 ) were used at final concentrations of 30 μM and 5 μM , respectively . DEX and NA-PP1 stock solutions were prepared in ethanol and DMSO , respectively . Equal volumes of ethanol or DMSO were used as negative controls . For observation of the effect of DEX on the GPT1-eYFP expression in the pollen grains of PGPT1:GPT1-eYFP DD plants , the inflorescences were sprayed with DEX solution or solvent negative control . After 36 hours , the microspores at the late uninucleate stage were isolated and observed . Application of NA-PP1 inhibitor effectively inhibits the activity of chemical-sensitized MPK6YG , giving rise to the activity null double mutant of MPK3 and MPK6 . To determine the pollen development and function of pollen grains from mpk3 mpk6 double mutant plants , we submerged the inflorescences of MPK6SR plants in NA-PP1 solution ( 5 μM ) for 10 seconds , and this treatment was repeated every 12 hours . Five days later , the mature pollen was collected , stained with Nile blue A and observed . To generate the GPT1 promoter-driven GPT1-eYFP construct ( PGPT1:GPT1-eYFP ) , we amplified the GPT1 genomic DNA by using GPT1-FP ( 5’-AAATGCACATGCTGATGCTATG-3’ ) and GPT1-BP ( 5’-CTGGTCAGTACGTTTCCAACAA-3’ ) primer pair . The PCR fragment was cloned into the pBlueScript II KS vector to generate pBS-PGPT1:GPT1 construct . A Sma I site was added in front of the stop code by PCR amplification of pBS-GPT1 construct using GPT1-Sma I-FP ( 5’-GGGTGATGCGAAAGACATAAGAGTGTA-3’ ) and GPT1-Sma I-BP ( 5’-GGGGAGCTTTGCCTGCAAAACAC-3’ ) primer pair . The DNA was end phosphorylated and ligated to generate pBS-GPT1-Sma I construct . The eYFP fragment was then inserted into the Sma I-digested pBS-PGPT1:GPT1-Sma I construct to generate pBS-PGPT1:GPT1-eYFP . GPT1 native promoter-driven GPT1-eYFP fragment was then cloned into pCambia3300 binary vector using Hind III and Bam HI sites to generate pCambia3300-PGPT1:GPT1-eYFP . To overexpress GPT1-eYFP protein in pollen specifically , we use a strong pollen-specific promoter , LAT52 [41] . We first introduced a Sma I site before the start code by PCR-amplifying pCambia3300-PGPT1:GPT1-eYFP without the GPT1 promoter using GPT1-eYFP-Sma I-FP ( 5’-GGGATGGTTTTATCGGTGAAGCAAAC-3’ ) and GPT1-eYFP-Sma I-BP ( 5’-GGGGGATCCACTAGTTCTA-3’ ) primer pair . LAT52 promoter fragment was then inserted into the Sma I-digested pCambia3300-GPT1-eYFP construct to generate pCambia3300-LAT52:GPT1-eYFP . To mutate all four W-boxes in the GPT1 promoter , we divided the pCambia3300-PGPT1:GPT1-eYFP construct into four fragments at the sites of W-boxes and amplified each fragment separately using primers with mutated W-boxes sequence . GBclonart Seamless Assembly Kit ( Genebank Biosciences Inc . Suzhou , China ) was used to assemble the four fragments into the vector to generate pCambia3300-PGPT1-mW:GPT1-eYFP . All the binary vectors were transformed into Agrobacterium strain GV3101 . Arabidopsis transformation was performed by the floral dip procedure [67] , and transformants were identified by screening for BASTA resistance . PGPT1:GPT1-eYFP+/- gpt1 plants were obtained by crossing PGPT1:GPT1-eYFP gpt1 plants with gpt1+/- heterozygous mutant plants . F2 progenies from PGPT1:GPT1-eYFP+/- gpt1 F1 plants with either PGPT1:GPT1-eYFP+/- gpt1 or PGPT1:GPT1-eYFP gpt1 genotype were used in our experiments . Fluorescence microscope was performed with a Nikon Eclipse 80i microscope equipped with a Nikon Intensilight C-HGFI and fluorescence filter sets . Fluorescence signal was recorded using a TRITC ( EX 540/25; DM 565; BA 605/55 ) filter set for propidium iodide ( PI ) , a FITC ( EX 465–495; DM 505; BA 515–555 ) filter set for eYFP , and a DAPI ( EX 340–380; DM 400; BA 435–485 ) filter set for DAPI . Images were captured utilizing the Nikon Digital Camera DS-Fi1c and imaged with NIS Elements 4 . 1 . Pollen viability was examined by staining pollen grains with 2 μg/ml PI [38] . Lipids in pollen grains was stained with 20 μg/ml Nile blue A [68] . For PI and Nile blue A staining , pollen grains were collected from the floral buds at the +1 stage as previously described [35] . DAPI was used to stain vegetative and generative/sperm nuclei and to determine the pollen development stage [35] . Floral buds at each stage were carefully dissected under stereoscope . Anthers were isolated and transferred to a drop of DAPI solution . A fine needle was used to gently break the anthers , and a cover slip was then used to carefully squeeze the anthers to release the pollen . Pollen germination assays were performed as described previously with slight modification [69 , 70] . The basic medium was composed of 1 mM KCl , 10 mM CaCl2 , 0 . 8 mM MgSO4 , 1 . 5 mM boric acid , 5 mM MES , 10 μm D-myo-inositol , 18% sucrose , 1 . 5% ( w/v ) low-melting agar , and the pH was adjusted to 5 . 8 with KOH . BODIPY 505/515 ( 4 , 4-difluro-1 , 3 , 5 , 7-tetramethyl-4-bora-3a , 4-adiaza-s-indacene; Invitrogen Molecular Probes , USA ) was dissolved in anhydrous dimethyl sulfoxide ( DMSO ) as a stock solution at a concentration of 1 . 0 mg/mL . For Arabidopsis pollen staining , a final concentration of 1 . 0 μg/mL was used . Lipid droplets in stained pollen were observed using a Nikon Eclipse 80i microscope or a confocal microscope system ( Carl Zeiss upright LSM 710 NLO ) . To quantify the fluorescence intensities of BODIPY 505/515 stained pollen , we first converted the images to grey scale images . The intensity of each pollen grain was then quantified using ImageJ . Anthers with pollen at bicellular pollen ( BCP ) or tricellular pollen ( TCP ) stage were detached and submerged in 0 . 3 M mannitol solution . A fine needle was used to gently break the anthers to release the pollen , and the pollen grains ( suspended in the mannitol solution ) were transferred into a 1 . 5 mL tube using a glass capillary tube . Pollen grains from 10 flowers of similar stages collected from 3 plants were pooled together in each of the three repeats . After centrifugation at 450 ×g for 5 min at 4°C , the pollen pellets were washed twice with 0 . 3 M mannitol solution . Total RNAs were isolated using TRIzol reagent . After DNase treatment , 0 . 5 μg of total RNA was reversely transcribed , and quantitative PCR analysis was performed using an Eppendorf real-time PCR machine . Relative levels of each transcript were calculated after being normalized to the EF1α control . Protein extraction was performed as previously described with modifications [34] . Anthers at mature pollen ( MP ) stage but before dehiscence were collected from the same plant . Anthers were ground in liquid nitrogen and extracted in 20 μl 1 . 5 × SDS loading buffer without bromophenol blue dye . Concentrations of protein samples were determined by bicinchoninic acid ( BCA ) assay suing BSA as standard . Due to the high similarity with GFP [71] , eYFP fusion proteins can be detected with a rabbit anti-GFP polyclonal antibody ( Abmart ) . Immunoblot detection of GPT1-eYFP was performed as previously described [72] . Images of mature pollen were taken . Length of pollen grains was measured using ImageJ after the scale tool was set to establish a 100 μm reference on the images . To quantify the fluorescence intensities of pollen grains with GPT1-eYFP transgene , we first converted the fluorescence images to grey scale images , and then the intensity of each pollen grain was quantified using ImageJ . All experiments were repeated independently at least three times , and representative results are shown . For the purpose of calculating percentages of pollen grains with a particular phenotype , at least 80 pollen grains ( indicated in the figure legends ) were analyzed in each of the repeats in order to obtain an average of the percentages with standard deviation . One-way ANOVA Tukey’s test was used for statistical analysis . One and two asterisks above the columns indicate differences that are statistically significant ( P ≤ 0 . 05 ) and very significant ( P ≤ 0 . 01 ) , respectively . | Plastids are important sites of plant metabolism including fatty acid and starch biosynthesis . At present , how the activities in the plastids are coordinated with those in the cytoplasm and the signaling pathway ( s ) involved are largely unknown . Previously , we demonstrated that WRKY2 and WRKY34 transcription factors play an essential role in pollen development downstream of mitogen-activated protein kinase 3 ( MPK3 ) and MPK6 in Arabidopsis . Here , we report that GLUCOSE-6-PHOSPHATE/PHOSPHATE TRANSLOCATOR 1 ( GPT1 ) is a key target gene of WRKY2/WRKY34 . GPT1 is localized on the membrane of plastids and transports glucose-6-phosphate ( Glc6P ) into plastids for starch and/or fatty acid biosynthesis depending on the plant species . Genetic analyses demonstrated that WRKY2/WRKY34 and their upstream MPK3/MPK6 are involved in regulating GPT1 expression , therefore , the accumulation of lipid bodies in mature pollen , which is critical to pollen viability , pollen germination , pollen tube growth , and male transmission in Arabidopsis . This study revealed a cytoplasmic/nuclear signaling pathway capable of coordinating the metabolic activities in plastids . High-level expression of GPT1 at late stages of pollen development drives Glc6P from cytosol into plastids , where Glc6P is used for fatty acid biosynthesis , an important step of lipid body biogenesis . The accumulation of lipid bodies during pollen maturation is essential to pollen fitness and successful reproduction . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"fluorescence",
"imaging",
"plant",
"anatomy",
"plastids",
"gene",
"regulation",
"plant",
"cell",
"biology",
"regulatory",
"proteins",
"brassica",
"dna-binding",
"proteins",
"pollen",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"plants",
"arabidopsis",
"thaliana",
"research",
"and",
"analysis",
"methods",
"lipids",
"imaging",
"techniques",
"animal",
"studies",
"proteins",
"gene",
"expression",
"biochemistry",
"eukaryota",
"plant",
"and",
"algal",
"models",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"fatty",
"acids",
"organisms"
] | 2018 | Regulation of pollen lipid body biogenesis by MAP kinases and downstream WRKY transcription factors in Arabidopsis |
To gain insight into female-to-male HIV sexual transmission and how male circumcision protects against this mode of transmission , we visualized HIV-1 interactions with foreskin and penile tissues in ex vivo tissue culture and in vivo rhesus macaque models utilizing epifluorescent microscopy . 12 foreskin and 14 cadaveric penile specimens were cultured with R5-tropic photoactivatable ( PA ) -GFP HIV-1 for 4 or 24 hours . Tissue cryosections were immunofluorescently imaged for epithelial and immune cell markers . Images were analyzed for total virions , proportion of penetrators , depth of virion penetration , as well as immune cell counts and depths in the tissue . We visualized individual PA virions breaching penile epithelial surfaces in the explant and macaque model . Using kernel density estimated probabilities of localizing a virion or immune cell at certain tissue depths revealed that interactions between virions and cells were more likely to occur in the inner foreskin or glans penis ( from local or cadaveric donors , respectively ) . Using statistical models to account for repeated measures and zero-inflated datasets , we found no difference in total virions visualized at 4 hours between inner and outer foreskins from local donors . At 24 hours , there were more virions in inner as compared to outer foreskin ( 0 . 0495 +/− 0 . 0154 and 0 . 0171 +/− 0 . 0038 virions/image , p = 0 . 001 ) . In the cadaveric specimens , we observed more virions in inner foreskin ( 0 . 0507 +/− 0 . 0079 virions/image ) than glans tissue ( 0 . 0167 +/− 0 . 0033 virions/image , p<0 . 001 ) , but a greater proportion was seen penetrating uncircumcised glans tissue ( 0 . 0458 +/− 0 . 0188 vs . 0 . 0151 +/− 0 . 0100 virions/image , p = 0 . 099 ) and to significantly greater mean depths ( 29 . 162 +/− 3 . 908 vs . 12 . 466 +/− 2 . 985 μm ) . Our in vivo macaque model confirmed that virions can breach penile squamous epithelia in a living model . In summary , these results suggest that the inner foreskin and glans epithelia may be important sites for HIV transmission in uncircumcised men .
The World Health Organization estimates that over 35 million people world-wide are currently infected with the human immunodeficiency virus ( HIV ) [1] . The majority of these infections are acquired through heterosexual transmission events , with female-to-male HIV transmission rates approaching that of male-to-female in some areas[2] . Male circumcision has been shown to effectively reduce the risk of HIV acquisition in men by 50–60% in three large African cohorts[3–5] . This protective effect appears to be long-lasting and extends to other sexually transmitted infections ( STIs ) such as human papillomavirus and herpes simplex virus-2[6] . In contrast , the benefits of male circumcision have not been so clearly defined for men who have sex with men[7 , 8] . Our lack of a scientific model for how HIV infects the man through the penis hinders our ability to explain how male circumcision protects against HIV infection , as well as to interpret these clinical disparities . In this study , we sought to explore potential sites of HIV transmission through the penis using tissue explants from adult donors and a living rhesus macaque model . In all primates , the penis is naturally covered with a prepuce or foreskin . The foreskin is composed of an “inner” aspect that is adjacent to the glans epithelia in the flaccid state . The inner foreskin attaches to the penis at the coronal sulcus . In the erect state , the foreskin retracts to expose the inner surface to the environment . The “outer” foreskin is continuous with the penile shaft and remains exposed to the environment in both flaccid and erect states . An initial hypothesis for HIV entry into the uncircumcised penis centered on differences in foreskin keratin layers ( or stratum corneum , SC ) [9 , 10] . A thin inner foreskin SC would allow the virus to more easily penetrate the skin and encounter a HIV susceptible immune cell ( e . g . , Langerhans cells , CD4+ T-cell lymphocytes or macrophages ) . However , quantitative studies using foreskins from donors in China , the USA , and Uganda found no biologically or statistically significant difference in SC thickness between foreskin areas[11–13] . Other studies have demonstrated that the surface area of the foreskin correlates with HIV incidence rates , suggesting that simple removal of this target cell-rich tissue would be sufficient to lower a man’s risk of HIV sexual acquisition[14 , 15] . This risk may also be influenced by factors that have been shown to differ between circumcised and uncircumcised men , such as hygiene practices , latent STIs , and bacterial colonizers[16–18] . Latent STIs may also alter target cell populations in the tissue by recruiting cells to the surface or activating them , and thus enhance HIV susceptibility[19 , 20] . Finally , the increased HIV incidence rates in vaccinated , uncircumcised male subjects in the Merck HIV-1 STEP trial support the idea that the vaccine elicited a mucosal response and subsequently enhanced HIV transmission in the male genital tract[21] . These studies collectively raise questions on how penile tissues change after circumcision and how these changes contribute to HIV transmission through the penis . In this report , we investigated how HIV-1 interacts with male penile stratified squamous epithelia by visualizing and characterizing the earliest interactions of photoactivatable GFP-labeled HIV-1 with human and rhesus macaque penile tissues . We also surveyed tissue-resident immune cells in these tissues and found potentially important differences between foreskin , glans , shaft , and urethral meatus tissues . This information will help guide future studies on how male circumcision affects HIV sexual transmission through the penis .
Fluorescently labeled CCR5-tropic ( R5-tropic ) HIV-1 was made by co-transfecting 293T cells with an HIV-1 provirus and photoactivatable GFP-Vpr constructs ( PA GFP HIV ) [22–24] . Foreskin tissues were obtained from local consenting adult donors and cultured with PA GFP HIV for 4 and 24 hours ( n = 10 and 12 , respectively ) . 1612 images of tissue cryosections were obtained using deconvolution epifluorescent microscopy . A subtraction method was used to determine true PA GFP HIV from tissue background autofluorescence , as previously described[24] . Many images captured did not contain virions ( ∼40% ) ; in those that did , we counted 15626 individual virions , the majority of which were found on the epithelial surface or in the stratum corneum ( SC ) ( Fig . 1A and 1B ) . Foreskin specimens inoculated with PA GFP HIV and a fluorescent fluid phase marker ( bovine serum albumin , BSA ) demonstrated that the virus diffused into the SC in a similar manner as BSA ( Fig . 1C ) . That is , there was heterogeneous distribution of both BSA and virions into the SC , with some areas allowing for shallower diffusion and other areas allowing for deeper diffusion . These patterns did not demonstrably differ between the inner and outer foreskin . On average , 1 per 100 virions visualized were seen past the SC , which we termed , “penetrators” ( Fig . 1D ) . The range of penetration depths seen in foreskin tissue was 0–96 . 69 μm ( S1A Fig . ) . Using wheat germ agglutinin to highlight epithelial cell surfaces , we determined that >80% of penetrators were found between rather than inside a cell ( inset , Fig . 1D ) . We also performed immunofluorescence imaging for tissue-resident immune cells by using foreskin tissues that had not been exposed to HIV-1 in culture and were immediately frozen upon arrival to the lab ( Fig . 2A ) . We focused on cell phenotypes likely important in HIV sexual transmission: Langerhans cells ( LCs ) and CD4+ T-cell lymphocytes and macrophages[25–27] . Probability distributions of depths from the epithelial surface for both virions and cells were then used to estimate the likelihood that a penetrator would encounter an immune cell in the inner and outer foreskin . Given our finite dataset , we graphed normalized distributions using kernel density estimations ( KDE ) and then calculated the overlap of virus and cells in each tissue type ( Fig . 2B ) . From this initial analysis , we found that the distribution of CD4+ cells in the tissue differed between the inner and outer foreskin , resulting in greater overlap of penetrating virions at 4 and 24 hours ( S2 Fig . ) . In fact , there was a >2-fold greater overlap between penetrators and CD4+ cells in the inner as compared to the outer foreskin at 24 hours ( S2G and S2H Fig . ) . We generally observed LCs abundantly in the epidermis , but no differences were seen in cell counts or depths between the inner and outer foreskin . To evaluate if these sentinel LCs might change in response to viral particles , we also analyzed their counts and depths after 24 hours of exposure to PA GFP HIV in a randomly selected subset of donors ( n = 4 ) . We found no difference in the overlap of virions and cells between the inner and outer foreskin in this subset at this time point ( overlap percentages = 21 . 3 and 21 . 0 , respectively , S2I Fig . ) . Since the KDE distributions did not reflect varying virus stock concentrations used in each donor sample , number of obtainable images per sample , and repeated measures within samples , we developed models to make statistical comparisons between tissue types and time points . Our first model was constructed to evaluate total counts of virions per image , adjusted for virus stock concentration used for each tissue sample . Initial analysis took into account all images taken , including those in which no virions were seen ( n = 1612 images ) . We found no difference between the inner and outer foreskin at 4 hours ( mean 0 . 0505 +/− 0 . 0116 and 0 . 0542 +/− 0 . 0167 virions/image , respectively , Fig . 1E ) . This changed at the later time point , with more virions remaining in the inner as compared to outer foreskin ( 0 . 0495 +/− 0 . 0154 and 0 . 0171 +/− 0 . 0038 virions/image , p = 0 . 001 ) . Correspondingly , a significant decrease in total virions from 4 to 24 hours was only seen in the outer foreskin ( 0 . 0542 +/− 0 . 0167 to 0 . 0171 +/− 0 . 0038 virions/image , p<0 . 001 ) . Our second model evaluated proportion of penetrators , adjusted for virus stock concentration . For this parameter , we evaluated only the subset of images in which at least one virion was seen , since no proportion could be calculated from an image where no virions were visualized ( n = 964 images ) . We found no significant differences in the proportion of penetrators across tissue types or time points ( Fig . 1F ) . We re-analyzed the first parameter with this subset of images and confirmed our findings from the initial analysis ( i . e . , more virions seen at 24 hours in the inner versus outer foreskin ) ( S1B Fig . ) . Our third model evaluated mean depths of penetration into the tissue and did not show any significant differences between the inner and outer foreskin ( Fig . 1G ) . However , we observed significantly greater virion penetration depths at 24 hours as compared to 4 hours in both tissue types . Using similar statistical models , we found more CD4+ cells in the inner as compared to the outer foreskin at baseline ( mean 3 . 583 +/− 1 . 613 and 1 . 185 +/− 0 . 526 cells/image , respectively , p = 0 . 001 ) , but no differences when comparing estimated mean depths between the two tissue types ( Fig . 2C and 2D ) . There was no difference between the inner and outer foreskin in regards to total LCs or their depths from the epithelial surface at baseline ( inner: 3 . 776+/−0 . 469 cells/image and 84 . 876 +/− 8 . 575 μm; outer: 4 . 060 +/− 0 . 689 cells/image and 89 . 240 +/− 11 . 869 μm , respectively ) . After 24 hours of virus exposure , we found a slight increase from baseline in the mean number of LCs in the inner foreskin ( paired donors in the subset selected ) ( 2 . 77 +/− 0 . 50 to 3 . 60 +/− 0 . 42 cells/image , p = 0 . 047 , Fig . 2E ) but the change in mean depths was not significant ( 89 . 70 +/− 3 . 26 to 74 . 31 +/− 9 . 63 μm , p = 0 . 129 , Fig . 2F ) . In the outer foreskin , we found no significant changes in LC counts or depths after 24 hours of virus exposure . Comparing inner to outer foreskin , there was no significant difference in LC counts at either time point . Although outer foreskin LCs were closer to the surface as compared to those in the inner foreskin in this donor subset , the relative ratios did not significantly change after virus exposure ( Fig . 2F ) . Beyond foreskin tissue , we sought to determine if differences existed between circumcised and uncircumcised penile tissues . We obtained 14 cadaveric penile specimens ( 7 uncircumcised and 7 circumcised ) through tissue donation organizations . Tissue samples were cultured ex vivo with R5-tropic PA GFP HIV for 4 hours ( we excluded longer incubation times due to potential tissue degradation from prolonged post-mortem tissue shipping ) or immediately snap frozen as negative controls for immune cell analysis as described above . A total of 600 images were evaluated in the virion analysis ( with 65% containing visible virions ) and 352 images were used in the immune cell analysis . Similar to what was seen in the foreskin tissues described above , most visualized virions were on the surface , though penetrators could occasionally be seen between epithelial cells ( average 3 . 4 per 100 virions ) ( Fig . 3A and S3 Fig . ) . The same parameters described above were assessed in the penile explants . We observed differences in potential virus-cell interactions between uncircumcised and circumcised glans tissues using KDE plots of penetrators and immune cells ( particularly with CD4+ cells ) ( Fig . 3B ) . However , calculated overlap percentages mainly showed differences between tissue and cell types , not circumcision status ( S4 Fig . ) . Using the statistical models described above , we compared estimated means of virions or cells between tissue types ( glans , shaft , +/− inner and outer foreskin ) or circumcision status . Again , because of potential tissue degradation after prolonged shipping times , we only included data from the 4 hour time point in this analysis . Data from the virion analysis are presented in Fig . 3C as ratios for ease of comparison across each variable ( ratios >1 correlate with significant interactions ) . In the uncircumcised donor tissues , we found more virions/image in the inner foreskin than glans or shaft tissue ( inner = 0 . 0507 +/− 0 . 0079 virions/image , glans = 0 . 0167 +/− 0 . 0033 virions/image , p<0 . 001 , shaft = 0 . 0205 +/− 0 . 0065 p = 0 . 036 ) . No difference was seen between inner and outer foreskins at this early time point , as was noted in the foreskin analysis from local donors . Re-analyzing the virion count with the subset of images that contained at least one virion confirmed this finding ( n = 368 images , S3D Fig . ) . A larger proportion of penetrators was seen in the uncircumcised glans as compared to inner and outer foreskin ( glans = 0 . 0458 +/− 0 . 0188 virions/image , inner = 0 . 0151 +/− 0 . 0100 virions/image , p = 0 . 099 , and outer = 0 . 0048 +/− 0 . 0019 virions/image , p<0 . 001 ) ( Fig . 3D ) . A significantly greater mean penetration depth was also seen in the uncircumcised glans tissue ( 29 . 162 +/− 3 . 908 μm ) , as compared to that in inner and outer foreskin tissues ( 12 . 466 +/− 2 . 985 , p = 0 . 002 and 18 . 253 +/− 2 . 481 μm , p = 0 . 014 , respectively ) . In this virion analysis , we observed no differences between the tissue types based on circumcision status for any of the parameters measured . For the immune cell analysis , we found that uncircumcised glans epithelia contained marginally more CD4+ cells than shaft epithelia ( 1 . 333 +/ 0 . 387 vs . 0 . 452 +/− 0 . 323 cells/image , p = 0 . 05 , Fig . 3F ) and were closer to the epithelial surface though this was not statistically significant ( 64 . 892 +/− 12 . 584 vs . 84 . 883 +/− 5 . 587 μm , p = 0 . 158 , Fig . 3G ) . We also observed CD4+ cells closer to the surface of shaft tissues from uncircumcised as compared to circumcised donors ( 84 . 883 +/− 5 . 587 vs . 114 . 500 +/− 8 . 437 μm , respectively , p = 0 . 003 ) and in the glans as compared to shaft tissue of circumcised donors ( 66 . 754 +/− 13 . 465 vs . 114 . 500 +/− 8 . 437 μm , respectively , p = 0 . 015 ) . We did not observe significant differences in LCs counts between tissue types or circumcision status , but found that they were closest to the surface of the glans as compared to shaft tissue of circumcised donors ( 44 . 234 +/− 2 . 258 vs . 58 . 110 +/− 4 . 571 μm , respectively , p<0 . 001 ) . We also explored the urethral meatus ( opening to the urethra , UM ) as a potential site of HIV transmission . This area is continuous with the glans and is composed of non-keratinized stratified squamous epithelia[28] . We analyzed samples from 4 cadaveric donors ( 2 circumcised and 2 uncircumcised donors , but grouped them together as circumcision status should not affect this area ) in which we could clearly delineate UM from the urethra and glans . The tissues were analyzed using the same methods as described above , except that we immunostained for CD68+ macrophages rather than LCs , as LCs are not found in the urethra . This subset included 48 images , with estimated means of 0 . 0319 +/− 0 . 0099 virions/image ( adjusted for virus stock concentration , comparisons shown in S5B Fig . ) and 0 . 0284 +/− 0 . 0229 penetrators/image; these values were not significantly different from those of other tissues analyzed . The mean penetration depth was significantly less than that observed in other tissues ( 7 . 583 +/− 1 . 729 μm , p≤0 . 001 ) except inner foreskin ( p = 0 . 133 ) , and the calculated overlap percentages from KDE plots of penetrators and immune cells was smaller than that observed in other tissues ( S5C Fig . ) . To determine if our observations may have been influenced by use of devitalized explant tissues , we sought an in vivo model to examine HIV interactions with intact penile epithelia[29] . To this end , we exposed 7 mature Indian male rhesus macaques ( macaca mulatta ) to PA GFP HIV using a “dunk” method as described in the methods section . Since these experiments were only intended to observe early interactions between virus and epithelium and to compare these observations to our ex vivo studies , we used the PA GFP HIV produced as described above . However , the animals were only exposed to viral supernatant for ∼15 minutes while anesthetized and allowed to resume normal activity for 4 hours prior to tissue collection . From these experiments , we obtained 1104 epifluorescent images of macaque penile tissues , which included 1552 individual visualized virions . We visualized PA GFP HIV interacting with macaque penile epithelia in vivo in a similar manner as with the ex vivo penile explant model ( Fig . 4A and 4B ) . That is , the majority of viral particles remained on the surface or in the SC with a proportion able to penetrate into the epithelium . We used the statistical models described above to analyze the virions across tissue types and found a higher number of virions/image ( 0 . 01326 +/− 0 . 01247 ) but lower proportion of penetrators ( 0 . 02459 +/− 0 . 01015 ) in the outer foreskin as compared to other tissues ( Table 1 ) . Penetrators also reached greater depths in the outer foreskin ( 20 . 9262 +/− 7 . 1562 μm ) , significantly more so than in the glans tissues ( p = 0 . 038 ) . In the shaft tissues , we also observed high proportions of penetrators going to greater depths in the tissue ( 0 . 3803 +/− 0 . 1688 virions/image and 18 . 4040 +/− 6 . 2753 μm ) , but these observations may be attributed to specific macaque penile characteristics as described in the Discussion section below .
While male circumcision has been shown to reduce HIV acquisition rates in men , we do not yet fully understand how this protection works , nor how the virus enters the male genital tract[3–5] . Plausible theories include the removal of a large surface area of tissue containing HIV-susceptible cells ( the foreskin ) , but circumcised men still acquire HIV and it is unknown how penile transmission occurs after male circumcision[22] . To explore potential sites of HIV transmission across penile surfaces , we utilized epifluorescent microscopy to study PA GFP-labeled HIV-1 interactions with human tissue explants as well as in an in vivo rhesus macaque model[22 , 23] . In all penile tissues studied in both the human explant and macaque model , we observed most virions in the epithelial SC , even after 24 hours of exposure in culture . Co-inoculation of foreskin explants with HIV-1 and a fluorescent fluid phase marker ( BSA ) demonstrated that virions diffuse into the SC in a heterogenous pattern that is similar to the fluid phase marker . As no tissue washing occurred prior to fixation , we believe that this observation accurately reflects the simple diffusion of virions and BSA in culture . Furthermore , we observed similar diffusion patterns in the female macaque model upon exposure to BSA in vivo[24] . In this study , we found significantly more HIV-1 viral particles remaining in the inner foreskin ( predominantly in the SC ) after 24 hours as compared to the outer foreskin . More viral particles were also seen within inner foreskin tissue as compared to other penile surfaces . We and others have demonstrated that inner foreskin SC thicknesses do not significantly differ from that of other foreskin areas and propose instead that a more physiological characteristic of the foreskin SC allows virions to perpetuate over time[13] . The persistence of virus in the inner foreskin may lead to infection in the uncircumcised male via two mechanisms: the first is that infectious viral particles are introduced into the urethral meatus after sexual intercourse , as the foreskin has been observed to cover the UM in the flaccid state in a proportion of uncircumcised men[30 , 31] . While our limited dataset of UM tissues did not indicate that this was a particularly vulnerable site , it is possible that larger datasets or analysis of other areas of the distal/anterior urethra may yield different results . ( Although Ganor et al . have suggested that the “middle” urethra may be a site for HIV transmission , it is unclear how the virus would reach this area during or after sexual intercourse[32] . ) The second possibility is that retained viral particles enhance the immune response in the inner foreskin and adjacent glans ( preputial space ) , which eventually leads to virion uptake by a superficial potential target cell in either tissue . The results of the Merck STEP study , where uncircumcised vaccine recipients exhibited the highest HIV acquisition rates , support such a dynamic preputial environment where immunologic changes in these tissues post-vaccination may have enhanced HIV transmission . Further supporting the existence of a dynamic preputial space are our observations that virions are able to penetrate the uncircumcised glans and inner foreskin epithelia to reach depths where LCs and CD4+ cells reside . In fact , we demonstrated that penetrating virions could be seen reaching depths in several tissue types where resident immune cells were also found , particularly after 24 hours of culture . While it has been shown that LCs can be transiently infected by HIV-1 and transfer virions to CD4+ T-cells via synapses , we observed many penetrators at depths were CD4+ lymphocytes and macrophages could also be found[33 , 34] . The overlap of penetrators and CD4+ cells was greater in the inner as compared to outer foreskin , and in the uncircumcised glans as compared to other cadaveric penile tissues . Although we were unable to calculate the statistical significance of these distribution overlaps , they visually suggest that the inner foreskin and uncircumcised glans epithelia may be key sites in HIV transmission . These two surfaces form the preputial space in the uncircumcised man and the persistence of viral particles in the inner foreskin SC may lead to a greater likelihood of virion-target cell interactions within either tissue . We therefore propose that male circumcision protects against HIV transmission by not only removing the foreskin , but also by changing the remaining glans epithelium . Supporting this model is the observation that more virions were seen penetrating the uncircumcised glans epithelia and to greater depths than in the inner foreskin tissue . To our knowledge , this is the first study comparing circumcised and uncircumcised penile tissues , and future studies specifically evaluating glans epithelia using in vivo models or freshly obtained tissues will further investigate the role of this site in HIV transmission . Viral transmission across the glans epithelia might also explain how circumcised men remain at risk of HIV acquisition through the penis . In deeper strata , dense intercellular junctions prevent the interstitial movement of foreign agents and accordingly , we observed only a small proportion of virions penetrating to these depths in all tissues evaluated[35] . The proportion of viral penetrators did not differ between the inner and outer foreskin from local donors at either early or late time points , nor in the cadaveric specimens at the only observed early time point . However , at the late time point , the absolute number of penetrators was higher in the inner foreskin of local donors given the greater total number of virions visualized there . Hypothetical mechanisms through which penetrators reached deeper epithelial strata include disruption of intercellular junctions , travel along or with LC processes , and/or epithelial cell trancytosis ( though this has only been shown to occur through M cells in rectal epithelium ) [33 , 36] . LC processes may also disrupt tight junctions themselves as they survey the external environment , and we have previously demonstrated that foreskin LCs can migrate in/out of foreskin epithelium in response to external agents[37 , 38] . However , we did not observe significant differences in LCs between inner and outer foreskin tissue , even after 24 hours of virus exposure , to explain the observed differences . We therefore hypothesize that at the early time point , most penetrators were quickly degraded by epithelial or immune cells and differences were only seen at the later time point after a saturation point between virions and cells had been achieved . With more virions persisting on the inner foreskin after 24 hours of culture , more would penetrate and be visibly intact in the tissue . Future studies evaluating live virus movement into fresh tissues will help to elucidate these potential mechanisms . Finally , the use of cadaveric specimens allowed us to uniquely compare tissues from circumcised and uncircumcised donors as well as between more penile sites , such as the UM . Due to the nature of our tissue collection process , we could not extend the cadaveric tissue cultures to the later time point as we did with freshly obtained foreskin tissues from local donors . However , the similarity in our observations between locally-obtained and cadaveric foreskins at the early time point suggest that longer term explant studies with freshly obtained penile tissues may uncover even greater differences between penile sites or donor circumcision statuses . One caveat to using tissue explants is that observations may not reflect in vivo occurrences[29] . The rhesus macaque model , though somewhat different from humans , allowed us to verify that our observations were not an artifact of tissue explant cultures . With this model , we confirmed that virions can enter intact penile squamous epithelia , occasionally within reach of abundant LCs and CD4+ cells in the epidermis . As the macaque tissues were immediately snap-frozen in OCT , our observations likely reflect in vivo responses to virion exposure , rather than trauma from tissue excision . Of note , the data collected from the macaque experiments should be interpreted with caution due to key differences between macaque and human penile anatomy as well as experimental conditions . Firstly , the macaque outer foreskin is continuous with the abdominal skin and contains hair follicles , which traps viral particles . The foreskin also covers the entire length of the penis ( starts at the proximal penis base ) , so the preputial space includes the shaft . This may lead to more virion accumulation and penetration in the macaque outer foreskin and shaft relative to other tissue types . Secondly , the animals were only exposed to viral supernatant while anesthetized . After this time and prior to necropsy , superficial virions were likely brushed off by the animal , resulting in fewer overall numbers of virions seen in the macaque model . Despite these differences , the use of the macaque model was important in verifying observations made in the tissue culture model . Furthermore , macaque models will be important in future studies examining infection of cells within the tissue , which require longer experimental times ( days rather than hours ) to achieve successful virus-tissue encounters and productive infection of the cell . We also caution against directly comparing the results of this study to that previously published by our group utilizing the same virus identification technology to study the female reproductive tract of women and macaques[24] . Differences in methodology , such as the use of stitched panels in this study ( as seen in S1 Fig . ) and only counting penetrators seen past the SC ( as many were seen within the SC ) resulted in different counts and recorded depths of penetration . Our biostatisticians ( AF and AR ) also developed complex statistical models to include all images captured in the analysis ( including many with zero counts ) to make the comparisons reported , which was different from what had been done previously . As noted in many other studies using donor tissues , we observed substantial heterogeneity between individuals in this study . For example , three specimens from three donors processed and inoculated on the same day with the same virus stock resulted in entirely different patterns of virus association and epithelial penetration . Factors contributing to this heterogeneity may include latent STIs such as HSV-2 or HPV , race , age , sexual activity or hygiene practices , which we did not collect information on in this study . Future studies examining these potentially confounding variables along with baseline skin structural/biological characteristics may help explain some of the observed inter-individual heterogeneity . Other drawbacks to our study include the use of tissues from men undergoing elective male circumcision or cadaveric donors . However , we took several measures to optimize the use of these specimens as described in the Materials & Methods section . We also saw no evidence of tissue degradation at the microscopic level in our image analysis . In summary , we present data supporting that the inner foreskin may allow prolonged survival of infectious HIV particles in the preputial space , and that the uncircumcised glans penis may also be permissive to HIV encounters with CD4+ cells . This provides a mechanism for how male circumcision changes HIV susceptibility in a man , though further studies are needed to define how the glans tissue changes after male circumcision , as well as to demonstrate actual infection of immune cells within the tissue . Once a more complete model of HIV penile transmission is established , we may be able to devise other effective prevention strategies for HIV acquisition in men .
All work described in this study was reviewed and approved by the Institutional Review Board ( IRB ) and Animal Care and Use Committee of Northwestern University , the IRB of Rush Presbyterian Hospital , or by the ACUC at Tulane National Primate Research Center ( TNPRC , protocol 0094 ) . All human subjects provided written informed consent and all research was conducted according to the principles expressed in the Declaration of Helsinki . This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health ( NIH ) and with the recommendations of the Weatherall report; “The use of non-human primates in research . " All procedures were performed under anesthesia using ketamine hydrochloride , and all efforts were made to minimize stress , improve housing conditions , and to provide enrichment opportunities ( e . g . , objects to manipulate in cage , varied food supplements , foraging and task-oriented feeding methods , interaction with caregivers and research staff ) . Animals were euthanized by ketamine hydrochloride injection followed by barbiturate overdose in accordance with the recommendations of the panel on Euthanasia of the American Veterinary Medical Association . Green fluorescent protein linked to the N-terminus of HIV-1 Vpr was made as described by McDonald et al . [22] . To circumvent previous issues encountered with tissue auto-fluorescence , photoactivatable ( PA ) GFP , developed in the laboratories of Dr . Jennifer Lippincott-Schwartz , was incorporated to produce non-fluorescing GFP molecules that could be “turned on” by excitation and thereafter remain fluorescent[23] . Plasmids encoding PA GFP-Vpr and an HIV-1 provirus were used to co-transfect 293T cells and highly infectious viral supernatant was obtained at four hour intervals . Viral replication and infectivity was measured with HIV p24 ELISA assays and infectivity assays ( mean 526 . 77 ng/ml p24 ) . Viral stocks were stored at −80°C until ready for use . Adult human foreskin tissues ( n = 10 for 4 hour time point , n = 12 for 24 hour time point ) were obtained from consenting adult donors . Donors were identified upon presentation for elective medical male circumcision through the Departments of Urology . We did not collect medical information such as presence of latent STIs on the subjects , though any subject with gross lesions were deferred for surgery . All tissue samples were de-identified prior to arrival to the laboratory . The de-identified tissue was processed within 2 hours of removal from the donor . Foreskin explants were washed with sterile 1X PBS ( Hyclone ) , separated into inner and outer aspects , dissected into 0 . 5 x 0 . 5 x 0 . 2 cm sections , and placed individually into 24-well plates ( Becton Dickinson ) . Sections were inoculated with 500 μl of PA GFP-Vpr HIVBal or HIVR7 supernatant and incubated at 37°C . At 4 and 24 hours , explants were removed from the supernatant , snap-frozen in OCT ( Optimal Cutting Temperature , Sakura Finetek , Torrance , CA ) compound in standard-sized plastic cryomolds ( Sakura ) , and kept at −80°C for storage . Tissues were also snap-frozen in OCT and cryomolds without virus as negative controls . To evaluate tissue permeability and how virions might move into tissues , foreskin tissues were co-inoculated with PA GFP HIV-1 and fluorescently labeled bovine albumin serum ( 1 mg/ml , BSA , Sigma ) at 37°C for 4 hours . BSA was labeled by direct conjugation to an amine reactive Alexa Fluor 594 dye ( Molecular Probes , Invitrogen ) . Tissues were immediately snap-frozen in OCT after 4 hours of culture ( with no washing prior to embedment ) and stored at −80°C . Slides of cryosections were prepared as described below . Cadaveric penile specimens ( n = 14 ) were obtained from three tissue donation organizations: Life Legacy , ScienceCare , and National Disease Research Interchange ( NDRI ) . Donors from these donation banks are screened at enrollment for pre-existing infections or medical conditions . To determine tissue viability , we inoculated tissue sections with 0 . 2% dinitrofluorobenzene ( DNFB , Sigma-Aldrich ) + RPMI + 10% fetal bovine serum for 4 hours at 37°C . Tissues were examined for CD1a cell expression , as this chemical is known to induce down-regulation of CD1a by LCs[38] . From these tests , we developed strict cut-offs of 36 hours post-mortem ( donor time of death to tissue arrival in our Chicago laboratory ) to ensure tissue viability for our ex vivo assays . Specimens were washed with sterile 1X PBS , separated into glans , shaft , and if applicable , inner and outer foreskin , and further dissected into 0 . 5x0 . 5x0 . 2 cm pieces . Tissue pieces were placed into separate wells in plastic plates and inoculated as described above for foreskin specimens . Only 4 hour time points were evaluated in the analysis due to potentially significant tissue degradation at longer time points . Thin ( ∼10μm ) cryosections were placed onto 1mm glass slides ( VWR ) for immunostaining , kept frozen or immediately fixed with a PIPES-formaldehyde mix ( 0 . 1M PIPES buffer , pH 6 . 8 and 3 . 7% formaldehyde ( Polysciences ) ) and washed with cold ( 4°C ) 1X PBS . Tissues were blocked with 10% Normal Donkey Serum ( NDS ) /0 . 1% Triton X-100/0 . 01% NaN3 . To examine virus location within the tissues , sections were immunofluorescently stained with Wheat Germ Agglutinin ( WGA , Alexa Fluor 647 , Invitrogen , 1μg/mL ) and counterstained with 4 , 6-diamidino-2-phenylindole ( DAPI , 1:25000 ) . LC immunofluorescent staining was performed with anti-human OKT6 ( 1:1 ) and anti-mouse donkey Rhodamine Red X ( Jackson-ImmunoResearch , 1 . 5μg/mL ) . CD4+ immunostaining was performed with monoclonal anti-human mouse anti-CD4 antibody ( Sigma , clone Q4120 , 1:350 dilution ) fluorescently conjugated with Alexa Fluor Zenon labeling kits ( Invitrogen ) . CD68+ immunostaining was conducted with monoclonal anti-human mouse anti-CD68 antibody ( Dako , clone EBM11 , 1:200 dilution ) and anti-mouse donkey Rhodamine Red X as above . Sections were counterstained with Hoescht as described above . Fluorescent mounting medium ( DAKO , Denmark ) and coverslips ( VWR coverglass No . 1 thickness ) were placed onto slides; slides were kept at 4°C until ready for imaging analysis . All imaging was conducted with DeltaVision RT epifluorescent microscope systems ( General Electric , Issaquah , WA ) . Virions were visualized with a 100x objective lens using either a 2x2 paneled fields of view ( 5 . 76e-2 mm2 ) to include the epithelial edge and as much of the epithelium as possible in the field of view , with roughly equal number of images obtained for each tissue type ( at 4 hours , inner foreskin n = 357 , outer foreskin n = 351; at 24 hours , inner foreskin n = 484 , outer foreskin n = 420 ) . For each donor sample , at least 10 images from three separate tissue sections taken ∼20 μm apart in the tissue block were obtained for analysis . Each area was surveyed three-dimensionally ( 30 z-stacks x 0 . 5 μm spacing ) . All images were deconvolved and analyzed with SoftwoRx software ( GE ) to identify virions and target cells in the tissue . PA GFP HIV-1 particles were identified using an inverse subtraction method: a pre-photoactivation image ( accounting for background fluorescence ) is captured and pseudo-colored as green , the field of interest is photoactivated with ∼495nm light , and a second post-photoactivation image is captured and pseudo-colored as red; the green and red images are overlaid and viral particles appear red whereas tissue auto-fluorescence appears yellow from the green and red overlap . Each viral particle identified was confirmed as such using the Line Profile feature on SoftwoRx , which allows individual measurements of fluorescence intensity at different wavelengths of light . Viral penetration into tissue was defined as visualization of a viral particle past the stratum corneum , since most virions were visualized in this layer . Measurements of viral penetration depth were taken using the two-point method , where a straight line was drawn between the viral particle and the closest point on the epithelial surface ( SoftwoRx ) . Depths were calculated from a clear epithelial edge and only images in which an epithelial edge could be visualized were used in the analysis . In the target cell analysis , slides of cryosections were prepared as described above and imaged using a 60X objective lens and epifluorescent microscopy . Cells were identified based on positive immunofluorescent signal , identification of a cell nucleus , and proper morphology . Cell counts were determined per image and area , and distances measured with the two-point method described above . The probability distributions of penetrators and immune cells within the tissue were calculated using kernel density estimates of the respective probability distributions in each tissue type[39] . This was done without weighting for virus stock concentrations or total numbers of virions , cells or images . Graphs and overlap area calculations performed using Interactive Data Language ( Exelisvus , CO , USA ) . Our analysis of virions in tissue focused on three parameters: 1 ) total virion count per image , 2 ) proportion of penetrators per image , and 3 ) depth of penetration into the epithelium ( from the epithelial surface ) per penetrator . Separate models for each parameter were developed by biostatisticians ( AF , FR ) to best fit the observed data and account for repeated measures in each donor . We also accounted for virus stock concentration used with each donor sample—this was done with an offset option in the model statement for the total virion count and adjusted for the proportion of penetrating virions; therefore , the calculated estimated means reported for total virion count and proportion of penetrators are per unit of virus concentration ( mean p24 = 526 . 77 ng/ml ) . For the first parameter , we used a Generalized Estimating Equation model ( GEE ) with a negative binomial distribution and logit link function for the foreskin tissue analysis and a GEE model with a zero-inflated negative binomial distribution and logit link function for the penile ( including UM ) analysis . Any virion seen on the epithelial surface , in the SC , or deep within the epithelium was included in this analysis . For the second parameter , we used a GEE model with a binomial distribution for the foreskin and penile analysis . This parameter was conditional for images in which at least one virion was visualized . To compare the first and second parameter , we analyzed the first parameter using the subset of images conditional on the presence of one virion . This was done with a zero-truncated negative binomial GEE model and the NLMixed function for both foreskin and penile datasets . We found no difference in the observations reported with the entire datasets for both the foreskin and penile data . The third parameter was analyzed using a GEE with a gamma distribution and log link model for the foreskin analysis and a GEE model with a gamma distribution for the penile analysis , conditional on images with at least one penetrator present . Each model was used to compare interactions between different tissue types , time points , and circumcision statuses . The target cell analysis was conducted with similar models . Specifically , foreskin and penile cell counts were analyzed with a GEE model with a negative binomial distribution and cell distances from the epithelial surface were analyzed with a GEE model with a gamma distribution . Each model was used to compare interactions between tissue types and circumcised statuses . All analysis was conducted with SAS 9 . 3 with an alpha of 0 . 05 . The animals used in this study were housed at the TNPRC in Covington , LA in accordance with the regulations of the American Association for Accreditation of Laboratory Animal Care . Animals were anesthetized as described above and inoculated with ∼1 . 5mls of PA GFP HIV-1 supernatant ( from the same stocks as described above in the tissue culture explant model ) for at least 15 minutes . Virus inoculations were performed by manually retracting the foreskin then submerging the penis in viral supernatant for the duration of the inoculation . The animals were then taken off sedation and allowed to resume normal activity for 4 hours before euthanization and necropsies were performed . Penile tissues were separated into glans , shaft , inner and outer foreskin and smaller sections individually snap frozen in plastic cryomolds containing OCT . Frozen tissue blocks were then shipped to Northwestern and cryosections obtained as described above . Images were acquired and analyzed using DeltaVision RT systems and SoftwoRx software as described above . | Although several clinical trials have demonstrated that male circumcision can protect men from becoming infected with HIV , we know very little about how men get infected through sex and how circumcision changes this . In this study , we explored possible sites of virus transmission across the penis by looking at how HIV interacts with adult male foreskins , penile tissues from circumcised and uncircumcised cadavers , and male rhesus macaques . Using epifluorescent microscopy , we captured images of individual HIV particles entering the penile skin , sometimes to depths where CD4+ ( potential target ) cells could be found . We found more virus in and on the inner aspect of the foreskin than the outer aspect of the foreskin after culturing for 24 hours . Additionally , there was more virus entering the glans penis as compared to foreskin tissues from uncircumcised cadaveric donors , and to greater depths in these tissues . We made similar observations of virus entering the tissue in living rhesus macaques , strengthening the results obtained from human tissues . This information should help us better understand how the virus moves into uncircumcised penile tissue placing uncircumcised men at higher risk for HIV infection during sex . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Visualization of HIV-1 Interactions with Penile and Foreskin Epithelia: Clues for Female-to-Male HIV Transmission |
Cis-regulatory modules that drive precise spatial-temporal patterns of gene expression are central to the process of metazoan development . We describe a new computational strategy to annotate genomic sequences based on their “pattern generating potential” and to produce quantitative descriptions of transcriptional regulatory networks at the level of individual protein-module interactions . We use this approach to convert the qualitative understanding of interactions that regulate Drosophila segmentation into a network model in which a confidence value is associated with each transcription factor-module interaction . Sequence information from multiple Drosophila species is integrated with transcription factor binding specificities to determine conserved binding site frequencies across the genome . These binding site profiles are combined with transcription factor expression information to create a model to predict module activity patterns . This model is used to scan genomic sequences for the potential to generate all or part of the expression pattern of a nearby gene , obtained from available gene expression databases . Interactions between individual transcription factors and modules are inferred by a statistical method to quantify a factor's contribution to the module's pattern generating potential . We use these pattern generating potentials to systematically describe the location and function of known and novel cis-regulatory modules in the segmentation network , identifying many examples of modules predicted to have overlapping expression activities . Surprisingly , conserved transcription factor binding site frequencies were as effective as experimental measurements of occupancy in predicting module expression patterns or factor-module interactions . Thus , unlike previous module prediction methods , this method predicts not only the location of modules but also their spatial activity pattern and the factors that directly determine this pattern . As databases of transcription factor specificities and in vivo gene expression patterns grow , analysis of pattern generating potentials provides a general method to decode transcriptional regulatory sequences and networks .
A central challenge in understanding metazoan genome sequences is to identify and annotate genomic regions that regulate the complex spatial and temporal patterns of gene transcription . Analysis of the regulatory regions for many individual genes has typically identified discrete enhancers or “cis-regulatory modules” ( CRMs ) that are approximately 1 Kbp long and located at distances ranging from immediately adjacent to the start of transcription to 100 Kbp away . These CRMs are composed of transcription factor binding sites that integrate information about the concentration of relevant factors to determine the quantitative contribution of each CRM to the expression of its target gene [1] . A variety of experimental approaches has been utilized to identify and characterize CRMs in single gene or genome-wide studies . For example , approximately 50 CRMs involved in the anterior-posterior ( A/P ) segmentation of the blastoderm stage Drosophila embryo [2] have been identified by reporter gene assays . A combination of genetic studies , CRM mutagenesis , and DNA binding assays has identified individual transcription factors ( TFs ) that influence the activity of these modules . Genome-wide identification of TF binding loci has been carried out using chromatin immunoprecipitation ( ChIP ) in a variety of systems , including yeast and cultured cells [3] , [4] . ChIP of TFs that act to regulate dorsal-ventral or anterior-posterior patterning in Drosophila embryos identifies a set of bound genomic regions that is highly enriched in functional targets but also includes many regions whose contribution to patterned gene expression is currently unclear [5]–[8] . Furthermore , while ChIP can identify targets in specific stages or cell types , a clear technical challenge for ChIP-based methods is how to systematically characterize the genome-wide occupancy of the large number of TFs in metazoans across the vast number of distinct expression states that occur during developmental and physiological processes . Computational analysis provides a complementary means to discover functional TF–CRM interactions in the genome . Past attempts to identify CRMs often searched for clusters of putative binding sites for combinations of TFs that act in common biological processes [9] and have been particularly successful in the identification of Drosophila segmentation modules [10] . The statistical power of these approaches is increased by filtering for evolutionary conservation of either individual sites or regions with clusters of sites [11]–[13] . In parallel , new methods to systematically determine TF-DNA binding specificities [14] , [15] have the potential to generate a relatively large number of binding specificities ( “motifs” ) in a short time . Spurred by these advances and the increasing availability of new genomic sequences , computational approaches could , in principle , be applied more globally to determine the transcriptional regulatory function of genomic sequences . However , several problems complicate the global computational annotation of CRMs and TF–CRM interactions . First , there is the problem of overlapping specificities; many TFs , particularly those in common structural families such as homeodomains , have highly similar DNA binding specificities [16] , making it difficult to assign conserved binding sites to an individual TF . Second , there is the problem of selecting the optimal combinations of TFs that should be tested together for clusters of sites; this becomes increasingly difficult as more expression states are considered . Third , there is the problem of TF pleiotropy; for example , a subset of TFs expressed during segmentation of the Drosophila blastoderm act again during cell fate specification in the nervous system . Genomic segments with overrepresentation of binding sites for these TFs might act during either developmental stage . A related problem is the identification of CRMs for genes with multiple expression domains; cluster-based analysis does not automatically attribute a specific expression domain to each CRM . Finally , there is the challenge of evaluating the relevance of individual TF–CRM interactions; while combining binding site scores for multiple TFs increases the sensitivity of CRM detection , the contribution of any individual TF to CRM function is typically smaller and more difficult to associate with a significance value . We describe a new approach for CRM identification and annotation that begins to address these issues . It employs a new method to estimate the potential of any genomic segment to drive a spatial expression pattern matching that of its nearby gene . This “pattern generating potential” is computed by combining information from experimentally determined TF binding motifs , TF expression patterns , and a comprehensive database of in situ gene expression images of the Drosophila embryo . For this approach , we developed an efficiently computable , regression-based model of expression patterns as a function of evolutionarily conserved binding sites , with parameters learned from a collection of experimentally characterized CRMs . By incorporating TF expression patterns into the model , the contribution of potential binding sites for a factor are only considered in the subset of cells that express that factor . Genomic regions are annotated as potential CRMs based on functional combinations of TF binding sites , while rejecting clusters of overrepresented binding sites that are inconsistent with the relevant gene expression pattern . Whether an individual CRM contributes to all or part of the expression pattern is an automatic result of the method . The contribution of any individual TF to this pattern can be quantitatively evaluated by examining the effect of disrupting the TF's expression pattern on the predicted activity of the CRM . We use this method to annotate genomic sequences with the potential to regulate the initial stages of segmentation in the Drosophila embryo . We exploit this approach to produce an associated transcriptional regulatory network model in which each TF–CRM interaction is associated with a confidence value . We demonstrate that this approach provides additional insights into how multiple CRMs contribute to expression patterns and how individual TFs can directly or indirectly regulate the expression of multiple target genes . This study represents a generalizable approach to produce predictive models of genome function and regulatory networks .
The availability of genome sequences for multiple Drosophila species provides an opportunity to optimize quantitative modeling of functional TF occupancy along the genome . The basic assumption of this approach is that CRMs with conserved activity across these species will maintain some binding activity for each requisite TF while binding sites in non-functional regions will be less conserved . We used genome-wide profiles of binding motif scores for 10 TFs ( BCD , CAD , HB , KNI , KR , GT , HKB , TLL , FKH , and CIC ) involved in the initial stages of anterior-posterior patterning or segmentation in the embryo . These profiles were generated using the Hidden Markov Model–based Stubb program [17] that captures both weak and strong motif matches in a probabilistic framework . We combined the motif profiles from D . melanogaster and 10 other Drosophila genomes [13] , by averaging scores from orthologous ∼500 bp regions , to create a multi-species motif profile that incorporates evolutionary conservation . Because species more closely related to D . melanogaster are better represented in the currently sequenced set of genomes , this phylogenetic comparison is weighted more heavily towards D . melanogaster than more distant species . In an alternative approach designed to reflect the evolutionary distances among the sequenced species , we modeled the motif score of a region as a random variable evolving through Brownian Motion dynamics along the branches of the evolutionary tree [18] , and computed the expected tree-wide average of this variable given its observed values in the extant species ( Methods ) . This computation is performed using a new “upward-downward” algorithm that scales linearly with the number of species . These single and multi-species motif profiles are made available through the “Genome Surveyor” interface [14] at http://veda . cs . uiuc . edu/lmcrm/ . We used published ChIP-on-chip data for eight TFs ( BCD , CAD , GT , HB , KNI , KR , HKB , and TLL ) [6] , [19] to compare the ability of different motif profiles to distinguish the top 100 bound regions from a random set of non-coding regions ( Methods ) . As Table 1 reveals , single species motif scores show significant discrimination between bound and random sequences ( p value <0 . 01 ) for each TF , with especially strong discrimination in the cases of BCD and HKB ( p value = 2 . 0E-25 and 5 . 7E-23 , respectively ) . We find a dramatic improvement in this discriminative ability when using multi-species motif profiles ( e . g . , the p value improves from 1 . 5E-5 to 1 . 9E-27 for CAD , from 1 . 8E-3 to 7 . 0E-15 for HB , and from 2 . 0E-4 to 3 . 1E-20 for TLL ) . The two schemes for combining multi-species profiles produce comparable results by this measure , which are significantly better than results produced by corresponding two-species ( D . mel . and D . pse . ) motif profiles . Both multi-species methods were also tested in CRM predictions below . We next used these binding site profiles to predict the potential transcriptional regulatory activity of any given genomic region . We reasoned that determining the potential of a region to generate patterned gene expression could help distinguish functional TF binding sites from regions that happened to have motif matches but were evolutionarily conserved for other reasons . A previous study [20] described a thermodynamic model that can recapitulate the expression activity of characterized CRMs . We developed a simpler , logistic regression model that could be readily adapted to multi-species analysis and genome-wide scanning and trained this model on a set of Drosophila CRMs . In any regression model , the parameters of the model are adjusted such that the output of the model ( e . g . , the predicted CRM activity at each A/P position in the embryo for the entire set of training CRMs ) shows the greatest agreement with the training data ( the experimentally determined expression profiles ) . Logistic regression models are a generalized version of linear regression where a sigmoidal ( “logistic” ) function is used to constrain the minimum and maximum output ( e . g . , CRM activity ) to 0 and 1 , respectively ( see Figure S11 ) . The logistic regression model used here combines weighted contributions from all TFs ( using their expression and binding sites ) . The contribution of each TF is calculated heuristically as the product of its concentration and its binding affinity to the CRM ( Figure 1A , middle panel and bottom right panel ) . The weight assigned to each TF indicates its regulating role—positive weights are used for activators and negative weights for repressors . We used this model to predict the anterior-posterior ( A/P ) expression profiles of 46 experimentally characterized CRMs in the segmentation network [2] , using multi-species motif profiles and expression patterns [2] , [21] of the 10 TFs mentioned above . A binary representation of a CRM's activity profile along the A/P axis was modeled as a function of ( i ) each TF's motif score in the CRM's sequence and ( ii ) each TF's concentration value at that position ( “bin” ) along the axis , the bins being labeled from 1 ( most anterior ) to 100 ( most posterior ) ( Figure 1A ) . The parameters of the model include a coefficient representing each TF's regulatory effect and a baseline expression value for each CRM ( which is constant across all bins ) . These parameters were trained on the known expression profiles from the 46 CRMs . Visual inspection of the results ( Figure 1B ) indicates that the expression patterns predicted by the model are in good or fair agreement with the observed expression patterns for most of the 46 CRMs . By this qualitative assessment ( which is consistent with the more quantitative assessment using “PGP scores” defined below ) , our method compares well with the results of the thermodynamic model , although a direct quantitative comparison is not feasible ( Table S1 ) . We tested for the possibility of the model “over-fitting” the data by comparing cross-validation results from the real data and randomized data and found a clear separation ( p value = 1 . 2e-34 ) between the two ( Figure S1 ) , ruling out any significant over-fitting . The above model provides “systems level” insights into the A/P network . We observed that coefficients for BCD , CAD , and FKH were fit to positive values while KNI , KR , GT , HB , TLL , HKB , and CIC were fit to negative values ( Table S2A ) , broadly consistent with the activator/repressor roles known for these factors . ( Although dual roles for some of these factors have been noted in the literature [22] , our model learns a single dominant role consistent with the dataset . ) We explored the effect of producing more complex relationships between TF expression and activity ( by adding “second order covariates , ” the squares of the term corresponding to each TF; see Methods ) and found that a second order term for BCD improved the model ( p value <E-16 ) by creating an anterior “dip” in the contribution of BCD to CRM activity ( Figure S2 ) . This broad anterior dip is not present in the BCD concentration gradient we used as input to the model . It may reflect previous observations that BCD levels appear higher than necessary for target gene activation by a simple BCD gradient model [23] , [24] . Our model may not completely account for some aspect of down regulation of BCD target genes by the terminal patterning system , either by converting BCD into a repressor [25] or through regulation of other repressors [23] , [24] . At the same time , the observation that second order covariates for the nine other TFs do not significantly improve the model's predictions suggests that the linear approximation provides a reasonable description of the CRMs' activities in terms of TF inputs . We assessed the effect of using single or multi-species motif profiles in our CRM activity pattern prediction model and found that the multi-species Brownian Motion averaging-based profiles provided the best fit ( Table 2 ) . Improved performance with multi-species scores is broadly consistent with previous studies demonstrating that A/P CRMs with conserved activity patterns and similar binding site composition can be identified in related species [11] , [26] . Interestingly , three individual modules , eve_stripe4_6 , gt_−1 , and kni_+1 , have better predictions from the model trained with single species motif profiles ( Figure S3 ) . In at least one case , this discordance between the single species and multi-species predictions is mirrored in evolutionary changes within the CRM: there is experimental evidence that the D . pseudoobscura ortholog of the gt_−1 module does not drive the posterior domain of gt expression that is observed for the D . melanogaster module ( S . Sinha et al . , manuscript in preparation ) . Thus , while the overall improvement in CRM activity predictions using multi-species profiles suggests that the majority of TF–CRM interactions in the A/P patterning network examined here are conserved , there are also examples of CRMs that have functionally diverged . One of the strengths of the A/P network as a model system is that many relevant TFs have been identified in previous molecular and genetic studies . A potential unidentified factor was suggested by the characterization of a sequence motif “TorRE” ( Torso response element ) that is overrepresented in CRMs active at the anterior or posterior termini [27] . This motif and a hypothetical concentration profile ( high at the two terminal regions ) was previously used in a thermodynamic model of CRM function [20] . We considered the hypothesis that the TF Capicua ( CIC ) acts through the TorRE motif , suggested previously by [28] based on genetic data , and further examined in a later study [23] . CIC is a transcriptional repressor that is post-transcriptionally regulated in the embryo via degradation at the anterior and posterior termini in response to Torso signaling [28] . We determined the DNA binding specificity of CIC ( Note 1 in Text S1 ) and found it to be similar to the TorRE ( p value = 0 . 0012 , Figure 2A ) , indicating that CIC can bind to most of the sites that contributed to identifying the TorRE . We found that the motif scores of TorRE and CIC are highly correlated across the 46 modules ( correlation coefficient 0 . 62; p value = 5 . 4E-6; Figure 2E ) and that CRMs with high motif scores ( i . e . , many potential binding sites ) for either factor are mostly found at the terminal regions ( Figure 2F ) . When the regression model is trained with either the TorRE motif or the CIC motif ( and their respective concentration profiles , Figure 2B ) , the quality of fit is comparable ( Figure 2C , 2D ) . Consistent with the complementary expression patterns for TorRE and CIC , CIC has a negative rather than positive coefficient , confirming that it generally acts as a transcriptional repressor . Adding TorRE to a model that already includes CIC leads to no significant improvement ( unpublished data ) . These results indicate that CIC is the TorRE binding factor and that this factor acts by repressing target CRMs in the center of the embryo rather than activating targets at the termini . Individual direct and indirect targets of CIC are discussed below . The ability to predict the spatial expression pattern driven by a module ( CRM activity ) suggests a method for discovery of novel CRMs: to scan the flanking genomic sequences of a gene for segments whose predicted activity pattern agrees with the gene's endogenous pattern . For this purpose , we developed a newly defined measure of similarity between expression profiles and its statistical significance; this measure is named the “Pattern Generating Potential” ( PGP ) ( Figure 3A , Methods , Note 2 in Text S1 ) . The scoring method was designed to: ( 1 ) be sensitive to both the shape and magnitude of the predicted expression profile , ( 2 ) avoid biases towards or against overly broad or overly narrow domains of expression , and ( 3 ) automatically allow sub-domains of a gene's expression pattern to be directed by the CRM ( Figure 3B ) . To compute this score , we first calculate the average predicted CRM activity in domains of gene expression ( the “reward” term ) and domains of non-expression ( the “penalty” term ) and subtract the penalty from the reward , followed by a linear transformation generating PGP values between −1 and 1 ( Figure 3C ) . An important feature of this score is that it can identify CRMs that contribute to only part of a gene's expression pattern ( see below ) . When applied to the 46 CRMs used in the regression model above , the PGP score was highly correlated with our visual assessments of prediction success ( Figure S4 ) . We tested this measure on the 22 genes ( henceforth called “A/P-22” ) regulated by the 46 CRMs described above . Expression data were obtained from whole embryo in situ hybridization images from BDGP ( http://www . fruitfly . org/cgi-bin/ex/insitu . pl ) and FlyExpress [29] ( data available at http://veda . cs . uiuc . edu/lmcrm/ ) . We scanned the control region of each gene ( Note 3 in Text S1 ) with a sliding window of size 1 Kbp , predicted the A/P expression profile based on the motif scores in that window , and calculated the PGP ( Figure 3A ) . An empirical p value representing the statistical significance of a putative module was estimated based on how frequently we observed a window with equivalent or greater PGP score in a genome-wide scan . Of the 62 modules predicted at a p value threshold of 0 . 015 , 34 had significant overlap ( >50% ) with known modules , indicating 55% specificity at 74% sensitivity ( Figure S5 ) . Seventeen of the remaining 28 predicted modules overlapped the bound regions of at least one TF ( ChIP data at 1% FDR from [6] , [19] ) , indicating that the majority of predicted CRMs are functional and/or biochemical targets of A/P factors . Overall , we did not observe any systematic biases in the score , and modules with broad ( “gap” ) as well as sharp ( “pair-rule” stripes ) patterns were correctly predicted . The genomic location and predicted expression activity for each of these CRMs are available at http://veda . cs . uiuc . edu/lmcrm . The 12 known modules not recovered included 10 that had either “bad” or “fair” predictions by the regression model ( Figure 1 ) , pointing out that CRMs whose expression is poorly predicted by the model are difficult to detect in the CRM search . For another CRM ( gt_−6 ) , the experimentally characterized activity pattern does not agree with the endogenous gene expression pattern we used ( Note 4 in Text S1 ) . In this case , the CRM activity pattern we used [2] may reflect either an experimental artifact or expression at a different embryonic stage . In only one case ( h_stripe1 ) , the PGP approach was unable to recover a module despite the training stage prediction being of high quality . Thus , most of the false negatives are likely to be due to the current limitations in the ability to predict CRM expression activity . The results of this search were compared to two previously described CRM prediction programs , Cluster Buster [9] and Stubb [17] , that search for clusters of binding sites for multiple TFs . To ensure that the performance of the PGP method was not influenced by including the same CRMs to train parameters that were then part of the predicted set , we used a cross-validation strategy where all known modules of a single gene were left out of the training phase before predicting CRMs within the control region of that gene . The PGP method performed better than both single and multi-species versions of Stubb and Cluster Buster ( Figure 3D ) . Unlike the other CRM prediction approaches , the PGP method predicts which aspect of the gene's pattern is regulated by an individual CRM , allowing the range of regulatory architectures for the A/P-22 genes to be examined: solitary CRMs , multiple CRMs contributing to distinct aspects of the pattern , or multiple “sibling” CRMs with a similar predicted activity . ( We use the term “sibling” to indicate CRMs that may have effectively redundant activity within the context of our model , but possibly distinct contributions to the magnitude , temporal regulation , or robustness of patterned gene expression in vivo . ) In our predictions , there was only one gene ( btd ) with a single predicted CRM; this prediction overlaps a known CRM ( btd_head ) driving the gene's expression . In all other cases , two or more modules were predicted in a single gene's control region . These included cases where distinct aspects of a gene's blastoderm expression pattern are captured by distinct predicted CRMs ( e . g . , five CRMs near the gene eve , including four known CRMs ) , a well-established phenomenon reported for primary pair-rule genes . We also found many cases of “sibling” CRMs , where multiple modules near a maternal/gap gene were predicted to drive highly similar expression patterns ( Figure 4A ) . We considered whether possible false positive predictions could account for this observation; if the occurrence of a second , functionally similar CRM prediction in a gene's control region is an artifact of false positives , they should also be found near other randomly selected genes . However , we find that enrichment of functionally similar CRMs near the target gene is highly significant ( p value = 4E-4 , Table S3 ) . Given the previous identification of “shadow” CRMs in the dorsal-ventral patterning network [30] , the utilization of functionally similar CRMs may be a more common theme of cis-regulatory organization than currently recognized . We applied the PGP method to a larger collection of 144 genes with patterned expression along the anterior-posterior axis [12] . ( A/P-22 genes were not included . ) We automatically extracted the A/P expression profiles of these genes from the FlyExpress database [29] , transformed the intensity values into binary expression domains ( Methods ) , and identified flanking sequences with PGP at the same empirical p value threshold used above . ( Predicted sequences that did not have above-average binding site presence for one of the activators in the model or for the broadly expressed activator Stat92E were discarded . ) We identified 123 putative CRMs from 68 genes , henceforth called the “FlyExpress” gene set ( data at http://veda . cs . uiuc . edu/lmcrm; the 60 most significant predictions are shown in Figure 4B ) . The distribution of PGP scores and their empirical p values is similar to that of A/P-22 and very different from that of bona fide non-modules that were identified as false positives in a cluster-based method to identify CRMs ( Figure 3E and Figure S6 ) . 44% of the predicted CRMs overlapped a ChIP-chip peak ( at 1% FDR; 65% when considering peaks at 25% FDR; Table S4 ) . The predictions included CRMs for genes with a single expression domain and genes with multiple expression domains ( e . g . , slp1 and ara , respectively; Figure 4B ) . Among CRMs corresponding to genes with multi-domain patterns , 53% capture only one of the domains of the endogenous pattern ( e . g . , drm; Figure 4B ) while 47% capture more than one domain ( e . g . , emc ) . Sixteen of the above CRM predictions overlapped previously verified modules , of which 12 have blastoderm stage expression that agrees with the predicted expression profile from our model ( Table S9 ) . These provide an independent experimental validation for our CRM and activity prediction pipeline . In addition , we tested seven CRM predictions using new reporter transgenes . These lines were created as part of an ongoing project to systematically examine regulatory regions surrounding a subset of Drosophila genes with patterned expression in the nervous system [31] . Only predictions in intergenic or intronic regions of at least 10 Kbp were chosen for analysis . Selections included regions flanking genes manually annotated as “strong” or “weak” A/P patterned expression . 4 of 7 tested regions exhibited reporter gene expression patterns resembling the predicted pattern ( Figure 5 ) . For one of these , Ubx , the anterior boundary of reporter expression is in the correct region of the embryo , but initiation of the pattern is delayed relative to the endogenous gene and more strongly resembles the endogenous gene expression pattern at this slightly later stage; it has more posterior expression and a striped pattern likely reflecting the activity of later acting repressors not included in our model . Two of the remaining tested reporters ( pdm2 and emc ) exhibit expression in the developing CNS , where many of the same TFs that regulate A/P patterning are re-expressed ( unpublished data ) . It is possible that the same combinations of TFs that predict an A/P pattern in our model can act to direct patterned expression in the developing CNS . We note that the specificity we observed here ( 57% ) is about the same as that recorded in our cross-validation tests on the A/P gene set . We also examined the genome-wide locations of all segments with high PGP scores ( and not just those located near the genes whose expression was modeled ) . We found these segments to be preferentially located near A/P patterned genes . However , we also observed a large number of segments ( with high PGP ) that are apparently not associated with patterned genes ( Table S12 ) . This suggests that the genome harbors a relatively large number of segments with PGP , and only a small subset of these actually realize this potential . This finding further supports our rationale of searching only in the neighborhood of a gene for segments with the potential to drive the gene's expression pattern . Unlike binding site clustering methods , the PGP method uses both the binding specificities of TFs and their expression pattern to predict the activity pattern of a CRM . Using the PGP method , it is possible to computationally assess the contribution of each TF to the CRM by asking if altering the expression of the TF affects the quality of the prediction . We used this strategy to infer direct regulatory interactions between TFs and CRMs , depicted as edges in the transcriptional regulatory network . To visualize the effect of removing an individual TF from the model , we simulated a “knock down” of the TF ( by setting its expression to 0 ) and compared the predicted CRM expression in this “in silico mutant” background and in “wild type” ( Figure 6A , knock down patterns shown in green ) . Unlike traditional in vivo genetic assays , where observed changes may be the indirect effect of mis-regulation of other genes , this approach examines the direct contribution of a TF to a specific CRM . In order to assign a statistical significance to this contribution , we developed an alternative procedure ( Methods and Figure 6A ) : CRM activity predictions were generated using random permutations of the TF's concentration profile and compared to the “true” activity , thus creating a null distribution of similarity scores ( depicted in blue ) . The score obtained with the actual profile ( black dot ) was compared to this distribution , generating an empirical p value . When there are few binding sites in the CRM , the TF pattern has little influence on CRM predictions and the null distribution of scores is very narrow ( unpublished data ) . When there are more binding sites in the CRM , there is a broader distribution of similarity scores from the random profiles , and the position of the actual profile within this distribution reflects the combined contribution of the binding sites and the normal TF expression pattern on CRM activity . Using this procedure to infer a p value for every TF–CRM combination , we constructed a transcriptional regulatory network ( Figure 6B , Figure 7 ) involving the 35 CRMs where the model's quality of fit had been “good” or “fair” ( Figure 1B ) . A total of 102 regulatory edges were predicted ( at p value <0 . 05 ) between the 10 TFs and 35 CRMs , revealing a very dense network . ( See http://veda . cs . uiuc . edu/lmcrm . ) 82 edges were supported by ChIP-based evidence of occupancy at the strongest level ( 1% FDR ) . 63 of the 102 edges have been previously reported in the literature , mostly by examination of CRM activity in mutant embryos lacking the TF ( Table S5 ) . In some cases , confidence in experimentally determined TF–CRM edges is further increased by in vitro confirmation of TF binding sites by DNaseI footprinting . For 12 of the 35 CRMs analyzed above , the FlyReg database [32] catalogs at least one such interaction with either BCD , CAD , KR , KNI , HB , GT , or TLL . These validated TF–CRM edges were significantly enriched in our network ( Hypergeometric test , p value = 0 . 0026 ) ( Figure S7 ) . This network model can address specific questions about the role of individual TFs in A/P patterning . For example , the concentration of the repressor CIC is a direct output of the terminal patterning system [33] , but it is not known whether this mechanism acts solely by determining the terminal expression patterns of TLL and HKB . Terminal gene expression could be either entirely regulated by these factors or the terminal system might also directly regulate additional targets via CIC . The regulatory network model predicts that CIC directly targets at least six CRMs corresponding to five distinct genes—tll and hkb as well as cnc , fkh , and kni ( Figure 2G ) . Thus , this result extends our observation above that CIC binds to the Torso response element and indicates that control of CIC levels by the terminal regulatory system indirectly regulates many genes via tll and hkb but also has some additional direct targets besides tll and hkb . Existing experimental evidence also points to a role of the terminal system in regulating these CRMs or their associated genes [34]–[36] , although evidence of direct influence has been missing . This finding complements our current understanding of the terminal patterning system , which has thus far been shown to act only through the TFs TLL and HKB [36] , [37] . We used the above statistical procedure to construct a regulatory network for all of our CRM predictions ( 62 in the A/P-22 set , 123 in the FlyExpress set ) ; the TF–CRM interactions composing this network are cataloged in Table S6 . Analysis of the predicted network reveals several common patterns . A recurring theme in the TF–CRM interactions is potential “auto-regulation” by activators . For example , all three predicted modules near the cad gene had significant regulatory input from CAD , and in each case , this predicted auto-regulation was supported by ChIP data ( at 1% FDR ) . Similarly , four out of five predicted modules for fkh are predicted to have FKH-driven activation . fkh auto-regulation ( in salivary glands ) has been experimentally shown [38] . On the other hand , auto-regulation by repressors is not seen in our predictions , as anticipated . Another common theme observed was that of mutual repression by pairs of TFs , e . g . , HB–KNI , GT–KR , KR–KNI , HB–KR , GT–KNI , and TLL–KR , some but not all of which were reported previously [22] , [39]–[41] . We also used edges of this network to characterize the “complexity” of CRMs . About three TFs on average had edges leading into each CRM , for most target patterns , except that CRMs driving expression at the anterior seem to have relatively low complexity and those active in bins 80–90 have slightly greater complexity ( Figure S8 ) . When we examine the degree distribution of the network from the perspective of the TFs , all 10 TFs contributed almost uniformly to the predicted CRMs ( Figure S9 ) . We compared the above-mentioned network to that predicted by Stark et al . [13] , which was based on the presence of conserved ( predicted ) binding sites in 2 Kbp promoters of all Drosophila genes . However , we found very little overlap between the two networks ( Note 5 in Text S1 ) , which we attribute to the fact that only a small percentage of our predicted CRMs ( and of experimentally validated A/P CRMs ) are located in the 2 Kbp regions immediately upstream of genes . Genome-wide ChIP assays provide the location and strength of TF occupancy in vivo . Compared with cell culture or yeast experiments , intact organisms represent potentially more challenging contexts to interpret ChIP data since TF expression can vary across space and time . A recent analysis of CRMs acting in mesoderm development demonstrates that time course ChIP data can predict multiple distinct classes of CRM activity patterns in whole embryos [8] . In contrast , complementary computational methods lack the in vivo context of ChIP but can provide a potentially general approach to predict regulatory networks , even in cells and tissues that are difficult to characterize experimentally . A high-quality genome-wide ChIP dataset is available for eight of the A/P TFs during early embryogenesis [6] , [19] . In the previous sections , we have used this dataset to confirm that a majority of the PGP-derived CRM predictions correspond to in vivo occupancy by one or more TFs . In this section , we evaluate if the TF–CRM interactions predicted by the PGP method can approach the quality of predictions derived from ChIP data .
As large numbers of genome sequences become available , annotation of how different genomic segments contribute to organismal function remains a central challenge . Despite the relative simplicity of the genetic code , annotation of the protein-coding regions of large genomes has undergone continued revision as new experimental datasets and computational approaches have been developed . Computational annotation of CRMs is significantly less advanced , in part due to the incomplete description and complexity of metazoan TF-DNA binding specificities . However , even after determining binding motifs for the central regulators of Drosophila anterior-posterior patterning network [14] , we found it difficult to use existing clustering strategies to systematically search for the targets of these factors . Here , we describe an alternative strategy—use binding site motifs to predict the A/P activity pattern for a given DNA sequence and determine the similarity of the predicted activity pattern to an experimentally determined expression pattern . The PGP can be used to annotate the non-coding genome , similar to the “regulatory potential” score of [42]; unlike the regulatory potential , which generally classifies non-coding sequences as regulatory or neutral , PGP ranks sequences by their ability to contribute to the specific expression pattern of a nearby gene . It further facilitates a quantitative inference of TF–CRM interactions , whose validity may then be assessed through in vivo observations . We have specifically applied this approach to the A/P network , but it should be applicable to any system in which adequate expression data are available for relevant TFs , CRMs , and target genes . One key distinction between using PGP to characterize CRMs instead of binding site clusters is that this method can automatically select appropriate combinations of TFs to contribute to a CRM's activity . By only considering TFs expressed at the appropriate time and place , this approach partly addresses issues associated with TF specificity overlap and pleiotropy . A second advantage is the rich class of expression patterns with which it may be used . The current implementation accommodates expression states composed of any combination of 100 positions along the A/P axis and can be expanded in a straightforward way to include additional spatial and temporal dimensions . These expression patterns can even be the result of automated image-processing pipelines , such as the one used here for A/P patterns . This is in contrast to the more limited classes of manually determined expression patterns considered in previous studies [8] . The regression model also has the advantage that the explicit activity pattern predictions are easily interpreted , compared to other machine-learning techniques such as Bayesian networks [43] or support vector machines [8] . As noted above , replacing binding site profiles with ChIP-based occupancy data in our model did not lead to superior predictions . This negative result is somewhat contrary to the findings of Zinzen et al . [8] , who exploited ChIP data on five TFs ( at five different time points ) to successfully predict spatio-temporal expression patterns of many CRMs involved in mesoderm specification . Integration of both ChIP and motif presence information may hold the key to significantly improved predictions and will be an exciting area for future research . Generating the experimental datasets required to apply this method more broadly should be feasible with current technologies . The bacterial one-hybrid system and other methods should generate DNA binding specificities for most Drosophila TFs [14]–[16] , [44] . Systematic determination of the temporal and spatial expression patterns of TFs is critical and not a minor task; however , it should be far more straightforward than applying genome-wide ChIP methods to the many different possible cell types present at different developmental stages . In addition to TF binding specificities and expression patterns , two other datasets are required . First , large-scale descriptions of gene expression patterns must be available; these are currently being generated for the Drosophila embryo [29] , [45] , [46] . Second , a training set of CRMs with diverse activity patterns must be identified; for the Drosophila embryo , these can be curated from the literature [47] or generated in moderate to high throughput reporter studies [31] . While we have treated CRM and gene expression patterns as binary values at a single developmental time point , quantitative spatial and temporal expression data are readily accommodated in this approach and should capture more comprehensive and subtle aspects of transcriptional regulatory networks . We note that the specific components of the model used here for predicting segmentation modules may change as more genomes and relevant TF motifs are characterized in the future . At the same time , our tests suggest ( Tables S10 , S11 ) that including more genomes ( and to some extent additional motifs ) may not lead to dramatic improvements . The logistic regression model used here is very similar to the more popular linear regression model [48] , combining weighted contributions from all TFs , except that the logistic model imposes the combined output to saturate at high values ( Figure S11 ) . This model is “simpler” than a previously published thermodynamic model to predict regulatory function from sequence [20] , in that it has fewer parameters to be trained from data . At the same time , it performs well compared to the thermodynamic model and has the added advantages of easily incorporating multiple species comparisons and of computation that is orders of magnitude faster . This enables fast , genome-wide prediction of other CRMs , examination of the effect of each motif on each putative CRM , and empirical assessment of its statistical significance through permutation tests . However , the regression model does not incorporate known mechanistic features of CRM function , such as cooperative TF binding . More detailed models of CRM function have been developed for individual enhancers [40] , [49] , [50] , which can accurately describe changes in CRM activity over developmental time or due to mutation . In principle , these models or other approaches to capture how binding site arrangements contribute to expression could replace the regression model in the overall framework to measure PGP . Models with additional parameters may provide better predictions but require additional prior knowledge , while models with fewer parameters may generalize better . We also note that the motif scores used in our model are based on evolutionary conservation at the ∼500 bp resolution and are thus robust to local turnover of sites [51] . The approach is also applicable with single-species motif scores ( although this led to poorer performance in our setting ) , which may be significant for discovery of CRMs that are not as well conserved [52] . This analysis of the Drosophila A/P patterning using PGP is the most complete description of this network to date . The quantitative descriptions of how TF inputs generate the activity pattern of specific CRMs and the explicit predictions of individual TF–CRM interactions provide a level of detail not typically generated by other approaches . In this study , we have highlighted a few specific novel observations on the predicted regulatory network , such as which genes regulated by the terminal system are direct or indirect targets of CIC . We have also identified and experimentally confirmed the activity of four new CRMs of the A/P patterning network , regulating the genes Antp , noc , SoxN , and Ubx . In addition , we identified several general conclusions about the network , including the frequent occurrence of positive autoregulation by activators and mutual repression by spatially adjacent repressors . One of the most striking results is how often individual genes have multiple CRMs predicted to direct the same embryonic expression pattern . Individual examples of such “sibling” CRMs have been previously described in both the A/P and D/V embryonic patterning networks , but the current analysis indicates that they may be more frequent than previously appreciated . ( A more complete experimental analysis of this aspect of cis-regulatory architecture will be required given the observation that at least some of these CRMs are in fact “cousins” that appear to use similar TF binding sites to drive patterned expression in a different tissue . ) Application of the PGP method to other transcriptional regulatory networks should reveal if similar overall regulatory themes act in other developmental contexts . Recent ChIP-chip analysis of multiple TFs regulating Drosophila embryonic patterning provides a quantitative dataset to compare with our computational approach [6] , [19] . Overall , ChIP data suggest far more binding events than expected to be required to directly control patterned gene expression [6] , consistent with earlier predictions of widespread genome binding by TFs [53] , [54] . Presumably , as long as occupancy does not interfere with patterning , it can be tolerated . In contrast , computational TF binding site profiles incorporate multiple species comparisons to probe where TF binding sites are under evolutionary selection , which should reflect a conserved role in patterning . In our comparison of ChIP data and TF binding site profiles for the A/P network , we find that evolutionarily conserved binding sites provide greater specificity and that this leads to better gene expression prediction models and a greater enrichment of known TF-CRM interactions . Interestingly , we found several examples of disagreement between motif-based and ChIP-based prediction of binding where the ChIP occupancy appears to be antagonistic to the known activity pattern of the CRM . For future work in cells where high-quality ChIP data are available , integrating ChIP scores and multi-species motif profiles may allow higher confidence predictions of CRM position , function , and regulation by combining both experimental evidence for availability and occupancy with evolutionary evidence for function [13] , [55] . For cells where ChIP data are not available , determining the PGP of genomic regions can provide a general strategy to annotate regulatory regions . In summary , this work presents a general computational framework for analyzing transcriptional regulatory networks through a systematic integration of sequence ( from multiple species ) , expression , and TF binding specificity data , all of which are hallmarks of the genomics toolkit available today . Application of the framework provides systems-level insights into the regulation of anterior-posterior patterning in the Drosophila embryo .
All motif profiles are made available through the Genome Surveyor interface at http://veda . cs . uiuc . edu/lmcrm . Given a phylogeny and a profile value for each leaf node or extant species , our task is to compute an evolutionary average of the given values . Following [56] , we consider a random variable evolving according to a Brownian Motion process along each branch , with the process on each branch being conditional on the value of the variable at the parent node of that branch . The temporal expectation of this random variable , over all branches , is the desired average . Exploiting the observation that the random variable has a Gaussian distribution at every ( non-root ) node with mean and variance defined by the value at the parent of that node , researchers have shown [18] how this temporal average may be calculated in time O ( n2 ) , where n is the number of species . We developed an alternative , O ( n ) algorithm for this purpose based on the upward-downward algorithm paradigm for trees ( Text S2 ) [57] . The top 100 bound regions of a TF , each defined as the 500 bp centered on a ChIP peak , were used , along with 5 , 000 length-matched sequences selected at random from non-coding regions . For each sequence , the motif score was computed , and the sets of scores for the bound and random regions were compared using the Wilcoxon Rank-Sum test . The basic model for predicting CRM expression patterns is as follows: ( 1 ) where is the expression value ( between 0 and 1 ) of the CRM l in bin b ( the A/P axis is divided into 100 equal bins ) , is the concentration of factor i in bin b , is the motif score of factor i in the CRM l , is the regression coefficient for factor i , is the “basal” expression level of CRM l , sig ( x ) is a “sigmoid” function 1/ ( 1+exp ( −x ) ) . The free parameters are ( one for each CRM ) and ( one for each factor ) . Use of the CRM-specific parameter is motivated by ( i ) the fact that the discrete ( 0/1 ) expression values that form the desired output do not reflect the variation in basal gene expression levels and ( ii ) an opportunity to compensate for , at least partially , the lack of complete knowledge of relevant TFs , especially of ubiquitous activators and/or repressors . Note that the concentration and motif score terms occur together ( ) and this product is called the “covariate” of factor i for CRM l in bin b . An additional higher order term , called “BCD2 , ” is used in our model . BCD2 is the square of the covariate “BCD” for the factor BCD . Utilizing the glm ( generalized linear model ) function in R's “stats” package [58] , we trained the parameters of the model using iteratively reweighted least squares ( IWLS ) to minimize the error between predicted and true expression values . The ratio between the trained parameter and its estimated standard error was treated as a z score for calculating its statistical significance . The overall quality of fit of the model to the data was measured by standard statistics such as the root mean squared error ( RMSE ) , mean Correlation Coefficient ( CC ) , and the Akaike Information Content ( AIC ) ( Note 6 in Text S1 ) . All analyses were performed within the R programming environment . Given a predicted expression profile ( real numbers between 0 and 1 for each bin along A/P axis ) and an endogenous expression profile ( 0 or 1 values for each bin ) , we defined the PGP score as follows: ( 2 ) where Eg , b is the expression value ( 0 or 1 ) of the gene g in bin b and is the predicted expression value ( between 0 and 1 ) . This score ranges from −1 to +1 . It rewards correctly predicted domains of expression and penalizes false prediction of expression . If the endogenous profile has multiple domains of expression , a subset of those domains are selected based on the predicted profile and then compared to the predicted profile using PGP . See Figure 3B , Figure 3C , and Note 2 in Text S1 for additional details . First , we computed the “root mean square error” ( RMSEwt ) between the model's prediction in “wild type” and the true expression . Next , the concentration profile of a factor was permuted randomly and the RMSE was determined for each permutation . Repeating this step 1 , 000 times , we obtained an empirical p value of RMSEwt , which represents how important this factor ( in particular , its concentration profile ) is to the CRM's expression pattern . A/P axis expression profiles were calculated from ∼36 , 800 BDGP in situ expression images ( lateral orientation ) . The images were first converted into a standardized format and aligned using the FlyExpress image processing pipeline [59] . The resultant 128×320 standardized images were then manually inspected and corrected , as necessary . For generating an A/P expression profile , the expression values were calculated for each of the 320 points along the A/P axis by taking the average of the intensity values within a window around the middle of the Dorsal-Ventral axis . All images for a given gene and developmental stage 4–6 were visually examined to identify those with A/P patterns and select a representative image , whose A/P profile was further discretized into domains of expression and non-expression along 100 equal-sized bins along the axis , as described in Note 7 in Text S1 . Genomic regions encompassing predicted CRMs were tested as transgene reporter constructs as described previously [31] using labeled RNA probes to assay expression of the Gal4 reporter gene by in situ hybridization to mixed stage embryos . Candidate CRMs for the following genes were tested ( with release 5 coordinates for the tested fragment ) : noc ( 2L:14487040–14490562 ) , SoxN ( 2L:8831242–8833223 ) , Antp ( 3R:2774228–2775839 ) , Ubx ( 3R:12503151–12505092 ) , apt , ( 2R:19455701–19458627 ) , emc ( 3L:745029–746591 ) , and pdm2 ( 2L:12669616–12672809 ) . | The developmental program specifying segmentation along the anterior-posterior axis of the Drosophila embryo is one of the best studied examples of transcriptional regulatory networks . Previous work has identified the location and function of dozens of DNA segments called cis-regulatory “modules” that regulate several genes in precise spatial patterns in the early embryo . In many cases , transcription factors that interact with such modules have also been identified . We present a novel computational framework that turns a qualitative and fragmented understanding of modules and factor-module interactions into a quantitative , systems-level view . The formalism utilizes experimentally characterized binding specificities of transcription factors and gene expression patterns to describe how multiple transcription factors ( working as activators or repressors ) act together in a module to determine its regulatory activity . This formalism can explain the expression patterns of known modules , infer factor-module interactions and quantify the potential of an arbitrary DNA segment to drive a gene's expression . We have also employed databases of gene expression patterns to find novel modules of the regulatory network . As databases of binding motifs and gene expression patterns grow , this new approach provides a general method to decode transcriptional regulatory sequences and networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"developmental",
"biology",
"computational",
"biology/transcriptional",
"regulation"
] | 2010 | Quantitative Analysis of the Drosophila Segmentation Regulatory Network Using Pattern Generating Potentials |
A sustained outbreak of leptospirosis occurred in northeast Thailand between 1999 and 2003 , the basis for which was unknown . A prospective study was conducted between 2000 and 2005 to identify patients with leptospirosis presenting to Udon Thani Hospital in northeast Thailand , and to isolate the causative organisms from blood . A multilocus sequence typing scheme was developed to genotype these pathogenic Leptospira . Additional typing was performed for Leptospira isolated from human cases in other Thai provinces over the same period , and from rodents captured in the northeast during 2004 . Sequence types ( STs ) were compared with those of Leptospira drawn from a reference collection . Twelve STs were identified among 101 isolates from patients in Udon Thani . One of these ( ST34 ) accounted for 77 ( 76% ) of isolates . ST34 was Leptospira interrogans , serovar Autumnalis . 86% of human Leptospira isolates from Udon Thani corresponded to ST34 in 2000/2001 , but this figure fell to 56% by 2005 as the outbreak waned ( p = 0 . 01 ) . ST34 represented 17/24 ( 71% ) of human isolates from other Thai provinces , and 7/8 ( 88% ) rodent isolates . By contrast , 59 STs were found among 76 reference strains , indicating a much more diverse population genetic structure; ST34 was not identified in this collection . Development of an MLST scheme for Leptospira interrogans revealed that a single ecologically successful pathogenic clone of L . interrogans predominated in the rodent population , and was associated with a sustained outbreak of human leptospirosis in Thailand .
Leptospirosis is a zoonotic infection caused by pathogenic members of the genus Leptospira . Human disease is usually acquired following environmental exposure to Leptospira shed in the urine of an infected animal [1] , [2] . Infection is acquired during occupational or recreational exposure to contaminated soil and water , organisms gaining entry to the accidental human host via abrasions or less commonly the conjunctiva [1] . Disease may also be acquired through direct contact with infected animals , and occurs in farmers , veterinarians and abattoir workers [1] . The disease has a worldwide distribution but is most common in tropical regions where incidence peaks during the rainy season [1] , [2] . Clinical manifestations are broad ranging and follow a biphasic pattern in which a septicemic phase lasting around one week is followed by an immune phase during which antibodies are raised and organisms localize in tissues and appear in urine . Much disease is sub-clinical or mild , but patients reaching medical attention usually have an acute febrile illness associated with one or more of chills , headache , myalgia , conjunctival suffusion , and abdominal symptoms which can include nausea , vomiting and diarrhea [1] . Leptospirosis has been described as anicteric or icteric; the former represents 85–90% of cases and is associated with a good prognosis , while the latter may be associated with multisystem disease involving particularly the kidneys , lung and heart , with a reported mortality rate of 5–15% [1] . Leptospirosis is an emerging infectious disease in Thailand [3] , [4] . Before 1996 , the number of cases reported to the Department of Disease Control ( DDC ) was approximately 200 per year . Leptospirosis was sporadic and reported mainly from central and southern regions . A marked change occurred in the subsequent decade , with a year-on-year rise from 398 cases in 1996 to a peak of 14 , 285 cases in 2000 . This was followed by a continuous decline with 2 , 868 cases reported during 2005 [5] . Reporting in Thailand is voluntary and probably represents a small proportion of true cases . There was also a shift in the geographical distribution , with the majority of cases being reported in the northeast . One explanation for the outbreak is that it was related to the emergence of a biologically successful clone of Leptospira . This possibility is supported by a study of 44 leptospiral strains obtained from humans during three outbreaks in Brazilian urban centers , in which typing using arbitrarily primed PCR demonstrated that 43 isolates exhibited very similar fingerprints suggestive of a clonal population of L . interrogans [6] . In addition , during a large urban outbreak in Brazil , L . interrogans serovar Copenhageni was isolated from 87% of cases with positive blood cultures [7] . Although it is currently unclear to what extent genetic relatedness can be informed by serotype alone , this observation is consistent with the majority of cases being caused by the expansion of a single outbreak clone . The aim of this study was to define the molecular epidemiology of Leptospira strains isolated from humans during the Thai outbreak , and to relate this to the maintenance animal host . To achieve this , an MLST scheme was developed for L . interrogans , the major cause of human disease . This approach has the advantage over existing typing schemes in that the data generated are amenable to detailed evolutionary analysis . MLST data are also readily comparable via the internet , and establishment of an MLST scheme therefore paves the way for future studies . Our results confirm the emergence of a dominant clone of L . interrogans serovar Autumnalis; this was the major cause of human disease , and was found in a maintenance host which was defined as the bandicoot rat .
A prospective study was undertaken at Udon Thani General Hospital in northeast Thailand to identify patients with leptospiremia . This 1 , 000 bed provincial hospital serves a predominantly rural population , >80% of whom are rice farmers and other agricultural workers who are repeatedly exposed to rats and water contaminated by rat urine . Patients were recruited during consecutive months from October 2000 to December 2002 , then for four months during each rainy season ( July to October inclusive ) during 2003 and 2005 . The reason for this pattern of recruitment is that leptospirosis is predominantly a rainy season disease . Consecutive adult patients ( ≥15 years ) presenting with fever ( >37 . 8°C ) of unknown cause were recruited following informed and written consent . Patients with a blood smear positive for malaria parasites or other definable infections such as pneumonia or urinary tract infection were excluded . The clinical features of leptospirosis are broad ranging and similar to other acute febrile illnesses common to this geographic area such as scrub typhus and dengue fever . In view of this , all adult patients presenting with acute undifferentiated fever were cultured to detect leptospiremia . A 10 ml blood sample was drawn on the day of admission into a sterile tube containing 250 units of sodium heparin for Leptospira culture . The study protocol was approved by the Ethical Committee of the Ministry of Public Health , Royal Government of Thailand . A further 24 unselected isolates cultured from the blood of patients with leptospirosis presenting to hospitals in 8 additional provinces in Thailand during the rainy seasons of 2003 and 2004 were obtained from strain collections . These provinces were: Lumpang ( situated in the north ) , Yasothon , Nakhon Ratchasima , Maha Sarakahm and Loei ( northeast ) , Ratchaburi ( central ) , Rayong ( east ) and Chumphon ( south ) . Seventy six reference strains representative of the species L . interrogans , L . kirschneri and L . borgpetersenii were obtained from the WHO/FAO/OIE Collaborating Center for Reference & Research on Leptospirosis , Australia , or National Institute of Health , Thailand . A total of 1 , 126 rodents were trapped in Nakhon Ratchasima , northeast Thailand during 2004 by the National Institute of Health , Thailand . Animals were identified and cultured for Leptospira , as described previously [8] . Ten animals were culture positive ( Bandicota indica 8 , Bandicota savilei 1 , and Rattus rattus 1 ) , while all samples from Rattus exulans , Rattus losea , Mus cervicolor , Mus caroli and Sancus murinus were culture negative for Leptospira . Eight unselected isolates remained viable and were evaluated in this study; of these , 6 were isolated from B . indica ( greater bandicoot rat ) and 1 each was isolated from B . savilei ( lesser bandicoot rat ) and Rattus rattus ( black rat ) . Culture of leptospires from human blood was performed using EMJH supplemented with 3% rabbit serum and 0 . 1% agarose , as described previously [9] . Positive cultures were sent to the WHO/FAO/OIE Collaborating Center for Reference & Research on Leptospirosis , Australia for serovar identification using the cross agglutinin absorption test ( CAAT ) [10] . Definitive identification of species was undertaken by amplification and sequencing of the near full-length 16S rRNA gene . Primers were designed to anneal to conserved regions of genes from pathogenic species L . interrogans , L . kirschneri , L . borgpetersenii , L . santarosai , L . alexanderi and L . fainei . The primers ( f - 5′ GTTTGATCCTGGCTCAG 3′ and r -5′CCGCACCTTCCGATAC 3′ ) amplified a 1 , 483 bp PCR product which was sequenced in its entirety using internal primer pairs ( primers available on request ) . Genomic DNA was extracted using the High Pure PCR Template Preparation Kit ( Roche Applied Science , Germany ) . In a pilot study , 14 housekeeping loci were selected using the whole genome sequence of L . interrogans serovar Lai strain 56601 ( 11 loci situated on chromosome I and 3 loci on chromosome II; loci and primer sequences available on request ) . Primers were designed using PrimerSelect software ( DNASTAR Inc . , Wisconsin , USA ) , and synthesized by Sigma-Proligo ( Proligo Singapore Pty Ltd ) . These were evaluated using 30 clinical or reference strains belonging to species L . interrogans , L . kirschneri or L . borgpetersenii , using standard MLST methodology [11] ( data not shown ) . Each of the 14 gene fragments were amplified by PCR , purified and sequenced using a MegaBACE 500 sequencer and DYEnamic ET Dye Terminator Cycle Sequencing Kit ( Amersham Biosciences , England ) . Seven loci were then selected based on performance of primers , number of alleles at a given locus and distribution of strain numbers between the alleles . These loci were pntA , sucA , fadD , tpiA , pfkB , mreA , & glmU , which are located on chromosome I with the exception of fadD . Primer sequences are shown in Table 1 . Amplifications were performed in 25-µl total volumes of PCR reaction mix contained 1–10 ng of genomic DNA , 5 pmol of each primer , 200 µM dNTP , ( eppendorf , Germany ) , 1 . 5 mM of MgCl2 , 1 . 25 unit of Taq DNA polymerase ( Promega , USA ) and 1× buffer . A PTC-200 Peltier Thermal Cycler ( MJ research , USA ) was used to perform PCR with an initial denature step at 94°C for 5 minutes , followed by 30 cycles of 94°C for 10 seconds , 52°C ( mreA , pfkB , pntA , sucA , and tpiA ) , or 50°C ( fadD and glmU ) for 15 seconds , 72°C for 50 seconds , then 72°C for 7 minutes . PCR product size ranged from 555 bp to 638 bp; the sequence start and end points used to define each MLST locus are shown in Table 1 . MLST was performed for the remaining isolates using these 7 loci . Following the standard MLST protocol , each allele was assigned a different allele number and the allelic profile ( string of seven integers ) was used to define the sequence type ( ST ) . A leptospira mlst website was established to provide public access to these data , and to provide a resource to other investigators who can use this to assign the ST of further strains . This can be accessed at http://leptospira . mlst . net . DNA sequences for the 16S rRNA gene have been deposited in the GenBank database with the accession numbers shown in Table S1 .
The number of leptospirosis cases reported to the Department of Disease Control , Thailand between 1990 and 2005 is shown in Figure 1 . An increase in cases of leptospirosis was also observed by clinicians working in northeast Thailand during 1999 ( personal communication , Dr R . Limaiboon , Udon Thani Hospital ) . A prospective study was commenced at Udon Thani Hospital in mid-October 2000 to identify and culture suspected cases and isolate the causative Leptospira . A total of 1 , 658 patients were recruited in Udon Thani , of whom 115 were culture positive for Leptospira . The number of cases of culture proven leptospirosis was greatest during 2001 ( there were only 2 study months during 2000 ) , followed by a decline to the end of the study in 2005 ( Figure 1 ) . There was a significant reduction over time in the proportion of patients presenting with fever who were leptospiraemic ( chi-squared for trend = 15 . 3 , p<0 . 0001 ) . This case load pattern mirrors the number of cases reported to the Department of Disease Control . Our data provides additional confirmatory evidence for a true increase in leptospirosis in northeast Thailand during the putative outbreak . The pathogenic strains of Leptospira obtained from patients presenting to Udon Thani hospital were characterized to determine whether the increased disease incidence was related to one or a number of different circulating bacterial clones . Of the 115 isolates obtained from patients in Udon Thani , 104 were available for molecular characterization . 16S rRNA sequencing was performed on at least one representative of each ST . One hundred isolates were L . interrogans , 3 were L . borgpetersenii and 1 was L . kirschneri ( Table S1 ) . Three strains ( L . borpetersenii serovar Javanica ) failed to amplify at five or six MLST loci but were identical to each other at glmU; these strains are not considered further . The 101 isolates from Udon Thani corresponded to 12 STs but a single ST predominated , with ST34 accounting for 77 ( 76% ) , all of which were identified as serovar Autumnalis . Of the remainder , 8 isolates belonged to ST46 , 4 isolates were ST49 , and the remaining nine sequence types consisted of one or two isolates ( Figure 2a , Table S1 ) . A single isolate defined as ST41 was also serovar Autumnalis but this strain is unrelated to ST34 , showing divergence at all seven alleles . Thus two strains sharing the same serovar can be distantly related [1] , possibly because serovars may arise independently in differently lineages by evolutionary convergence , or by horizontal gene transfer . To explore the role of the dominant clone ST34 in the putative outbreak of leptospirosis , the proportion of Leptospira ST34 was determined for each year of the study in Udon Thani ( Figure 1 ) . This demonstrated that the dominance of ST34 declined over time , being replaced by a range of other sequence types ( Table 1 ) . The proportion of clinical Leptospira isolates that were ST34 fell from 85% in 2000/2001 to 64% in 2002/2003 and 56% in 2004/2005 ( years combined because of small numbers in 2004 and 2005 - χ2 for trend = 6 . 61 , p = 0 . 01 ) . To define the extent to which ST34 was distributed across Thailand , a further 24 unselected isolates obtained in 2003 and 2004 from human cases of leptospirosis from across the country were evaluated . The total proportion of isolates corresponding to ST34 was 17/24 ( 71% ) ( two strains were non-typable L . borpetersenii ) . The geographic distribution was as follows: Lumpang , 1/2 isolates; Rayong , 1/1 isolate; Chumphon , 2/4 isolates; Loei , 9/9 isolates; Ratchaburi , 1/1 isolate; Yasothon , 1/1 isolate; Nakon Ratchasima , 2/5 isolates; and Maha Sarakham , 0/1 isolate . This is not significantly different from the proportion of ST34 in isolates from Udon Thani in the same years ( Fisher's exact p = 0 . 37 ) , and confirms that the outbreak clone ST34 was widely distributed throughout Thailand and formed the predominant virulent strain at the time of the outbreak . A further six STs were identified in this collection , four of which were not observed in the Udon Thani collection . These data provide further support for the picture of a single dominant clone ( ST34 ) associated with an increased incidence of human disease , within a “background” population of higher genotypic diversity . One strain ( ST22 obtained in Lumpang province ) was serovar Autumnalis , but showed divergence at 5/7 alleles from ST34 . To determine whether a link could be identified between ST34 and a maintenance host , 8 isolates available from rodents captured in northeast Thailand were characterized . Seven strains ( from B . indica ( 6 ) and B . savilei ( 1 ) ) , were L . interrogans ST34 . This confirms the predominance of the outbreak strain in a maintenance host , which in this case appears to be the bandicoot rat . The remaining isolate from R . rattus was L . interrogans , ST49 which was also isolated from human cases in Udon Thani in 2001/2 ( n = 4 ) and Nakhon Ratchasima in 2004 ( n = 1 ) . This does not exclude the possibility of additional maintenance hosts , but rodents trapped in agricultural areas reflect the species to which farmers are commonly exposed . To place the Thai isolates within a global context , we selected a total of 76 reference strains representative of the species of the Leptospira strain population in Thailand but recovered from diverse geographical sources ( L . interrogans 65 , L . borgpetersenii 3 , L . kirschneri 8 ) ( Table S1 ) . From our Thai sample of 123 clinical isolates , 16 STs were identified ( 0 . 13 ST per isolate; 5 strains of L . borgpetersenii being non-typable by MLST ) . In contrast , MLST revealed 59 STs for 73 reference strains ( 0 . 81 ST per strain ) , revealing that the reference strains are far more diverse , and that only a small fraction of the global diversity was recovered in the Thai sample . The reference L . borgpetersenii strains did not amplify at all seven loci , and so are again scored as non-typable . The largest clones within the reference collection were ST17 and ST37 ( both with 4 isolates ) ; one ST contained 3 isolates , 7 STs contained two isolates and the remaining 49 sequence types had one representative strain ( Figure 2a , Table S1 ) . ST34 was not represented in the reference collection . One strain was serovar Autumnalis ( Akiyami A , ST27 ) but this was unrelated to ST34 and was much more similar to the non-ST34 Autumnalis strain isolated in Udon Thani ( ST41 ) . This analysis indicates that the strains causing human disease in Thailand are more clonally restricted than reference strains from variable hosts and geographical locations , and that the population genetic structure of L . interrogans is highly diverse when considering non-ST34 isolates . This further supports the argument that the predominance of ST34 during the Thai outbreak does not reflect a clonal population structure , and is consistent with a temporary selective advantage . A phylogenetic analysis was performed to shed light on the emergence of ST34 . All 204 typable strains ( excluding 8 non-typable L . borgpetersenii isolates ) were evaluated to identify the close relatives of ST34 . Figure 2 shows two neighbour-joining trees based on the concatenated sequences of the seven MLST genes ( 3165-bp ) . Figure 2a was constructed using all 204 isolates . There was a clear distinction between the two species L . interrogans and L . kirschnerii which was also noted in loci individually ( not shown ) . ST34 isolates accounted for almost half of the tree , illustrating the numerical dominance of this clone . As the branching order of this tree is unclear , Figure 2b shows a neighbour joining tree for just the STs highlighted in Figure 2a . Of the four Autumnalis STs , ST27 and ST41 appear closely related in both Figure 2a and 2b , but unrelated to the other Autumnalis STs ST22 and ST34 . This latter pair appears to be closely related in Figure 2a , but Figure 2b reveals this is an artifact of the poorly resolved topology of this tree . The different clones sampled from Thailand in this study did not form a single cluster but were dispersed throughout the tree . This suggests that they have not all diverged from a single common Thai ancestor . The lack of evidence for strong geographical structure is consistent with high rates of migration via the rodent ( or possibly human ) host . Figure 2b identifies ST29 ( reference strain Bangkinang 1 ) as a close relative of ST34; this was isolated from a human in Indonesia . Other close relatives of ST34 are also reference strains from Indonesia and Malaysia ( not shown ) , although the significance of this is unclear as the tree is not robustly supported .
Human outbreaks of leptospirosis are well documented in the literature , as are clusters of cases linked by specific water-related activities or occupations [1] , [2] . Outbreaks in Thailand and elsewhere are often linked to climatic events such as flooding and the concomitant increase in human exposure to environments contaminated by Leptospira . The precipitous increase in reported cases of leptospirosis in Thailand commencing in 1999 , followed by the sustained incidence during the ensuing years , could not be explained by persistent climatic change or sequential episodes of regional flooding . Changes in reporting practice can lead to marked changes in the perceived disease incidence , although this does not explain the marked rise and fall in reported cases over time . An alternative explanation is that this was associated with the presence of a biologically successful clone of pathogenic Leptospira . In this study , we developed and applied robust typing methods to provide several lines of evidence in support of this hypothesis . This clone is likely to harbour an adaptive ( competitive ) advantage , albeit transiently . Possible explanations include a selective advantage for ST34 in the maintenance host ( the bandicoot rat ) leading to a higher bacterial load and higher shedding from urine , or a survival advantage once shed into environment , such as increased resistance to desiccation . Both possibilities are amenable to testing in the laboratory setting . Alternatively , ST34 may have a greater propensity to cause human disease compared with other circulating clones . Although difficult to test , the finding that ST34 co-existed in the environment with a large number of other STs but caused most disease would be supportive of this hypothesis . The virulence of ST34 as reflected by severity of human disease was not assessed in patients presenting to Udon Thani hospital , since the comparator group was small and caused by 11 other STs . The emergence of ST34 may have predated the outbreak , and this is difficult to refute since no strains were available from the period prior to the outbreak . However , the decline in frequency of ST34 as a cause of leptospirosis over time is consistent with the suggestion that there is a direct link between the clone and the outbreak . Previous studies of human outbreaks have largely relied on serological methods to confirm clinical cases and to define indirectly the infecting isolate [1] . The standard serological method ( microscopic agglutination test , MAT ) provides a broad idea of serogroups responsible for leptospirosis in a given geographic area , but in one study the predominant serogroups at a titer of ≥100 correctly predicted less than 50% of serovars [13] . Arbitrarily primed PCR has been used successfully to study human outbreaks in Brazil [6] , and to characterize 40 isolates recovered from humans between 1995 and 2001 on the Andaman and Nicobar Islands in India , 32 of which were a clone with a fingerprint matching that of L . interrogans sensu stricto [14] . Here , we use the more discriminatory and robust method of MLST to identify clusters of closely related isolates . The use of multiple gene loci is essential , as frequent recombination within the population would make inferences based on single gene loci unreliable [15] . This study clearly demonstrates the advantages of bacterial isolation in that it permits detailed typing studies to characterize local populations and outbreaks . The MLST scheme presented here was developed primarily to characterize the isolates responsible for the outbreak of leptospirosis unfolding in Thailand in the early 2000s ( i . e . L . interrogans and the closely related L . L . kirschneri ) , and is not designed for the characterization of the genus as a whole . Nevertheless , the scheme presented here demonstrates the utility of MLST for Leptospira for characterizing isolates from a clinical perspective . For more taxonomic or genus-wide evolutionary studies , or for disease caused by other Leptospira species , the primer sequences could be refined in order to broaden the phylogenetic range over which they amplify , or alternatively the loci used by Ahmed et al . may be employed [16] . In conclusion , our observations provide strong support for the hypothesis that the ST34 clone was associated with the 1998–2003 outbreak of leptospirosis in northeast Thailand . The existence of this strain collection now provides a unique opportunity to study the basis for pathogenicity and disease acquisition . | A sustained outbreak of human leptospirosis occurred in northeast Thailand between 1999 and 2003 , the basis for which was unknown . Leptospirosis is a potentially serious infection cause by bacteria known as Leptospira; infection usually occurs following environmental exposure to pathogenic Leptospira shed in the urine of an infected animal . The purpose of this study was to obtain bacterial isolates from humans with leptospirosis around the time of the Thai outbreak for genotyping , and to relate these to the maintenance host animal . To achieve this , a bacterial typing scheme ( multilocus sequence typing , MLST ) was developed for L . interrogans , the major cause of human disease . This approach has the advantage over existing typing schemes in that the data generated are amenable to detailed evolutionary analysis , and are readily comparable via the internet . Our results demonstrated the emergence of a dominant clone of L . interrogans serovar Autumnalis; this was the major cause of human disease during the outbreak , and was found in a maintenance host which was defined as the bandicoot rat . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/bacterial",
"infections"
] | 2007 | A Dominant Clone of Leptospira interrogans Associated with an Outbreak of Human Leptospirosis in Thailand |
The power of forward genetics in yeast is the foundation on which the field of autophagy research firmly stands . Complementary work on autophagy in higher eukaryotes has revealed both the deep conservation of this process , as well as novel mechanisms by which autophagy is regulated in the context of development , immunity , and neuronal homeostasis . The recent emergence of new clustered regularly interspaced palindromic repeats/CRISPR-associated protein 9 ( CRISPR/Cas9 ) -based technologies has begun facilitating efforts to define novel autophagy factors and pathways by forward genetic screening in mammalian cells . Here , we set out to develop an expanded toolkit of autophagy reporters amenable to CRISPR/Cas9 screening . Genome-wide screening of our reporters in mammalian cells recovered virtually all known autophagy-related ( ATG ) factors as well as previously uncharacterized factors , including vacuolar protein sorting 37 homolog A ( VPS37A ) , transmembrane protein 251 ( TMEM251 ) , amyotrophic lateral sclerosis 2 ( ALS2 ) , and TMEM41B . To validate this data set , we used quantitative microscopy and biochemical analyses to show that 1 novel hit , TMEM41B , is required for phagophore maturation . TMEM41B is an integral endoplasmic reticulum ( ER ) membrane protein distantly related to the established autophagy factor vacuole membrane protein 1 ( VMP1 ) , and our data show that these two factors play related , albeit not fully overlapping , roles in autophagosome biogenesis . In sum , our work uncovers new ATG factors , reveals a malleable network of autophagy receptor genetic interactions , and provides a valuable resource ( http://crispr . deniclab . com ) for further mining of novel autophagy mechanisms .
Autophagy is an umbrella term for a broad family of trafficking pathways that transport cytoplasmic material to the lysosome for destruction . Its most studied form , macroautophagy ( hereafter “autophagy” ) , involves the formation of a double-membrane vesicle ( the autophagosome ) that sequesters cytoplasm and delivers it to the lysosome by vesicle fusion . The first factors identified in autophagosome biogenesis , the so-called autophagy-related ( ATG ) factors , were originally identified by genetic screens in yeast [1–3] . This list eventually matured to its current state that includes approximately 40 yeast ATGs and distinguishes “core” factors required for all forms of autophagy from those required only for specific substrates ( e . g . , a damaged mitochondrion ) [4] . The inventory of factors involved in autophagy has been further expanded by the discovery of additional autophagy factors not found in yeast , such as ATG101 , EPG5 , and vacuole membrane protein 1 ( VMP1 ) [5–8] . Whether additional autophagy factors have yet to be discovered remains an open question . The molecular functions of ATGs in autophagosome biogenesis are still being delineated . Initiation involves construction of a cup-shaped precursor membrane ( the phagophore ) in close apposition to the endoplasmic reticulum ( ER ) [9–13] . This process is dependent on the RB1 inducible Coiled-Coil 1 ( RB1CC1 ) scaffolding protein and post-Golgi vesicles containing ATG9A [14–16] . Next , recruitment of an autophagy-specific phosphoinositide-3 kinase ( PI3K ) complex results in localized phosphatidylinositol-3 phosphate ( PI3P ) formation and recruitment of PI3P effectors such as WD repeat domain , phosphoinositide interacting 2 ( WIPI2 ) [17 , 18] . Nascent phagophores are then elongated concurrent with lipid conjugation of microtubule-associated protein 1 light chain 3B ( LC3 ) family proteins ( autophagy-related protein 8 [Atg8] in yeast ) , although the requirement for LC3 lipidation has recently been challenged [19–21] . Efficient encapsulation of specific cell material within autophagosomes is mediated by autophagy receptors , adaptor proteins whose defining feature is their ability to bridge cargo with lipidated LC3 present on the autophagosomal membrane [22] . Interactions between cargos , receptors , and lipidated LC3 leads to their mutual capture by autophagosomes and subsequent lysosomal destruction . This feature of LC3 has been exploited to establish the most widely used reporter of autophagic flux: tandem-fluorescent ( tf ) LC3 ( tfLC3 ) . In this approach , 2 fluorescent protein tags ( red fluorescent protein [RFP] and green fluorescent protein [GFP] ) are appended to the amino terminus of LC3 . Upon lysosomal delivery of tfLC3 , low pH conditions within the lysosomal lumen selectively quench GFP fluorescence , resulting in a dramatic increase in the observed Red:Green fluorescence ratio [23 , 24] . In a similar manner , tf sequestosome 1 ( SQSTM1 ) has been used to measure the flux of this heavily studied autophagy receptor [24] . Here , we further expanded this approach to include 3 additional SQSTM1-like receptors ( nuclear dot protein 52 [NDP52] , tax1 binding protein 1 [TAX1BP1] , and neighbor of BRCA1 gene 1 [NBR1] ) broadly implicated in selective autophagy . With this expanded toolkit in hand , we performed genome-wide clustered regularly interspaced palindromic repeats ( CRISPR ) screens to identify selective autophagy factors in mammalian cells . We recovered virtually all known ATG factors and uncovered several uncharacterized proteins for further study . We validated our list using quantitative cell microscopy and biochemical analyses to define transmembrane protein 41B ( TMEM41B ) as an integral ER membrane protein that is required for phagophore maturation in higher eukaryotes . TMEM41B shares a broadly conserved transmembrane domain ( pfam09335 ) with an established autophagy factor , VMP1 , and our data show that these 2 factors play related , though not fully overlapping , roles in autophagosome biogenesis .
K562 cells ( a human myelogenous leukemia line ) can be readily cultured in cell suspension , which has led to their frequent use in pooled CRISPR screens . To devise a general strategy for monitoring autophagic flux in K562 cells , we began by constructing 6 homologous gene cassettes: 1 ( tfLC3 ) encoding an N-terminal , tf ( RFP-GFP ) fusion with LC3B , 4 ( tfSQSTM1 , tfNDP52 , tfTAX1BP1 , and tfNBR1 ) with N-terminal fusions to SQSTM1-related autophagy receptors , and 1 ( tfEmpty ) as a negative control ( S1A Fig ) . The RFP to GFP fluorescence ratio ( Red:Green ratio ) of tfLC3 is a widely used metric for autophagic flux predicated on the selective quenching of GFP in low pH environments , such as the lysosomal lumen ( Fig 1A ) . Following cassette integration at the adeno-associated virus integration site 1 ( AAVS1 ) locus , we analyzed cells by flow cytometry under basal conditions and following treatment with either Bafilomycin A1 ( BafA1 ) , an inhibitor of autophagosome–lysosome fusion , or torin , a small-molecule inducer of autophagy . Corresponding changes in the observed Red:Green ratio of tfLC3 and tfReceptors provided strong evidence for both basal and induced autophagy in K562 cells ( Fig 1B ) . By contrast , the Red:Green ratio of tfEmpty cells was unresponsive to either drug treatment , arguing that nonselective reporter engulfment makes a negligible contribution to our flux measurements . We obtained similar results using adherent human embryonic kidney HEK293T cells ( S1B Fig ) . Furthermore , we confirmed that torin-induced increases in Red:Green ratio were due to selective GFP quenching ( S1C Fig ) and that both basal and induced autophagy were dependent on the known autophagy-related factors RB1CC1 and ATG13 ( S1D and S1E Fig ) . Collectively , these data validated a broad panel of autophagic flux reporters for subsequent use as genetic screening tools in K562 cells . Next , we performed a series of genome-wide , pooled CRISPR knockout screens in K562 cells co-expressing CRISPR-associated protein 9 ( Cas9 ) with each of our autophagic flux reporters . To this end , we utilized the Brunello single guide RNA ( sgRNA ) library containing 76 , 441 sgRNAs covering 19 , 114 genes [25] . Following transduction of the sgRNA library , we used fluorescence activated cell sorting ( FACS ) to collect the top and bottom third of cells ranked on the basis of their Red:Green ratio ( Fig 2A ) . Read counts of sgRNAs derived from these cell fractions were obtained by Illumina sequencing and computationally analyzed by model-based analysis of genome-wide CRISPR-Cas9 knockout ( MAGeCK ) [26 , 27] . The resulting output for each gene includes a beta score ( similar to log-fold change ) as a proxy for its strength as an autophagy effector ( Fig 2B ) . To facilitate public mining of these data , we have made them freely available using an interactive interface at http://crispr . deniclab . com . From the global analysis of gene hits ( Fig 2B ) , we made multiple observations that validated our genetic screening approach . First , virtually all known ATG factors were identified as required for reporter flux . Second , we observed the expected phenotype for positive and negative regulators of mammalian target of rapamycin complex 1 ( mTORC1 ) , a potent inhibitor of autophagy signaling . Third , we could distinguish protein complexes implicated in autophagy from related complexes with shared subunits . For example , our analysis distinguished the autophagic role of the homotypic fusion and vacuole protein sorting ( HOPS ) tethering complex from the related class C core vacuole/endosome tethering ( CORVET ) complex . Similarly , hits in the BLOC-one-related complex ( BORC ) disrupted autophagy , but we found no hits among subunits specific to the related biogenesis of lysosome-related organelles complex ( BLOC-1 ) ( Fig 2B ) . Our screening approach also identified several uncharacterized factors as strong modifiers of autophagy . To examine these hits further , we retested them as individual sgRNAs alongside a broad panel of known ATG factors ( Fig 3A; S2A–S2D Fig ) . Indeed , the vast majority of individual sgRNAs increased or decreased flux proportionate to their beta score ranking ( compare Fig 3B and Fig 2B ) . By contrast , we found that sgRNAs targeting uroporphyrinogen decarboxylase ( UROD ) were spurious hits that modified the Red:Green ratio by drastically enhancing RFP fluorescence ( S2E Fig ) . We also confirmed that tfNBR1 flux in K562 cells was largely independent of ATG factors required for LC3 lipidation ( e . g . , ATG7 ) , which agrees with our original screen findings ( Fig 3B ) . ATG7-independent autophagosome formation has been reported in other cell types , and our work suggests that NBR1 might provide a useful new marker for further dissecting this process [28–30] . In sum , our genetic screening strategy and hit validation robustly uncovered known ATG factors and unveiled new candidates for further study . TMEM41B stood out among our uncharacterized hits because of its strong phenotype ( particularly for tfNDP52 and tfTAX1BP1 ) in K562 cells . This protein is predicted to span the membrane 6 times and to carry a di-lysine , C-terminal Golgi-to-ER retrieval signal ( S3A Fig ) [31] . In line with these predictions , an unbiased proteomic analysis of subcellular fractions previously assigned TMEM41B’s residence to the ER [32] . Indeed , when we immunoprecipitated TMEM41B we found that N-terminally tagged , but not C-terminally tagged , TMEM41B co-immunoprecipitated a coat protein I ( COPI ) complex component of Golgi-to-ER vesicles from detergent-solubilized cell lysates ( S3B Fig ) . Furthermore , when we tagged the N-terminus of endogenous TMEM41B with the 11th beta strand of GFP ( GFP11 ) and co-expressed a complementary GFP fragment ( beta strands 1 through 10 [GFP1–10] ) , we observed a reticular GFP fluorescence signal that colocalized with the ER marker calnexin ( S3C Fig ) . To test whether TMEM41B is critical for autophagy in other human cell lines , we knocked out its gene in HEK293T and HCT116 cells , where we observed even more robust effects on autophagy than in our original K562 background . Specifically , immunoblotting ( IB ) analysis of endogenous SQSTM1 , TAX1BP1 , NDP52 , and LC3 revealed their accumulation in the absence of TMEM41B consistent with the possibility of reduced autophagic turnover ( Fig 4A; S3D and S3E Fig ) . Two complementary measures of autophagic flux further support this interpretation . First , treatment of TMEM41BKO cells with BafA1 was unable to induce further accumulation of lipidated LC3 ( LC3-II ) ( Fig 4B ) . Second , tfLC3 and tfSQSTM1 flux were strongly repressed in TMEM41BKO cells , comparable to ATG7KO control cells ( S3F and S3G Fig ) . Taken together , these data argue that autophagic flux is substantially inhibited in the absence of TMEM41B . While autophagy was dramatically impeded in the absence of TMEM41B , it was unclear whether this defect was caused by the inability to properly initiate autophagy , the failure of phagophores to mature into autophagic vesicles , or the failure of mature autophagosomes to properly traffic and/or fuse with lysosomes . To define the stage of autophagosome formation that is arrested in TMEM41BKO cells , we used a series of complementary approaches to systematically probe the autophagosome biogenesis pathway . The accumulation of lipidated LC3 in TMEM41BKO cells is inconsistent with an initiation defect ( Fig 4A ) . To further support this view , we analyzed the kinase activity of Unc-51 like autophagy activating kinase 1 ( ULK1 ) , a component of the RB1CC1 complex that regulates phagophore nucleation [33] . Upon autophagy induction , ULK1 phosphorylates numerous substrates including ATG13 . Consequently , the phosphorylation of serine-318 ( p-S318 ) in ATG13 can serve as a proxy for ULK1 activity [34 , 35] . We found that activation of autophagy by torin stimulated the appearance of p-S318 comparably in wild-type and TMEM41BKO cells ( S4A Fig ) . For comparison , this assay robustly detected loss of ULK1 activity in cells lacking RB1CC1 , a known ULK1 kinase coactivator ( S4A Fig ) . These data argue that the primary role of TMEM41B is downstream of initiation . Next , to determine whether initiated phagophores matured properly in TMEM41BKO cells , we turned to a protease protection assay ( Fig 5A ) [36] . In this assay , properly sealed autophagosomes protect their cargo from proteolysis by an exogenous protease . Cells were treated with BafA1 for 18 h to accumulate autophagosomes prior to lysis by mechanical disruption . As expected , in wild-type extracts , the autophagy receptor NDP52 was protease resistant until membranes were solubilized with Triton X-100 ( a nonionic detergent ) . By contrast , NDP52 was protease sensitive in native extracts derived from TMEM41BKO cells and RB1CC1KO control cells , consistent with a failure to properly form completed autophagosomes ( Fig 5B ) . We obtained similar results when we analyzed protease protection of LC3-II ( S4B and S4C Fig ) . In sum , these data suggest that TMEM41B is required for phagophore maturation . To better define the arrested stage of autophagosome biogenesis in TMEM41BKO cells , we used confocal microscopy to monitor the recruitment of various ATGs associated with specific steps in autophagosome biogenesis . To validate this approach , we monitored the formation of LC3-positive ( LC3+ ) punctae . Consistent with our earlier biochemical approaches ( Fig 4 , Fig 5; S3 Fig , S4 Fig ) , we found that TMEM41BKO cells displayed approximately 5 . 5-fold more LC3+ punctae than wild-type cells under basal conditions ( Fig 6A and 6B ) . We reached a similar conclusion when we measured the abundance of LC3+ structures that colocalized with SQSTM1 as an additional marker of autophagosome biogenesis ( S5A and S5B Fig ) . Furthermore , torin treatment induced an approximately 4-fold increase in the level of LC3+ punctae in wild-type cells while having no effect on TMEM41BKO cells ( Fig 6A and 6B ) . These data are broadly consistent with an inhibition of autophagic flux in TMEM41BKO cells . We next analyzed LC3+ structures for WIPI2 , which associates with PI3P-rich phagophore intermediates but not mature autophagosomes [17 , 37] . We found that the majority of LC3+ structures in TMEM41BKO cells colocalized with WIPI2 ( 79%; n = 5 , 032 LC3+ structures ) while those in wild-type cells overlapped poorly with WIPI2 ( 8%; n = 1 , 394 ) ( Fig 6A and 6B ) . Genetic ablation of RB1CC1 abolished accumulation of LC3+ or WIPI2+ structures in TMEM41BKO cells , arguing that these structures represent bona fide intermediates in the process of phagophore maturation ( Fig 6C and 6D; S5C Fig ) . In addition , we visualized syntaxin 17 ( STX17 ) , a SNARE protein that is recruited to late phagophore intermediates prior to vesicle closure [38] . This analysis did not find appreciable amounts of STX17 on the LC3+ structures that accumulate in TMEM41BKO cells ( S5D and S5E Fig ) . For comparison , we recapitulated the published observation that STX17+ structures accumulate in the absence of ATG7 ( S5D and S5E Fig ) [20] . Collectively , our biochemical and microscopy-based data argue that ablation of TMEM41B arrests phagophore maturation during membrane elongation—specifically , after recruitment of LC3 or WIPI2 but prior to WIPI2 dissociation or STX17 recruitment . To substantiate this interpretation , we used transmission electron microscopy ( TEM ) as an alternative method of visualizing arrested autophagosomal intermediates at an ultrastructural level . Analysis by TEM revealed the accumulation of single-membrane vesicles ( approximately 140 nm in diameter ) in TMEM41BKO cells but not in wild-type cells ( S6A Fig ) . However , correlative light and electron microscopy ( CLEM ) demonstrated that fluorescent LC3 punctae did not correlate with these vesicles but rather with a distinct population of smaller ( approximately 50 nm ) vesicles ( Fig 7A ) . Three-dimensional reconstruction of electron miscroscopy tomographs containing LC3+ structures revealed a rudimentary collection of membranous sheets , tubules , and vesicles , reminiscent of immature isolation membranes [15] ( Fig 7B; S6B and S6C Fig ) . For comparison , in wild-type cells , LC3 punctae correlated with structures resembling fully-fledged autophagosomes ( Fig 7C ) . To gain insight into the biochemical function of TMEM41B , we focused on its broadly conserved transmembrane domain ( pfam09335 ) [39] ( S3A Fig; S7A Fig ) . VMP1 is another broadly conserved ER-membrane protein that also contains a pfam09355 domain and is required for an early stage of autophagy [6 , 7 , 14 , 15 , 40] . On the basis of these similarities , we hypothesized that TMEM41B and VMP1 have related activities . To examine this possibility , we looked for increased association of certain ATG factors with ER membranes in the absence of TMEM41B , a hallmark of arrested autophagosome biogenesis in VMP1KO cells [41] . Indeed , following differential centrifugation of cell lysates ( Fig 8A ) , we observed a shift in the migration of WIPI2 , RB1CC1 , and ATG9A from the high-speed pellet ( p100 ) to lower-speed pellets containing ER-derived membranes ( p3 and p20 ) in both VMP1KO and TMEM41BKO cells ( Fig 8B ) . Importantly , we could suppress this effect on WIPI2 and ATG9A fractionation in TMEM41BKO cells by genetically ablating RB1CC1 , highlighting that this is an autophagy-dependent effect . To additionally test the hypothesis that TMEM41B and VMP1 have related activities , we performed a genetic complementation analysis . Specifically , we used our tf gene cassette system to overexpress tagged versions of TMEM41B and VMP1 and evaluated suppression of the TMEM41BKO phenotype by 2 assays: monitoring receptor accumulation by IB and cargo encapsulation by protease protection . Both tfTMEM41B and tfVMP1 ( approximately 15-fold over-expressed [S7B Fig] ) complemented TMEM41BKO phenotypes ( Fig 8C–8E ) . By contrast , only tfVMP1 was able to suppress VMP1KO phenotypes . Taken together , these data strongly argue that VMP1 and TMEM41B have partially overlapping roles during phagophore maturation .
The power of forward genetics in yeast is the foundation on which the field of autophagy research firmly stands [1–3 , 42–46] . Complementary work on autophagy in higher eukaryotes has revealed both the deep conservation of this process , as well as novel mechanisms by which it is regulated in the context of development , immunity , and neuronal homeostasis . Much of this diversification is enabled by autophagy receptors , proteins that connect diverse cellular targets with the core autophagy machinery [47] . However , the potential of these receptors as genetic handles for identifying new factors and adaptations of human autophagy has only begun to be explored . CRISPR-based screens have recently demonstrated their power over previous technologies in identifying novel mammalian autophagy factors and pathways [48 , 49] . Here , we combined CRISPR screening with an expanded toolkit of tf reporters to identify novel factors required for mammalian autophagy . These data are freely available in their entirety as an interactive resource at http://crispr . deniclab . com . Our screens recovered virtually all known ATG factors as well as uncharacterized factors , including TMEM41B . In follow-up studies , we showed that TMEM41B is an ER integral membrane protein critical for autophagic flux in multiple human cell lines . Using complementary biochemical and fluorescence-microscopy–based assays , we detected the accumulation of unsealed , LC3+ , WIPI2-positive autophagy intermediates in cells ablated for TMEM41B ( Figs 5 and 6 ) . Furthermore , our ultrastructural analysis by CLEM found that these intermediates correspond to a collection of thin ( approximately 50 nm ) membranous sheets , tubules , and vesicles , which are hallmarks of immature isolation membranes ( Fig 7 ) [15] . TMEM41B contains a conserved transmembrane domain ( pfam09335 ) also found in VMP1 , a factor previously known to regulate phagophore maturation [15 , 41 , 50] . Correspondingly , overexpression of VMP1 restored autophagosome formation in TMEM41BKO cells ( Fig 8 ) , consistent with the possibility that TMEM41B and VMP1 may be structurally and/or functionally related . VMP1 is absent in yeast but is generally conserved in higher eukaryotes . In Dictyostelium and other model eukaryotes , cells lacking VMP1 display diverse phenotypes , including defects in autophagy , membrane contact sites , ion homeostasis , lipid metabolism , and phosphoinositide distribution [6 , 40 , 41 , 50–52] . In light of such pleotropic effects , it remains unclear whether the autophagy defects observed in cells lacking VMP1 or TMEM41B are direct or sequelae of a broader defect in ER-organelle dynamics . During revision of this manuscript , 2 groups independently reported TMEM41B as a novel modifier of mammalian autophagy [53 , 54] . Consistent with the results herein , both observed similarities between TMEM41B and VMP1 deletions . On the basis of these phenotypic similarities , one immediate hypothesis is that TMEM41B and VMP1 function as a complex . Indeed , Morita and colleagues were able to isolate a VMP1/TMEM41B-containing co-complex in the presence of n-dodecyl-β-D-maltoside/cholesteryl hemisuccinate ( DDM/CHS ) [53] , although such a complex was not observed in other detergents [54 , 55] . It remains an important future goal to establish whether TMEM41B’s function is dependent on its interaction with VMP1 . A potential clue comes from studies of DedA proteins in Escherichia coli , which share pfam09335 with TMEM41B and VMP1 [56] and are distantly related to bacterial LeuT transporters [57] . E . coli cells lacking the DedA homologs YqjA and YghB have a defect in proton import that can be rescued by overexpressing a Na+/K+-H+ antiporter [58 , 59] . Future biochemical and structural studies of VMP1 and TMEM41B will provide new tools for testing their potentially conserved role in ion homeostasis . A deeper understanding of autophagy has the potential to provide therapeutic insight into numerous pathologies , including neurodegenerative disease , cardiometabolic disease , and cancer [60–62] . TMEM41B , for example , has been implicated in diverse pathologies , including spinal muscular atrophy [63] , pulmonary carcinoid tumors [64] , and coronavirus infection [65] . Our work argues that autophagy should be further examined as a potential etiological factor in these contexts . The identification of TMEM41B as a novel ATG factor is one facet of this work , but other aspects of our data remain to be explored . These include novel inducers of autophagy , receptor-specific modulators , and additional putative ATG factors . Vacuolar protein sorting 37 homolog A ( VPS37A ) and VPS25 are 2 factors in the latter category known to assemble into endosomal sorting complexes required for transport ( ESCRT ) . ESCRT components have long been implicated in autophagy and may function at several steps during autophagosome biogenesis , as well as during autophagosome–endolysosomal fusion [66 , 67] . Many ESCRT components are required for cell division , which complicates analysis of their mutant phenotypes . By contrast , VPS37A is nonessential and offers a potential future handle for further dissecting the mechanistic details of ESCRT’s role in mammalian autophagy . In sum , the resource presented herein presents 3 concrete advances . First , it illustrates the robust nature of tf flux reporters as genetic screening tools that can be further applied to other receptors and receptor targets . Second , it yields a rich data set ( http://crispr . deniclab . com ) for further hypothesis testing . And lastly , it provides a new molecular handle ( TMEM41B ) for further dissecting the enigmatic role of the ER membrane in the process of autophagosome biogenesis .
The following antibodies were used in this study . For IB , all primary antibodies were used 1:1 , 000 except where otherwise noted; secondary antibodies were used 1:3 , 000 ( HRP ) or 1:10 , 000 ( fluorescent ) . For immunofluorescence ( IF ) , antibody dilutions are noted following each antibody; secondary antibodies were used 1:500 . Primary antibodies include the following: mouse anti-SQSTM1 ( [1:200 –IF] ab56416 , Abcam , Cambridge , United Kingdom ) , rabbit anti-TAX1BP1 ( #5105 , Cell Signaling Technology , Danvers , MA ) , rabbit anti-NDP52 ( #9036 , Cell Signaling Technology ) , rabbit anti-LC3B ( [1:2 , 000 –IB] NB100-2220 , Novus Biologicals , Centennial , CO ) , rabbit anti-LC3A/B ( [1:100 –IF] #12741S , Cell Signaling Technology ) , rat anti-tubulin ( [1:500 –IB] sc-53030 , Santa Cruz , Dallas , TX ) , anti-p70-S6K ( #9202S , Cell Signaling Technology ) , anti-p70-S6K ( pT389 ) ( #9205S , Cell Signaling Technology ) , rabbit anti-ATG13 ( ABC344 , MilliporeSigma , Burlington , MA ) , rabbit anti-ATG13 ( p-S318 ) ( NBP2-19127 , Novus Biologicals , Centennial , CO ) , anti-WIPI2 ( [1:150 –IF] ab105459 , Abcam ) , rabbit anti-ATG7 ( #8558 , Cell Signaling Technology ) , anti-RB1CC1 ( #12436S , Cell Signaling Technology , Danvers , MA ) , mouse anti-GFP ( #11814460001 , MilliporeSigma ) , mouse anti-SERCA2 ( ab2861 , Abcam ) , and rabbit anti-Calnexin ( #2679 , Cell Signaling Technology ) . Secondary antibodies for IB include the following: goat anti-mouse IgG HRP ( #170–6516 , Bio-Rad , Hercules , CA ) , goat anti-rabbit IgG HRP ( #170–6515 , Bio-Rad , Hercules , CA ) , goat anti-rat IgG Alexa Fluor 488 ( A11006 , Thermo Fisher Scientific , Waltham , MA ) , goat anti-mouse IgG Cy5 ( A10524 , Thermo Fisher Scientific ) , and goat anti-rabbit IgG Alexa Fluor 546 ( A11010 , Thermo Fisher Scientific ) . Secondary antibodies ( for IF ) include the following: goat anti-rabbit IgG Alexa Fluor 568 ( A11036 , Thermo Fisher Scientific ) and goat anti-mouse IgG Alexa Fluor Plus 488 ( A32723 , Thermo Fisher Scientific ) . The following chemicals and reagents were used in this study: torin1 ( sc-396760 , Santa Cruz , Dallas , TX ) , BafA1 ( tlrl-baf1 , InvivoGen , San Diego , CA ) , poly-l-lysine ( P4707-50ML , MilliporeSigma ) , Lipofectamine 3000 ( L3000008 , Thermo Fisher Scientific ) , nucleofector kit T ( VACA-1002 , Lonza , Basel , Switzerland ) , polybrene ( H9268-5G , MilliporeSigma ) , normocin ( ant-nr-1 , InvivoGen ) , puromycin ( ant-pr-1 , InvivoGen ) , blasticidin ( ant-bl-1 , InvivoGen ) , zeocin ( ant-zn-1 , InvivoGen , San Diego , CA ) , normal goat serum ( NGS; ab7481 , Abcam , Cambridge , United Kingdom ) , Phusion High-Fidelity DNA polymerase ( M0530L , NEB , Ipswich , MA ) , T5 exonuclease ( M0363S , NEB , Ipswich , MA ) , and Taq DNA ligase ( M0208L , NEB , Ipswich , MA ) . pHR-SFFV-GFP1-10 ( Addgene plasmid # 80409 ) and pCDNA CMV mCherry-GFP-11 were gifts from Bo Huang . pMRXIP GFP-Stx17TM was a gift from Noboru Mizushima ( Addgene plasmid #45910 ) . Human Brunello CRISPR knockout pooled library was a gift from David Root and John Doench ( Addgene #73178 ) . lentiCRISPRv2 puro was a gift from Brett Stringer ( Addgene plasmid #98290 ) . lentiGuide-puro was a gift from Feng Zhang ( Addgene plasmid #52963 ) . ptfLC3 was a gift from Tamotsu Yoshimori ( Addgene plasmid #21074 ) . pX330-U6-Chimeric_BB-CBh-hSpCas9 was a gift from Feng Zhang ( Addgene plasmid #42230 ) . AAVS1-CAG-hrGFP was a gift from Su-Chun Zhang ( Addgene plasmid #52344 ) . psPAX2 was a gift from Didier Trono ( Addgene plasmid #12260 ) . pCMV-VSV-G was a gift from Bob Weinberg ( Addgene plasmid #8454 ) . pUMVC was a gift from Bob Weinberg ( Addgene plasmid #8449 ) . pFUGW-EFSp-Cas9-P2A-Zeo ( pAWp30 ) was a gift from Timothy Lu ( Addgene plasmid #73857 ) . PCR fragments were generated using 2X phusion master mix ( M0531S , NEB , Ipswich , MA ) and insert-specific primers that appended a 30 bp overlap with target DNA . Vector backbones were linearized by restriction enzyme and dephosphorylated by calf intestinal phosphatase ( M0290S , NEB , Ipswich , MA ) . Prior to assembly , all DNA fragments were gel purified ( D4002 , Zymo Research , Irvine , CA ) . Linearized vector DNA ( 50 ng ) was combined with isomolar amounts of purified insert ( s ) ; 5 ul of the resulting DNA mix was added to isothermal assembly master mix and incubated at 50°C for 20 min [68] . Assembled product was transformed into NEB Stable competent cells ( C3040H , NEB , Ipswich , MA ) and plated on LB + agar plates ( plus appropriate antibiotics ) to isolate single isolates . Single isolates were grown in LB broth + antibiotics , and plasmid DNA was purified using a Qiagen miniprep kit ( #27106 , Qiagen , Hilden , Germany ) . Sequences were verified by Sanger sequencing ( Eton Bioscience , San Diego , CA ) . sgRNA oligonucleotides ( oligos ) were ordered from Eton Bioscience ( San Diego , CA ) . Oligo sequences are listed in S1 Table . To generate the necessary overhangs , all oligos were in the form: Forward: 5′- CACCGNNNNNNNNNNNNNNNNNNNN–3′; Reverse: 5′- AAACNNNNNNNNNNNNNNNNNNNNC–3′ . Oligos were diluted to 10 uM in distilled water . A total of 50 pmol each of forward and reverse oligo was combined in a 25 ul reaction and phosphorylated with T4 polynucleotide kinase ( M0201S , NEB , Ipswich , MA ) in 1X T4 DNA ligase buffer ( B0202S , NEB , Ipswich , MA ) for 30 min at 37°C . Phosphorylated oligos were boiled for 5 min at 95°C and slow cooled ( 0 . 1°C/s ) to facilitate annealing . Annealed oligos were diluted 1:100 , and 2 μl of insert was ligated into 20 ng digested vector ( pLentiuGuide-puro , BsmBI site; pLentiCRIPSR version 2 , BbsI site; pX330-derivatives , BbsI site ) using T4 DNA ligase ( M0202S , NEB , Ipswich , MA ) . Ligation was allowed to proceed for 15 min at room temperature . Ligated products were transformed into NEB Stable cells ( NEB , Ipswich , MA ) . Genomic DNA ( gDNA ) was extracted using QuickExtract buffer ( QE0905T , Epicentre , Madison , WI ) according to the manufacturer’s instructions . gDNA was subsequently normalized to 200 ng/μl . Per 100 μl PCR , 600 ng gDNA was used as template to amplify the targeted region of interest . Primers are listed in S2 Table . The resulting PCR amplicon was purified using a Zymoclean Gel DNA Recovery Kit ( D4002 , Zymo Research , Irvine , CA ) and normalized to 20 ng/μl in 19 μl 1X NEBuffer 2 ( B7002S , NEB , Ipswich , MA ) . Amplicon was boiled and cooled ( −1°C/s ) to allow for hybridization . T7 endonuclease I ( M0302 , NEB , Ipswich , MA ) was added at 1 μl per 20 μl reaction and incubated for 15 min at 37°C . Reaction was quenched by adding 1 . 5 μl 0 . 25 M EDTA and was analyzed in a 2% UltraPure Agarose gel ( #16500500 , Thermo Fisher Scientific , Waltham , MA ) . The hrGFP fragment was excised from AAVS1-CAG-hrGFP using SalI/EcoRV and was replaced with RFP-GFP subcloned from ptfLC3 to generate pCS418 tfEmpty ( puro ) . LC3B ( NP_074058 . 2 ) , SQSTM1 ( NP_003891 . 1 ) , NDP52 ( NP_005890 . 2 ) , NBR1 ( NP_006015 . 4 ) , or TAX1BP1 ( NP_005822 . 1 ) was PCR amplified and subcloned into the KpnI site of pCS418 . Primers are listed in S3 Table . To generate blasticidin-resistant versions of each cassette , the puromycin resistance gene was excised from pCS418 with an XhoI/SpeI digest and replaced with an analogous geneblock fragment ( Integrated DNA Technologies , Coralville , IA ) encoding a blasticidin resistance gene ( BSD ) . All cells were grown in a standard water-jacketed incubator with 5% CO2 . K562 cells were grown in IMDM media ( #30–2005 , ATCC , Manassas , VA ) with 10% FBS ( #30–2020 , ATCC , Manassas , VA ) and 1x penicillin/streptomycin . Cells were maintained below 1 million cells per milliliter . HEK293T cells were grown in DMEM media ( #30–2002 , ATCC , Manassas , VA ) with 10% FBS and 1x penicillin/streptomycin . Normocin ( 1:500 ) was used as a common additive . All cells were passaged less than 25 times . For passaging , cells were trypsinized with Trypsin-EDTA ( #25300–054 , Thermo Fisher Scientific , Waltham , MA ) . Puromycin ( 2 μg/ml ) , blasticidin ( 5 μg/ml ) , and zeocin ( 50 μg/ml ) were added when necessary for selection . Hct116 cells were grown in McCoy’s 5a modified media ( #30–2007 , ATCC , Manassas , VA ) with 10% FBS ( #30–2020 , ATCC , Manassas , VA ) and 1x penicillin/streptomycin . gDNA was isolated from HEK293T and K562 cells using the GenElute Mammalian Genomic DNA Miniprep Kits ( MilliporeSigma , Burlington , MA ) . Short tandem repeat ( STR ) profiling and allele identification were performed by the Molecular Diagnostics Laboratory of Dana-Farber Cancer Institute . Briefly , isolated gDNA was analyzed with the GenePrint 10 STR profiling kit ( Promega , Madison , WI ) and Amelogenin for gender identification . GeneMapper version 4 Fragment Analysis software ( Thermo Fisher Scientific , Waltham , MA ) and GenePrint10 allele panel ( Promega , Madison , WI ) custom bin files were used to identify the alleles at 8 STR loci ( TH01 , TPOX , vWA , CSF1PO , D16S539 , D7S820 , D13S317 , and D5S818 ) . The ATCC STR Profile Database was used to verify that the identified alleles matched those of the expected cell type . Prior to transfection , HEK293T cells were seeded in OptiMEM Reduced Serum media ( #51985–034 , Thermo Fisher Scientific , Waltham , MA ) . At 90% confluency , cells were transfected overnight using Lipofectamine 3000 reagent according to the manufacturer’s recommendations . The following morning , media were exchanged , and cells were passaged for another 24 h prior to drug treatment . K562 cells were nucleofected using a Nucleofector 2b Device ( Lonza , Basel , Switzerland ) using nucleofector kit T ( VACA-1002 , Lonza , Basel , Switzerland ) and protocol T-016 . Transfected/nucleofected DNA was prepared using a ZymoPURE Plasmid Midiprep Kit ( D4200 , Zymo Research , Irvine , CA ) . For the generation of stable knockouts , HEK293T and K562 cells were transfected or nucleofected , respectively , as described above . Oligo sequences for sgRNAs were generated by CHOPCHOP or were extracted from the Brunello library and cloned into the indicated vectors as outlined above under “sgRNA oligonucleotide ligation protocol” [25 , 69] . Oligos are listed under S2 Table . Cells were diluted by limiting dilution or cell sorting into 96-well plates and clonally expanded . Expanded cells were confirmed for knockout by western blot or PCR/T7 endonuclease testing . HEK293T cells were transduced with a lentiviral GFP1–10 expression cassette . Clonal cell lines were established by limiting dilution and were validated for GFP1–10 insertion by transient transfection with mCherry-GFP-11 . Four successful clones were saved , and 1 was used for all further experiments . To tag TMEM41B , cells were transfected with pCS651 ( co-expressing Cas9-T2A-BFP , sgTMEM41B ) and oVD6217 ( an oligo containing 5′ homology arm-GFP11-linker-3′ homology arm ) . Successful integrants were identified and sorted by FACS . Cell lines were co-transfected with pX330-U6-Chimeric_BB-CBh-hSpCas9-AAVS1- gRNA and AAVS1-tfReporter template vectors . Forty-eight h post transfection , cells were incubated with appropriate selection media and passaged for 14 d . Red+/Green+ cells were sorted and propagated . Lentivirus was generated in HEK293T cells using Lipofectamine 3000 . Cells were grown overnight in OptiMEM media ( 5% FBS , no antibiotics ) ( #31985062 , Thermo Fisher Scientific , Waltham , MA ) to 90% confluency . Cells were than transfected with pVSV-G , pSPAX2 , and packaging constructs at a 1:3:4 ratio . Transfection proceeded for 6 to 8 h before media were refreshed . Virus was collected and pooled at 24 and 48 h post transfection . Virus was pelleted at 1 , 000 g 2× 10 min , aliquoted , and frozen in single-use aliquots . For retrovirus production , all methods were the same except that pUMVC was exchanged for pSPAX2 . Cells were incubated in appropriate media containing 8 μg/ml polybrene and lacking penicillin/streptomycin . Cells were transduced overnight . Media were exchanged for media lacking polybrene for 24 h prior to antibiotic selection . Our protocol for mammalian protease protection assay was based on Zhao and colleagues [36] . Cells were seeded so they would be 75% confluent at 5 PM . Media were then exchanged into media containing 100 nM BafA1 . Cells were incubated for 15 h . After incubation , cells were trypsinized and pelleted . Cells were washed 1 time with cold PBS and resuspended in prechilled lysis buffer ( 20 mM Hepes KOH [pH 7 . 4] , 0 . 22 M mannitol , 0 . 07 M sucrose ) . Cells were lysed by extrusion through a 26-gauge needle 20 times . Samples were pelleted 2 times at 500 g for 10 min at 4°C to pellet debris . When indicated , samples were incubated with 1X trypsin ( T1426-100MG [MilliporeSigma , Burlington , MA]; 100X stock: 1 mg/ml ) and/or 0 . 5% Triton X-100 for 35 min at 30°C . Reactions were quenched in 1X hot Laemmli sample buffer and held at 65°C for 10 min . Cells were harvested from two to four 15-cm dishes of cells at 95% confluency by scraping and washing them with 5 mL of cold PBS per plate in the cold room . Cells were pelleted at 150 g for 10 min at 4°C . PBS was removed , and cells were resuspended in 3X volume of hypotonic lysis buffer ( 10 mM HEPES , 10 mM KOAc , 1 . 5 mM Mg ( OAc ) 2 , 1X protease inhibitor tablet ) . Cells were incubated on ice for 20 min and then pelleted . Pelleted cells were resuspended in Buffer E ( 2X volume of packed cell pellet; 20 mM HEPES [pH 7 . 4] , 250 mM sucrose , 1 mM EDTA , 1X protease inhibitor tablet ) . Cells were mechanically lysed through a 26-gauge needle using a 1 mL syringe ( prechilled at −20°C ) on ice ( up/down 8 times , letting cells settle in between ) . Lysate was diluted 2X in Buffer E prior to pelleting . Lysate was cleared cell debris and nuclei by pelleting at 1 , 000 g for 10 min at 4°C . Sequential pelleting was performed at 3 , 000 g for 10 min , 20 , 000 g for 20 min , and 100 , 000 g for 60 min . All pellets were resuspended in Buffer E . Cells were collected and resuspended in IP buffer ( 50 mM HEPES [pH 7 . 4] , 150 mM NaCl , 2 mM EDTA , 1% Triton X-100 ) . Cells were incubated on ice for 30 min and pelleted twice at 5 , 000 g for 5 min at 4°C . Supernatant was applied to prewashed GFP-Trap or RFP-Trap magnetic agarose ( Chromotek , Planegg-Martinsried , Germany ) and incubated for 1 h at 4°C . Beads were washed 4× 5 min with 2 tube changes . Protein was eluded by boiling at 70°C in 1X SDS buffer . Cells were trypsinized at 75% confluency and quenched in an equal amount of media . Cells were lysed for 15 min on ice in lysis buffer ( 50 mM HEPES [pH 7 . 4] , 150 mM NaCl , 2 mM EDTA , 1% Triton X-100 , 2X complete protease inhibitor tablet [Roche , Basel , Switzerland] ) . For phosphorylation analysis , lysis buffer was supplemented with phosphatase inhibitors ( 10X: 100 mM NaF , 10 mM Na3VO4 , 100 mM NaPPi ) . Lysates were cleared 2X at 1 , 000 g for 5 min . Post-spin supernatants were used as input . Protein levels in supernatants were normalized using a BCA protein assay ( #23227 , Thermo Fisher Scientific , Waltham , MA ) . Normalized samples were boiled in 1X ( final concentration ) Laemmli Loading Buffer ( 3X stock: 189 mM Tris [pH 6 . 8] , 30% glycerol , 6% SDS , 10% beta-mercaptoethanol , bromophenol blue ) . Gel electrophoresis was performed at 195 V for 70 min in Novex 4%–20% Tris-Glycine gels . Total protein analysis was performed using SYPRO Ruby ( Thermo Fisher Scientific , Waltham , MA ) according to the manufacturer’s recommendations ( short protocol ) . For western blotting , samples were transferred for 60 min to 0 . 2 μm PVDF membranes ( #ISEQ00010 , MilliporeSigma , Burlington , MA ) using a Semi-dry transfer cell ( Bio-Rad , Hercules , CA ) . Membranes were blocked for 20 min in TBS-T with 5% milk . Primary antibodies were incubated overnight at 4°C . Blots were then rinsed 3× 5 min . Secondary antibodies were incubated for 1 h at room temperature . Blots were rinsed 4× 10 min in TBS-T and imaged using fluorescence ( Typhoon Trio Imager , GE Healthcare , Chicago , IL ) or chemiluminescence ( SuperSignal West Femto Maximum Sensitivity Substrate , #34095 , Thermo Fisher Scientific , Waltham , MA ) . If necessary , stripping of membranes was performed using Restore Western Blot Stripping Buffer ( #21059 , Thermo Fisher Scientific , Waltham , MA ) for 10 min . Coverslips ( #12-548A , Thermo Fisher Scientific , Waltham , MA ) were placed in 6-well tissue culture plates ( #62406–161 , VWR , Radnor , PA ) and coated with poly-L-lysine ( #P4707 , MilliporeSigma , Burlington , MA ) per the manufacturer’s recommendations . Cells were seeded onto coverslips overnight so that they would be 15% confluent at the time of fixation . When reported , cells were treated with 250 nM torin for 1 to 3 h prior to fixation . Fresh 16% PFA ( #15710 , Electron Microscopy Sciences , Hatfield , PA ) was diluted to 4% in 1X Dulbecco’s phosphate buffered saline with calcium chloride and magnesium chloride ( #14080–055 , Thermo Fisher Scientific , Waltham , MA ) . Coverslips were removed with forceps and placed into 4% PFA for 15 min . PFA was aspirated and washed twice with PBS ( D8537 , MilliporeSigma , Burlington , MA ) . For LC3 IF , cells were transferred to wells containing prechilled ( −20°C ) methanol for 5 min . Slides were returned to PBS and washed 2× for 5 min . Slides were blocked at RT for 1 h in blocking buffer ( 0 . 3% Triton X-100 , 5% NGS in PBS ) and washed once in PBS . Primary antibody was diluted in 5% NGS at the dilutions described elsewhere in Materials and methods . The amount of 75 ul of antibody mixture was spotted on parafilm in a humidified chamber , and inverted coverslips were incubated with antibody overnight at 4°C . After incubation , coverslips were washed 3× 10 min in PBS . Secondary antibodies were diluted 1:500 in 5% NGS , and inverted coverslips were incubated with antibody mixture for 45 min . Cells were stained with a 1:10 , 000 dilution of Hoechst 33342 ( H3570 , Thermo Fisher Scientific , Waltham , MA ) for 5 min . Coverslips were washed 4× 10 min in PBS and mounted on coverslips ( #294875X25 , Corning , Corning , NY ) using Prolong Diamond ( P36965 , Thermo Fisher Scientific , Waltham , MA ) . Fluorescent images were obtained using a confocal microscope with Airyscan detectors ( LSM880 with Airyscan , Zeiss ) and a 63X PlanAPO oil-immersion objective lens ( Zeiss , Oberkochen , Germany ) and were processed with Zeiss Blue software ( Zeiss , Oberkochen , Germany ) . Fluorescence microscopy images were processed using a newly developed Python analysis pipeline built around the pyto_segmenter analysis package [70] . First , regions of images containing cells were identified . To do so , we first fit a Gaussian distribution to the fluorescence intensity distribution for a smoothed green channel ( 488 nm excitation ) z-stack from an empty field . Using this Gaussian fit , we predicted the probability that each pixel in the smoothed green channel z-stack for each field corresponded to background ( noncell ) or foreground ( cell ) . We assigned each pixel with a p ( background ) < 10−5 to the cells , thus creating a “cell mask . ” After removing small specks ( <100 , 000 pixels volume ) to eliminate debris , we removed out-of-focus planes from the cell mask using a Support Vector Machine ( SVM ) classifier as described previously [71] . Next , nuclei were segmented from the blue ( DAPI ) channel by slice-by-slice relative thresholding followed by watershed segmentation using the pyto_segmenter package . Cells were segmented using watershed segmentation from nuclei seeds . Cell edges were eroded to eliminate blurred edge excess included during the p-value transformation . Cells contacting the edge of the field were removed from analysis . Next , punctae were segmented in the green and red ( 561 nm ) channels using the pyto_segmenter package with empirically determined Canny edge detection thresholds . The number of total punctae and punctae overlapping with objects in the other fluorescence channel were counted , and tabulated data were saved in . csv format . Plotting was performed using R and the ggplot2 package . See the image analysis package for details . Scripts for image analysis and data plotting can be found at https://github . com/deniclab/csth-imaging/tree/pub_version . For microscopy experiments , 40 images were collected , and the number of cells in each sample was counted using segmentation scripts . Cell counts are indicated above each quantitation . Samples were masked prior to data collection and analyzed using automated scripts to eliminate bias during quantification . Replicates represent biological replicates in which strains were subjected to identical preparations on different days . For sequencing experiments , the number of replicates ( 2–4 ) are indicted in S2 Data . Each replicate was a biological replicate in which strains were transfected and taken through the entire experiment on separate days . For flow cytometry experiments , n is indicated for each experiment in the figure legend; >1 , 000 cells were used for each experiment . Brunello library ( 2-vector system ) was purchased from addgene ( item #73178 ) . The amount of 50 ng of library was electroporated into 25 μl Endura electrocompetent cells ( 60242–2 , Lucigen , Middleton , WI ) . Cells from 8 electroporations were pooled and rescued in 8 ml of rescue media for 1 h at 37°C . Eight milliliters of SOC ( 2% tryptone , 0 . 5% yeast extract , 10 mM NaCl , 2 . 5 mM KCl , 10 mM MgCl2 , 10 mM MgSO4 , and 20 mM glucose ) was added to cells , and 200 μl of the final solution was spread onto 10 cm LB plates containing 50 μg/ml carbenicillin ( 80 plates total ) . Through a dilution series , 500 million colonies were estimated , representing 7 , 000X coverage of the library . Cells were manually scraped off plates , and a GenElute Megaprep kit ( NA0600-1KT , MilliporeSigma , Burlington , MA ) was used to purify plasmid DNA . Lentivirus was generated by lipofection ( Lipofectamine 3000 ) of HEK293T cells with 5 μg psPAX2 ( Addgene Plasmid #12260 ) , 1 . 33 μg pCMV-VSV-G ( Addgene plasmid Plasmid #8454 ) , and 4 μg library vector per 10 cm plate . Transfection was performed according to the manufacturer’s specifications . Briefly , low-passage HEK293T cells were grown in OptiMEM + 5% FBS medium to 95% confluency by time of transfection . Cells were transfected for 6 h , and then media were replaced with fresh OptiMEM + 5% FBS . At 24 h post transfection , supernatant was collected and replaced . At 48 h post transfection , supernatant was again collected , pooled with the 24 h supernatant , and clarified 2× 1 , 000 g for 10 min . Viral RNA was purified using a Macherey Nagel viral RNA purification kit ( Macherey-Nagel , Düren , Germany ) . Viral RNA was quantified using the Lenti-X qRT-PCR Titration Kit ( Clontech , Mountain View , CA ) . A value of 849 copies/IFU , derived from a control virus expressing BFP , was used to calculate viral titer . For CRISPR screening experiments , K562 cells were passaged to maintain cell density between 500 , 000 and 2 million cells/ml . Cells were propagated in IMDM + 10% FBS + penicillin/streptomycin + appropriate antibiotics ( blasticidin 5 μg/ml , zeocin 50 μg/ml ) until 200 million cells were obtained ( approximately 8–10 d ) . For infection , 200 million cells were pelleted and resuspended in IMDM + 10% FBS + 8 μg/ml polybrene . Date of infection was day 0 . An MOI of 0 . 4 was used to minimize multiple infection events per cell . Cells were infected overnight , pelleted , and exchanged into fresh media . After 24 h , cells were split , and 2 μg/ml puromycin was added . Cells were continually passaged in puromycin . At day 10 , cells were removed from puromycin , and at day 12 , cells were sorted for Red:Green fluorescence . The amount of 100 M unsorted cells were pelleted and processed as input . The top and bottom 30% of cells ( based on Red:Green ratio ) were taken; 100 million cells were sorted for each experimental condition . Cell sorting was performed using a FACSAria ( Becton Dickinson , Franklin Lakes , New Jersey ) or BioRad S3 ( Bio-Rad , Hercules , CA ) sorter . Cells were pelleted and stored at −80°C until processing . gDNA was purified from collected cells using the NucleoSpin Blood XL kit ( #740950 . 1 , Macherey-Nagel , Düren , Germany ) according to the manufacturer’s instructions . Illumina sequencing libraries were created by PCR amplifying the genomically integrated sgRNA sequences . All gDNA was used for each PCR . A pool of 8 staggered-length forward primers was used in each PCR reaction to avoid monotemplating during Illumina sequencing . Reverse primers contained unique barcodes designed to allow for sequencing and differentiation of multiple libraries on a single chip during Illumina sequencing . Each 50 μL PCR reaction contained 0 . 4 μM of each forward and reverse primer mix ( Integrated DNA Technologies ) , 1X Phusion HF Reaction Buffer ( NEB , Ipswich , MA ) , 0 . 2 mM dNTPs ( NEB , Ipswich , MA ) , 40 U/mL Phusion HF DNA Polymerase ( NEB , Ipswich , MA ) , 5 μg of gDNA , and 3% v/v DMSO . The following PCR cycling conditions were used: 1X 98°C for 30 s; 25X ( 98°C for 30 s , 56°C for 30 s , and 72°C for 30 s ) ; and 1X 72°C for 10 min . The resulting products were pooled to obtain the sgRNA libraries . The pooled PCR products were size selected by adding 0 . 95X magnetic bead slurry as outlined by DeAngelis and colleagues [72] . The High Sensitivity D1000 ScreenTape system ( Agilent Technologies , Santa Clara , CA ) was used to confirm the absence of primer dimers after purification . Sample libraries were quantified by qPCR using the NEBNext Library Quant Kit for Illumina ( NEB , Ipswich , MA ) . Four to five libraries were pooled to a total concentration of 10 nM for simultaneous Illumina sequencing on a single chip . The libraries were sequenced using either the HiSeq 2000 or 2500 system ( Illumina ) . The HiSeq 2000 system was run on High Output Run Mode with the TruSeq SBS version 3 kit ( Illumina , San Diego , CA ) . The HiSeq 2500 system was run on either the Rapid Run Mode with the HiSeq Rapid SBS version 2 kit ( Illumina , San Diego , CA ) or the High Output Run Mode with the HiSeq SBS version 4 kit ( Illumina , San Diego , CA ) . Sequencing was performed per recommendations of the manufacturer with custom sequencing and indexing primers ( Integrated DNA Technologies , Coralville , IA ) . For primer sequences , see S4 Table . The 5′ end of NGS reads were trimmed to 5′-CACCG-3′ using Cutadapt . The count function of MAGeCK ( version 0 . 5 . 3 ) was used to extract read counts for each sgRNA . Raw read counts can be found in S2 Data . The mle function was used to compare read counts from cells displaying increased and decreased Red:Green ratios . The output included both beta scores and false discovery rates . All beta scores can be found in S3 Data . Beta scores for each sgRNA for each tfReporter were averaged across 2 to 4 experiments . Across all experiments , average read counts were 200 to 400 per sgRNA . To generate heat maps for each reporter ( e . g . , Fig 2 ) , the beta scores for each gene were normalized by the beta score for ATG9A . All samples were pelleted , washed 1× in cold PBS , and filtered through strainer cap tubes ( 21008–948 , VWR , Radnor , PA ) prior to analysis . Flow cytometry data were collected on an LSRII flow cytometer . Data were analyzed in FlowJo ( FlowJo LLC , Ashland , Oregon ) and R . The biomodality of populations was determined in an unbiased manner using the BifurGate tool in FlowJo . At least 1 , 000 cells were collected for all samples . For CLEM , HEK293T TMEM41BKO or wild-type cells stably expressing GFP-mCherry-LC3 were grown on photo-etched coverslips ( Electron Microscopy Sciences , Hatfield , PA ) . Cells were fixed in 4% formaldehyde , 0 . 1% glutaraldehyde/0 . 1 M PHEM ( 240 mM PIPES , 100 mM HEPES , 8 mM MgCl2 , 40 mM EGTA [pH 6 . 9] ) for 1 h . The coverslips were washed with 0 . 1 M PHEM buffer and mounted with Mowiol containing 1 μg/ml Hoechst 33342 . The cells were examined with a Zeiss LSM780 confocal microscope ( Carl Zeiss MicroImaging GmbH , Jena , Germany ) utilizing a Laser diode 405–30 CW ( 405 nm ) and an Ar-Laser Multiline ( 488 nm ) . Cells of interest were identified by fluorescence microscopy; a z-stack covering the whole cell volume ( voxel size: 83 nm × 83 nm × 438 nm ) was acquired . The relative positioning of the cells on the photo-etched coverslips was determined by taking a low magnification DIC image . The coverslips were removed from the object glass , washed with 0 . 1 M PHEM buffer , and fixed in 2% glutaraldehyde/0 . 1 M PHEM overnight . Cells were postfixed in osmium tetroxide and potassium ferry cyanide , stained with tannic acid and uranyl acetate , and thereafter dehydrated stepwise to 100% ethanol followed by flat-embedding in Epon . Serial sections ( 200 nm ) were cut on a Ultracut UCT ultramicrotome ( Leica , Wetzlar , Germany ) and collected on formvar-coated slot-grids . Sections were observed at 200 kV in a Thermo Scientific Talos F200C microscope and recorded with a Ceta 16M camera . Consecutive sections were used to align electron micrographs with fluorescent images in X , Y , and Z . For tomograms , image series were taken between −60° and 60° tilt angles with 2° increment . Single-tilt axes series were recorded with a Ceta 16M camera . Tomograms were computed using weighted back projection using the IMOD package . Display , segmentation , and animation of tomograms were also performed using IMOD software version 4 . 9 [73] . | Eukaryotic cells use autophagy to eliminate unwanted structures—such as protein aggregates , intracellular pathogens , and damaged organelles—that are too large to be handled by the proteasome . This unusual vesicle transport pathway begins with packaging of cytoplasmic targets into a double-membrane vesicle ( autophagosome ) and ends with their degradation in the lysosome . A deeper understanding of autophagy as a regulated mechanism for protein turnover has the potential to provide new therapies for diverse human pathologies , including neurodegenerative disease , cardiometabolic disease , and cancer . Here , we set out to build new reporters for studying mammalian autophagy by genetic screening . This approach allowed us to carry out pooled genome-wide clustered regularly interspaced palindromic repeats ( CRISPR ) knockout screens and recover virtually all known mammalian autophagy-related ( ATG ) factors . In addition , we uncovered several uncharacterized proteins , including the endoplasmic reticulum ( ER ) transmembrane protein 41B ( TMEM41B ) , which we went on to show is required for normal growth of autophagosome precursor membranes around their targets . More broadly , our data set provides a comprehensive resource of genes that affect autophagosome trafficking to the lysosome for further hypothesis testing . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"flow",
"cytometry",
"genome",
"engineering",
"cell",
"death",
"autophagic",
"cell",
"death",
"methods",
"and",
"resources",
"vesicles",
"engineering",
"and",
"technology",
"synthetic",
"biology",
"cell",
"processes",
"green",
"fluorescent",
"protein",
"synthetic",
"bioengineering",
"crispr",
"membrane",
"proteins",
"luminescent",
"proteins",
"cellular",
"structures",
"and",
"organelles",
"synthetic",
"genomics",
"research",
"and",
"analysis",
"methods",
"bioengineering",
"synthetic",
"genome",
"editing",
"lipids",
"proteins",
"cell",
"membranes",
"spectrophotometry",
"biochemistry",
"cytophotometry",
"cell",
"biology",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"spectrum",
"analysis",
"techniques"
] | 2019 | CRISPR screening using an expanded toolkit of autophagy reporters identifies TMEM41B as a novel autophagy factor |
When multiple samples are taken from the neoplastic tissues of a single patient , it is natural to compare their mutation content . This is often done by bulk genotyping of whole biopsies , but the chance that a mutation will be detected in bulk genotyping depends on its local frequency in the sample . When the underlying mutation count per cell is equal , homogenous biopsies will have more high-frequency mutations , and thus more detectable mutations , than heterogeneous ones . Using simulations , we show that bulk genotyping of data simulated under a neutral model of somatic evolution generates strong spurious evidence for non-neutrality , because the pattern of tissue growth systematically generates differences in biopsy heterogeneity . Any experiment which compares mutation content across bulk-genotyped biopsies may therefore suggest mutation rate or selection intensity variation even when these forces are absent . We discuss computational and experimental approaches for resolving this problem .
The somatic genotypes of cancerous and pre-cancerous tissues are frequently assayed by taking biopsies containing a substantial number of cells and genotyping each biopsy as a whole ( via SNP chip , exome or genome sequencing , or other techniques ) . For example , in a study of Barrett’s esophagus genotypes were derived from biopsies containing approximately one million epithelial cells ( e . g . [1] ) . We will refer to this type of data collection as bulk-biopsy genotyping . It is generally much less expensive and technically difficult than single-cell genotyping . In this study we examine the use of multiple biopsies from a single tissue or tumor . It is tempting to think that differences in mutation content observed in bulk-biopsy genotyping reflect underlying differences in the number of mutations per cell , which could be informative about the spatial or temporal evolution of the tissue . But is this really true ? The roles of mutations in cancer have often been assessed by analyzing single tumor samples or paired tumor/normal samples from many patients . However , recently it has been recognized that multiple samples from a single patient , separated in time or space , offer additional information . Spatial heterogeneity of clones implies that sampling a single region of a neoplasm may not be representative of the entire neoplasm . The force of natural selection may vary in different parts of a tissue ( edge versus center , primary tumor versus metastasis ) or over time ( early versus late progression , before versus during or after chemotherapy ) . Multiple samples from a single individual also offer the possibility of phylogenetic analysis to infer relationships among different lineages and reconstruct past events in the history of the tissue . Table 1 shows a sampling of recent studies in which multiple cancer samples per patient were obtained , and phylogenetic methods were either used or could have been used . These studies considered both spatial separation–different parts of a tumor or neoplasm , a tumor and its metastases– and temporal separation– samples taken at different times , such as early and late in progression to cancer , or before and after chemotherapy . They show the potential power of the multiple-sample approach , which we expect will become increasingly important as genotyping costs decrease . Existing methods do not have the resolution to detect all variants present in a million-cell sample . Variants present in just a few cells will go undetected . Peiffer et al . [17] found minimum frequencies of 33% to 50% for reliable detection of copy-number variants in heterogeneous tumor data using a SNP array . Deep sequencing can detect single nucleotide variants in cancer samples at lower frequencies , down to 1% [18] for 40x sequencing , but the threshhold for reliable detection is much higher: even for an average of 2000x sequencing , single nucleotide variants were reproducibly detected only at >15% allele frequency and indels at >5% allele frequency [19] . Even when low-frequency variants are detected , they are often disregarded as they are difficult to quantify and assign to haplotypes . Thus , the result of bulk-biopsy genotyping is generally a survey of locally high-frequency variants only . In a well-mixed tissue such as blood , variants which are at high frequency in one sample will generally be at high frequency in all samples . In such tissues , bulk genotyping will miss low-frequency variants , and will thus be biased toward detecting older rather than younger mutations . However , this bias will affect all samples equally and will not tend to produce spurious evidence of non-neutrality . However , solid tissues are not well-mixed . We will consider the behavior of bulk-biopsy genotyping in a simulated tissue similar to Barrett’s esophagus ( BE ) : a sheet of tissue rolled into a cylinder , with very limited mobility of cell lineages except during initial development . While our simulations are inspired by BE , our conclusions should apply directly to neoplasms in two-dimensional epithelial sheets such as colon , skin , bladder and lung , and conceptually similar effects are also likely in three-dimensional tumors . The key factor is growth with limited mixing . An increasingly common objective in taking multiple biopsies from a neoplastic tissues is to look for evidence of natural selection or heightened mutation acting on specific clones . This is distinct from standard methods of detecting selection or enhanced mutation via comparison of single samples from many different tumors . A straightforward statistical approach to detecting perturbing forces from multi-sample data would be to infer the evolutionary tree connecting samples from the same individual , and test if that tree conforms to a molecular clock . We simulate this experiment on data which do have a molecular clock , and show that bulk-biopsy genotyping very often leads to the spurious rejection of the clock , and thus to a conclusion of non-neutrality , even when the underlying data are completely neutral . We emphasize that the bias we observe is not specific to the use of a phylogeny-based molecular clock test , but will influence any formal or informal comparison of apparent mutation content differences among biopsies . For example , if researchers use bulk genotyping to identify a biopsy with an unusually high number of mutations , and conclude that the highly mutant biopsy represents a genetically unstable lineage , they are implicitly assuming that bulk data have a molecular clock in the absence of perturbing forces . As we will show , this is not the case .
We model the BE segment as a 300 x 300 grid of crypts rolled into a cylinder , approximating the size of a typical BE segment . We treat a crypt as the basic replicative unit , since genetic drift is expected to rapidly homogenize the genotype within each crypt . Crypts have a birth rate representing crypt fission . According to Totafurno’s model of the crypt cycle , when stem cells double in number crypt fission is triggered , which results in halving the doubled stem cell population into two new daughter crypts [20] . In this study , we model the crypt fission cycle by allowing a crypt to either eliminate a neighboring crypt or fill in a space lacking crypts . Crypts also have a death rate , a dead crypt leaving an empty space . The BE segment is thought to expand from the gastro-esophageal junction . An alternative possibility is that cells gain a BE phenotype and spread clonally from squamous duct glands that are situated throughout the esophagus [21] . In both scenarios , BE segments must be rapidly established , since BE segments have not been endoscopically observed in the process of expanding . However , when we simulated the esophagus beginning from a uniform field of non-mutant cells no tree structure arose even after many simulated decades . We do not show results for this case as it is trivially predictable from coalescent theory: our simulations cover approximately 50 generations , and a thoroughly mixed population of size 90 , 000 will have an average of 1765 distinct ancestors 50 generations ago and will thus appear as approximately 1765 unrelated patches . A less thoroughly mixed sheet of cells will be even patchier . This is not consistent with actual BE data [1] which show mutations shared among biopsies . It is possible that BE arises in situ and is then “overwritten” by an early selective sweep; but if so , this seems little different from the BE segment itself arising by growth from one or a few ancestral crypts . We therefore model the establishment of BE as an expansion from the gastro-esophageal junction . To simulate BE data we used the agent-based forward simulator of [22] . While this simulator provides for loci whose mutant alleles modify the growth or mutation rates , in the majority of experiments presented here we used a purely neutral model . We simulated 1000 neutral loci for phylogeny inference . Mutations were scored as number of changes from ancestral state; there was no back mutation . We considered neutral mutation rates per locus per crypt per year ( μ ) of 0 . 001 and 0 . 002 . Data with the lower rate are fairly sparse , while data with the higher rate are highly polymorphic . We set the probability that a dividing crypt could displace a neighbor at 1 . The crypt birth rate was 0 . 02 and death rate 0 . 001 . We also did simulations with 100 neutral loci and mutation rates of 0 . 001 , 0 . 002 , and 0 . 004 , presented in Supporting Information . For illustrative purposes we also did a small number of simulations with five potentially selected loci , each having a mutation rate of 10−7 per locus per crypt per year and a twofold selective advantage for the mutant type over the wild type . The mutation rates given here do not correspond directly to per-cell mutation rates since , when a mutation arises in a crypt , it may be lost rather than fixed . The per-cell mutation rate would be higher by a factor of the mean number of stem cells in the crypt . In any case our mutation rates are chosen in order to give ample mutations for phylogenetic analysis with a limited number of loci . Real data would have fewer mutations per locus , but far more loci . We do not expect this to substantially change the results . Numerical estimates of BE crypt birth and death rates are not available . Our chosen numbers , which were roughly inspired by values measured for human colon crypts [23] , produce an initial spread which is slower than in BE , and a subsequent steady state which probably has faster turnover . However , this is conservative for our conclusions: a faster initial spread and slower subsequent turnover with the same expected amount of mutation would show even greater distortion of the molecular clock . We believe that the details of our parameter choice will not affect our qualitative conclusions as long as the pattern of rapid spread followed by slow turnover is conserved . We started with a single randomly placed crypt and simulated 20 years of growth . This was generally enough to allow crypts to fill the lower esophagus . A small proportion of simulations resulted in the death of the nascent Barrett’s epithelium; these were discarded . We then randomly chose 10 biopsies which were squares of 10x10 crypts , constrained not to overlap . Rarely , a biopsy was found to contain no live crypts; in such cases the entire simulation was discarded . Additional simulations were run to replace discarded simulations . These simulation conditions imply a molecular clock , as the mutation rate is the same in all crypts . We tested for presence of a clock in single crypt samples and in biopsies of different sizes using PAUP* 4 . 0 [24] . For analytic purposes we treated all loci with one or more mutations as one state , and loci with zero mutations as another state . This corresponds to the presence/absence scoring typically used for BE data . To enable use of available phylogenetic software , these states were coded as purine and pyrimidine ambiguity codons . We tested both estimation of the state frequencies from the data using the EMPIRICAL algorithm in PAUP* ( results shown in paper ) , and setting the frequencies equal ( results shown in Supporting Information ) . We performed maximum-likelihood analyses of the recoded data with and without the clock constraint , and assessed the difference in log-likelihoods using a likelihood ratio test [25 , 26] with a 5% significance cutoff . When multiple tied trees were produced , we used the first listed tree for analysis . This use of the likelihood ratio test can be criticized as it assumes that the clocklike and non-clocklike best trees had the same topology [26] , which was not always the case . We applied the test to all pairs of trees , even those differing in topology . Our argument is that when the topological difference is trivial ( rearrangement across branches of near-zero length ) the result of the test will be almost exactly the same as it would for identical topologies; and when the topological difference is non-trivial rejection of the clock is justified even though an exact statistical test is not available . To measure the influence of biopsy size on detection of rate heterogeneity , we subsampled our biopsies . That is , to produce a 4x4 biopsy we took a 4x4 subsample from the original 10x10 biopsy . To avoid dead crypts , we examined subsamples in turn starting in the upper left and chose the first one in which at least 1 live crypt was found . To measure the influence of detection threshholds , we used the same sets of simulated biopsies , but varied the cutoff used to establish the biopsy “genotype . ” For example , when the cutoff was 30% , we scored a mutation as present if it appeared in 30% or more of the sampled living crypts from the biopsy , and absent otherwise . In the simulations with 100 loci and μ = 0 . 001 , which had the smallest amount of information per phylogeny , a few cases with large biopsies and stringent cutoffs could not be run . Stringent cutoffs can generate biopsies with no detectable mutations , and having too many such biopsies in a single tree causes failure of the phylogeny analysis . Such runs were discarded . No more than 15/500 runs failed for any combination of conditions; the number of failed runs for each condition are given in the legends to S5 and S6 Tables . Our simulated data is archived on Dryad at http://dx . doi . org/10 . 5061/dryad . hf93c .
Our simulations were inspired by Barrett’s esophagus ( BE ) , a neoplastic condition in which the lower esophagus is colonized by a tissue organized into crypts . We treat crypts as the fundamental unit of our simulation , and assume that all spread of genotypes results from reproduction ( fission ) of crypts which either replace their neighbors or spread into unoccupied areas . The details of the simulator are described in [22] . At the beginning of the simulation each crypt began with an identical genome of 100 or 1000 loci . Mutations in these loci were selectively neutral: they were used solely to infer the relationships among biopsies . The first striking effect of bulk sampling was seen when the simulation was seeded with a completely filled grid of crypts . At the end of the simulation the tissue consisted of tiny patches of related crypts , each patch unrelated to its neighbors . This reflects the very low gene flow in a static crypt-organized tissue without natural selection . In a tissue of this kind , bulk genotyping would lead to the incorrect conclusion that there are few or no mutations present . Bulk biopsy sampling of actual BE segments shows abundant mutations [1] . We therefore considered a theory of BE origin in which it spreads from a few crypts . We represented this by seeding the simulation with a single randomly placed crypt . Biopsies sampled from such a tissue did contain genetic variants detectable with bulk genotyping , consistent with actual BE data . The spatial distribution of mutations in real BE segments is poorly known , as normally only a few biopsies are analyzed per individual . In our simulations we could readily examine the entire pattern , as well as taking simulated biopsies . The simulated BE segments developed a strongly sectored pattern , with small diverse patches of cells near the original seeding area , and larger , more homogeneous patches far from it . Sharp borders between genetically distinct lineages were seen; these borders ran vertically along the simulated esophagus , roughly parallel to the direction of tissue growth . A typical example , captured partway through colonization of the simulated esophagus , is shown in Fig 1 . These patterns reflect the effect of “gene surfing” [27] . Gene surfing is a phenomenon in population genetics , observed when a population is rapidly expanding into a new geographical region but the mobility of individuals is limited . Colonization is therefore driven by a few individuals on the leading edge of the population , and their genotypes will be disproportionately represented in the newly colonized area . Patterns visually similar to our simulations can be seen when two different strains of bacteria are mixed and seeded onto a plate: sectors of pure strains are generated by replication of the few individuals on the colony edge [28] even in the absence of any selective advantage . Our simulations , seeded with a single crypt , thus produced data that were broadly consistent with observations of actual BE segments . We next asked whether biopsies sampled from these purely neutral simulations would pass tests for neutrality . Based on ten biopsy samples from each of 100 simulated BE segments , we inferred phylogenetic trees and tested whether those trees rejected the molecular clock at the 5% level . We considered biopsies of sizes from 1 ( a single crypt ) to 10x10 ( 100 crypts ) . For biopsies of size greater than 1 , we also considered detection cutoffs from 10% ( mutations present in 10% or more of crypts were scored ) to 100% ( only mutations present in all crypts were scored ) . If biopsy sampling provided a phylogenetically unbiased sample of mutations occuring in our data , we would expect to see a molecular clock in our inferred trees with any size of biopsy . The proportion of inferences ( out of 500 ) rejecting the molecular clock are shown in Figs 2 and 3 and are presented in table form in S1 and S2 Tables . In these figures , the white color seen at the left-hand edge ( single-crypt samples ) represents an acceptable clock rejection rate of 5% . ( Note that detection cutoff does not affect the results from single crypts , and thus all of the left-hand results represent the same analyses . ) All larger biopsy sizes , even 2x2 biopsies with only 4 crypts , rejected the clock at high rates for all conditions studied . The choice of cutoff had a noticable impact on clock rejection . Cutoffs in the 30%-50% range were better than higher or lower cutoffs; the larger the biopsy , the lower the optimal cutoff . However , no cutoff tested restored the clocklike nature of the underlying data . Superficially satisfactory results can be obtained by using only 100 neutral loci and inferring frequencies of the mutant and non-mutant states ( S3 Fig ) . However , this apparent improvement merely represents lack of statistical power to detect clock violations , as seen by the dramatic worsening of results with 1000 neutral loci and the same model ( Fig 2 ) . Use of equal frequencies of mutant versus non-mutant states produces higher clock rejection: results are shown for completeness in S1 and S2 Figs for 1000 neutral loci and S4 , S5 and S6 Figs for 100 loci . We show a randomly selected pair of inferred trees from the simulation of S6 Table in Fig 4 . The topologies of the two trees differed in ordering of the short bottommost branches . Larger discrepancies were seen in the branch lengths . The single-crypt tree ( A ) showed some heterogeneity of branch lengths , but it was well within the expected range for a data set of this size , and the clock was not rejected . The 10x10 crypt tree ( B ) was much more distorted , and rejected the molecular clock . It would be tempting to conclude that biopsy 10 , in particular , had a higher mutation rate than biopsy 8; yet they arose from a simulation with perfectly equal rates .
We have presented our results in terms of rejection of the molecular clock in a formal test . However , their significance is not limited to such tests . When we first examined bulk-genotyped BE data for multiple biopsies per patient we saw a striking difference in the mutation content of different biopsies . It was natural to read this as a difference in the underlying mutation rate . After further thought we realized that it could also reflect a difference in the growth rate , since rapidly growing cells will form more uniform samples and therefore appear to have more mutations . Only after performing simulations did we discover that hetereogeneity in apparent mutation content is a general feature of this type of data and should be expected even when neither mutation rate nor growth rate varies . Comparison of Figs 2 and S5 shows that the more informative the data , the stronger this tendency to reject the clock . The clock test formalizes a scientist’s intuition , but both the test and the intuition are liable to error in this case . We stress that our findings do not challenge the important role non-neutral processes play in the development of cancer . Instead , they warn us against drawing conclusions about non-neutrality that cannot be supported . Two factors combine to produce this spurious evidence for non-neutrality . Gene surfing causes biopsies taken near the origin of the growing population to be much more heterogeneous than those taken far from the origin ( see Fig 1 ) . Bulk-biopsy sampling then translates this difference in diversity into a difference in detectable mutation content: a homogeneous cell sample will have more high-frequency mutations than a heterogeneous one , and bulk sampling detects only high-frequency mutations . This is most easily understood by considering common ancestry . Consider , as an example , a detection cutoff of 50% . Mutations which reach this cutoff must exist in 50% or more of the cells in the biopsy , and thus , barring convergent evolution , must be inherited from a common ancestor of 50% of the cells . If this common ancestor existed early in the development of the tissue , it likely had relatively few mutations , so few mutations will be shared by its descendants . If it existed more recently , it likely had more mutations ( since mutations accumulate over time ) and its descendants will have more shared mutations . Cells from a biopsy whose common ancestor is ancient will , individually , have just as many mutations , but a much larger proportion will be at low frequency in the biopsy . Such mutations are difficult to detect with bulk genotyping . We did not model complicating factors in analysis of bulk data such as differences in ploidy among lineages or typing errors . However , when a bias is present in analysis of clean , error-free genotypic data , there is no reason to believe that better results would come from dirtier data . Regrettably , no tested frequency cutoff rule was successful in resolving this problem . We suggest three possible approaches . Once subclones within a biopsy have been detected via approaches ( 2 ) or ( 3 ) , this information needs to be incorporated into the analysis . For analytic methods involving phylogenetics , mixed samples are particularly challenging because when two variants are found at similar frequency in a sample , there is no easy way to determine whether they represent one lineage with two mutations or two lineages with one mutation each [33] . In principle it would be possible for a statistical analysis to sum over these possibilities using an approach analogous to that of [34] . The high computational burden of this approach will have to be compared with the experimental burden of small-sample typing . Alternatively , one could use the minority alleles to estimate biopsy diversity , without attempting to reconstruct minority genotypes . One potential use of diversity estimations would be as the basis for corrected mutational distances to be used in phylogeny inference . Simulations or heuristics could be used to establish the relationship between observed mutational distance between two biopsies , internal diversity of each biopsy , and the true mutational distance . Distances corrected according to this relationship could then be used in a distance-based phylogeny algorithm to produce trees whose branch lengths more accurately represented the underlying mutational frequencies . We are currently developing such an algorithm . Accurate inference of branch lengths is important in distinguishing , for example , a mutation rate increase in a specific lineage ( presumably due to a mutator mutation or epigenetic change ) from a mutation rate increase at a given time across the entire tissue ( presumably due to an environmental change , since it manifests in unrelated lineages ) . Naive tree-drawing based on uncorrected data , as shown by the results in this paper , cannot answer such questions as the branch lengths of its trees are not proportional to time . This is shown dramatically in real BE data , where no separation is seen between data collected from time points many years apart [1] . This situation makes it difficult to draw conclusions about changes in rate over time , though in some cases coherent patterns have been detected [1] . Genetic distances corrected for the bias inherent in bulk-biopsy sampling could allow much more accurate separation of neutral from non-neutral processes in the development of tissues and cancers . One further positive finding from this study is that the spread of a growing tissue tends to produce a characteristic fan-shaped pattern , as seen in Fig 1 . As dense sampling of cancer and pre-cancer tissues becomes more feasible , it will become possible to detect this pattern or deviations from it which may indicate selection . An example is shown in Fig 5 , which shows four typical results from a simulation with selected as well as neutral mutations . Note the disruption of the fan pattern by lateral growth of selected clones . The spatial distribution of clones within an expanding tumor or neoplasm may therefore reveal selective processes , as has been explored by [11] in colorectal cancer . In other words , these simulations provide predictions for the nascent field of tumor phylogeography . More work is needed both to detect these patterns and to assess their significance . | Researchers who take multiple samples from a cancer or pre-cancer tissue and find that some samples show far more mutations than others are likely to conclude that the high-mutation samples reflect cells with an abnormal mutation or growth rate . We considered the common practice of testing a bulk sample for mutations , which finds only mutations that are common within the sample . Our computer simulations show that even when all cells have identical mutation and growth rates , testing bulk samples frequently leads to spurious detection of rate differences . This can lead to false conclusions about the causes and progress of cancer . We discuss possible solutions involving either genetic testing of single cells or the use of computer algorithms to detect rare mutations within a sample . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"biopsy",
"cloning",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"phylogenetics",
"data",
"management",
"phylogenetic",
"analysis",
"gastroenterology",
"and",
"hepatology",
"molecular",
"biology",
"techniques",
"genotyping",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"evolutionary",
"systematics",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gastrointestinal",
"tract",
"barrett's",
"esophagus",
"anatomy",
"gene",
"identification",
"and",
"analysis",
"genetics",
"mutation",
"detection",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"esophagus"
] | 2016 | Bulk Genotyping of Biopsies Can Create Spurious Evidence for Hetereogeneity in Mutation Content |
Thermodynamics dictates the structure and function of metabolism . Redox reactions drive cellular energy and material flow . Hence , accurately quantifying the thermodynamics of redox reactions should reveal design principles that shape cellular metabolism . However , only few redox potentials have been measured , and mostly with inconsistent experimental setups . Here , we develop a quantum chemistry approach to calculate redox potentials of biochemical reactions and demonstrate our method predicts experimentally measured potentials with unparalleled accuracy . We then calculate the potentials of all redox pairs that can be generated from biochemically relevant compounds and highlight fundamental trends in redox biochemistry . We further address the question of why NAD/NADP are used as primary electron carriers , demonstrating how their physiological potential range fits the reactions of central metabolism and minimizes the concentration of reactive carbonyls . The use of quantum chemistry can revolutionize our understanding of biochemical phenomena by enabling fast and accurate calculation of thermodynamic values .
In order to understand life we need to understand the forces that support and constrain it . Thermodynamics provides the fundamental constraints that shape metabolism [1–5] . Redox reactions constitute the primary metabolic pillars that support life . Life itself can be viewed as an electron transport process that conserves and dissipates energy in order to generate and maintain a heritable local order [6] . Indeed , almost 40% of all known metabolic reactions are redox reactions [7 , 8] . Redox biochemistry has shaped the study of diverse fields in biology , including origin-of-life [9] , circadian clocks [10] , carbon-fixation [11] , cellular aging [12] , and host-pathogen interactions [13] . Previous work has demonstrated that a quantitative understanding of the thermodynamic parameters governing redox reactions reveals design principles of metabolic pathways . For example , the unfavorable nature of carboxyl reduction and carboxylation explains to a large degree the ATP investment required to support carbon fixation [1] . Developing a deep understanding of redox biochemistry requires a comprehensive and accurate set of reduction potential values covering a broad range of reaction types . However , only ~100 reduction potentials can be inferred from experimental data , and these suffer from inconsistencies in experimental setup and conditions . Alternatively , group contribution methods ( GCM ) can be used to predict a large set of Gibbs energies of formation and reduction potentials [14] . However , the accuracy of this approach is limited , as GCM do not account for interactions between functional groups within a single molecule and GCM predictions are limited to metabolites with functional groups spanned by the model and experimental data . Quantum chemistry is an alternative modeling approach that has been used to predict redox potentials in the context of numerous applications , such as redox flow batteries , optoelectronics , and design of redox agents [15–27] . Unlike GCM , whose smallest distinct unit is a functional group , quantum chemistry directly relates to the atomic and electronic configuration of a molecule , enabling ab initio prediction of molecular energetics . Here , we adopt a quantum chemistry modeling approach from the field of redox flow battery design [25 , 26 , 28] to predict the reduction potentials of biochemical redox pairs . Our approach combines ab initio quantum chemistry estimates with ( minimal ) calibration against available experimental data . We show that the quantum chemical method can predict experimentally derived reduction potentials with considerably higher accuracy than GCM when calibrated with only two parameters . We use this method to estimate the reduction potentials of all possible redox pairs that can be generated from the KEGG database of biochemical compounds [7 , 8] . This enables us to decipher general trends between and within groups of oxidoreductase reactions , which highlight design principles encoded in cellular metabolism . We specifically focus on explaining the central role of NAD ( P ) as electron carrier from the perspective of the redox reactions it supports and the role it plays in lowering the concentration of reactive carbonyls .
To facilitate our analysis we divided redox reactions into several generalized oxidoreductase groups which together cover the vast majority of redox transformations within cellular metabolism ( Fig 1A ) : ( G1 ) reduction of an unmodified carboxylic acid ( -COO ) or an activated carboxylic acid–i . e . , phosphoanhydride ( -COOPO3 ) or thioester ( -COS-CoA ) –to a carbonyl ( -C = O ) ; ( G2 ) reduction of a carbonyl to a hydroxycarbon ( -COH , i . e . , alcohol ) ; ( G3 ) reduction of a carbonyl to an amine ( -CNH3 ) ; and ( G4 ) reduction of a hydroxycarbon to a hydrocarbon ( -C-C- ) , which usually occurs via an ethylene intermediate ( -C = C- ) . We note that this categorization corresponds to the treatment of carbon oxidation levels in standard organic chemistry textbooks [29] . We developed a quantum chemistry method for predicting the standard transformed redox potential of biochemical redox reactions . We explored a range of different model chemistries , including combinations of DFT ( density functional theory ) functionals or wave-function electronic structure methods , basis sets , choice of implicit solvent , and choice of dispersion correction . We found that a DFT approach that uses the double-hybrid functional B2PLYP [32 , 33] gave the highest prediction accuracy ( see Methods for detailed model chemistry description; other model chemistries also gave high accuracy as discussed in the Supplementary Information and S1 Fig ) . As each biochemical compound represents an ensemble of different chemical species–each at a different protonation state [31] –we applied the following pipeline to predict E’m ( Fig 1 , see also Methods ) : ( i ) a quantum chemical simulation was used to obtain the electronic energies of the most abundant chemical species at pH 0; ( ii ) we then calculated the difference in electronic energies ΔEElectronic between the product and substrate of a redox pair at pH 0 , thus obtaining estimates of the standard redox potential , Eo; ( iii ) next , we employed empirical pKa estimates to calculate the energetics of the deprotonated chemical species and used the extended Debye-Huckel equation and the Alberty-Legendre transform [31] to convert Eo to the standard transformed redox potential E’m at pH = 7 and ionic strength I = 0 . 25 M ( as recommended [34] ) , where reactant concentrations are standardized to 1 mM to better approximate the physiological concentrations of metabolites [1 , 35] . Finally , ( iv ) to correct for systematic errors , the predicted E’m values , of each oxidoreductase group , were calibrated by linear regression ( two-parameter calibration ) against a set of 105 experimentally measured potentials obtained from the NIST Thermodynamics of Enzyme-Catalyzed Reactions database ( TECRDB ) [30] and the Gibbs formation energy dataset of Robert Alberty [31] ( Supplementary Information ) . We note that we observe empirically that the difference in electronic energies ΔEElectronic is strongly correlated with the Gibbs reaction energy ΔGr for these redox systems ( S5 Fig ) and so we estimate redox potentials using the former in order to reduce computational cost ( see SI for details ) . We also note that the two-parameter calibration is needed mainly since we ignore vibrational enthalpies and entropies of the compounds ( Supplementary Information ) . As exemplified in Fig 2A and 2B and S2 Fig , the calibration by linear regression significantly improves the accuracy of our quantum chemistry predictions . As shown in Table 1 , the predictions of quantum chemistry have a lower mean absolute error ( MAE ) than those of GCM for all reaction categories . ( GCM has a higher Pearson correlation coefficient for category G1 , but this is an artifact introduced by a single outlier value , S3 Fig ) . The improved accuracy is especially noteworthy as our quantum chemical approach derives reduction potentials from first principles and requires only two calibration parameters per oxidoreductase group ( α and β in Fig 1E ) , as compared to GCM which uses 5–13 parameters while achieving lower prediction accuracy ( Table 1 ) . Therefore , our quantum chemistry approach can be extended to predict reduction potentials for a wide domain of redox reactions since it does not depend as heavily on empirical measurements . While the quantum chemistry method is computationally more expensive than GCM–with a cost that scales with the number of electrons per molecule ( Supplementary Information ) –it can still predict the potentials for several hundreds of reactions when run on a typical high-performance computing cluster . Inconsistencies between our predictions and experimental measurements can be used to identify potentially erroneous experimental values . However , as such discrepancies might stem from false predictions , we used an independent method to estimate redox potentials . We reasoned that consistent deviation from two very different prediction approaches should be regarded as indicative of potential experimental error . The second prediction approach we used is based on reaction fingerprints [38] , where the structure of the reactants involved is encoded as a binary vector ( 166 parameters without regularization , Supplementary Information ) . These binary vectors are then used as variables in a regularized regression to correlate structure against a physicochemical property of interest , such as redox potential [38 , 39] . This approach is similar to the group contribution method ( GCM ) in that it is based on a structural decomposition of compounds; however , unlike GCM , fingerprints encode a more detailed structural representation of the compounds . To detect potentially erroneous experimental measurements , we focused on redox potentials of category G2 ( carbonyl to hydroxycarbon reduction ) as we have abundant experimental information for this oxidoreductase group ( see S4 Fig for results with the other categories ) . As shown in Fig 2C , we normalized the prediction errors by computing their associated z-scores ( indicating how many standard deviations a prediction error is from the mean error across all reactions ) . Two redox reactions stand out as having significantly different experimental and predicted values for both methods ( Z>2 ) : indolepyruvate reduction to indolelactate ( indolelactate dehydrogenase , EC 1 . 1 . 1 . 110 ) and succinate semialdehyde reduction to 4-hydroxybutanoate ( succinate semialdehyde reductase , 1 . 1 . 1 . 61 ) . We suggest an explanation for the observed deviation of the first reaction: in the experimental study , the K’eq of indolelactate dehydrogenase was measured using absorbance at 340 nm as an indicator of the concentration of NADH [40] . However , since indolic compounds also have strong absorption at 340 nm [41] , this method probably resulted in an overestimation of the concentration of NADH , and thus an underestimation of K’eq . Indeed , the experimentally derived E’m is considerably lower ( -400 mV ) than the predicted one ( -190 mV , via quantum chemistry ) . With regards to the second reaction , succinate semialdehyde reductase , we note that re-measuring its redox potential is of considerable significance as it plays a central role both in carbon fixation–e . g . , the 3-hydroxypropionate-4-hydroxybutyrate cycle and the dicarboxylate-4-hydroxybutyrate cycle [11] –as well as in production of key commodities–e . g . , biosynthesis of 1 , 4-butanediol [42] . We used the calibrated quantum chemistry model to predict redox potentials for a database of natural and non-natural redox reactions . We generated this dataset by identifying pairs of metabolites from KEGG [7 , 8] that fit the chemical transformations associated with each of the four different oxidoreductase groups ( Methods ) . We considered only compounds with fewer than 7 carbon atoms , thus generating a dataset consisting of 652 reactions: 83 reductions of category G1; 205 reductions of category G2; 104 reductions of category G3; and 260 reductions of category G4 ( Supplementary Dataset 1 ) . Some of these redox pairs are known to participate in enzyme-catalyzed reactions while others are hypothetical transformations that could potentially be performed by engineered enzymes . We note that our approach to generate reactions is similar to that of the comprehensive Atlas of Biochemistry [43] , but we focus solely on the four redox transformations of interest . Fig 3A shows the distribution of all predicted redox potentials at pH = 7 , I = 0 . 25 M and reactant concentrations of 1 mM , i . e . , E’m [14 , 36] . Fig 3 demonstrates that the value of E’m is directly related to the oxidation state of the functional group being reduced . The general trend is that “the rich get richer” [1 , 44 , 45]: more reduced functional groups have a greater tendency to accept electrons , i . e . , have higher reduction potentials . Specifically , the reduction potential of hydroxycarbons ( G4 , <E′m> = −15 mV ) is higher than that of carbonyls ( <E′m> = −225 mV for both G2 and G3 ) and the reduction potential of carbonyls is higher than that of un-activated carboxylic acids ( G1 , <E′m> = −550 mV ) . Categories G2 and G3 ( reduction of carbonyls to hydroxycarbons or amines , respectively ) have very similar potentials because the oxidation state of the functional groups involved is identical ( note that this holds for the physiological E′m but not for E′o because reactions in the G3 category are balanced with an ammonia molecule as a substrate , thus introducing a factor of RTln ( 10−3 ) when converting to the mM standard state ) . For category G1 , activation of carboxylic acids significantly increases their reduction potential ( orange line in Fig 3 ) as the energy released by the hydrolysis of the phosphoanhydride or thioester ( ~50kJ/mol ) activates the reduction: ΔE=50nF≅250mV ( n being the number of electrons , F the Faraday constant ) . The quantum chemical predictions further enable us to explore detailed structure-energy relationships within each of the general oxidoreductase groups . To exemplify this we focus on the G2 category , as shown in Fig 4 . While we find no significant difference between the average E’m of aldehydes and ketones , we can clearly see that the identity of functional groups adjacent to the carbonyl has a significant effect on E’m , as expected . Alpha ketoacids and dicarbonyls have a significantly higher E’m than alpha hydroxy-carbonyls ( Δ <E′m> ≅ 20 mV , p < 0 . 005 ) and carbonyls adjacent to hydrocarbons ( Δ <E′m> ≅ 35 mV , p < 0 . 0005 ) . Carbonyls next to double bonds or aromatic rings have a significantly lower E’m values than alpha hydroxy-carbonyls and carbonyls that are next to hydrocarbons ( Δ <E′m> ≅ −50 mV , and Δ <E′m> ≅ −40 mV respectively , p < 0 . 0001 ) . Lactones ( cyclic esters ) , have redox potentials that are significantly lower than any other subgroup within the G2 category . As another validation of the predicted potentials , we found that the reduction potentials of open-chain sugars are significantly higher than those of closed-ring sugars that undergo ring opening upon reduction , where Δ <E′m> ≅ 60 mV ( p < 10−5 ) . This is consistent with the known thermodynamics of closed-ring sugar conformations , e . g . , the Keq of arabinose ring opening is ~350[46] , which translates to ΔE=RTln ( 350 ) nF≅75mV , close to the observed average potential difference between the subgroups ( R is the gas constant , and T the temperature ) . While myriad natural electron carriers are known to support cellular redox reactions , NAD ( P ) has the prime role in almost all organisms , participating in most ( >50% ) known redox reactions [7 , 8] . The standard redox potential of NAD ( P ) is ~ -330 mV ( pH = 7 , I = 0 . 25 ) , but as [NADPH]/[NADP] can be higher than 50 and [NADH]/[NAD] can be lower than 1/500 , the physiological range of the NAD ( P ) reduction potential is between -380 mV and -250 mV [35 , 47–51] . Most cellular redox reactions are therefore constrained to a limited reduction potential range determined by the physicochemical properties and physiological concentrations of NAD ( P ) . By examining the fundamental trends of redox potentials of the different oxidoreductase groups we will show that NAD ( P ) is well-matched to the redox transformations most commonly found in cellular metabolism . Fig 3 demonstrates that the reduction potentials of activated acids ( activated G1 ) and carbonyls ( G2 and G3 ) are very similar , such that NAD ( P ) can support both the oxidation and reduction of nearly all redox couples in these classes . Although the distributions associated with these redox reactions are not entirely contained in the NAD ( P ) reduction potential range ( marked in grey ) , the reduction potential of a redox pair can be altered by modulating the concentrations of the oxidized and reduced species . As the concentrations of metabolites usually lie between 1 μM and 10 mM [1 , 4 , 35 , 52] , the reduction potential of a redox pair can be offset from its standard value by up to ±RTln ( 104 ) nF≅±120mV ( assuming two electrons are transferred ) . Therefore , NAD ( P ) can support reversible redox reactions of compound pairs with E’m as low as −380 − 120 = −500 mV and as high as −250 + 120 = −130 mV ( indicated by the light grey regions in Fig 3 ) , a range that encompasses almost all activated acids ( activated G1 ) and carbonyls ( G2 and G3 reactions ) . Outside this range , however , NAD ( P ) ( H ) can only be used in one direction of the redox transformation–either oxidation or reduction , but not both . Fig 3 shows that NAD ( P ) H can support irreversible reductions of hydroxycarbons to hydrocarbons and NAD ( P ) supports irreversible oxidation of carbonyls to carboxylic acids . Next , we focus on a small set of redox reactions found in the extended central metabolic network that is shared by almost all organisms: ( i ) The TCA cycle , operating in the oxidative or reductive direction [53] , as a cycle or as a fork [54] , being complete or incomplete [54] , or with some local bypasses ( e . g . , [55] ) ; ( ii ) glycolysis and gluconeogenesis , whether via the EMP or ED pathway [56] , having fully , semi or non-phosphorylated intermediates [57]; ( iii ) the pentose phosphate cycle , working in the oxidative , reductive or neutral direction; and ( iv ) biosynthesis of amino-acids , nucleobases and fatty acids . As schematically shown in Fig 5 , and listed in Supplementary Dataset S2 , the ≈ 60 redox reactions that participate in the extended central metabolism almost exclusively belong to one of the following groups: ( i ) reduction of an activated carboxylic acid to a carbonyl or the reverse reaction oxidizing the carbonyl ( 9 reactions , G1 ) ; ( ii ) reduction of a carbonyl to a hydroxycarbon or its reverse oxidation ( 20 reactions , G2 ) ; ( iii ) reduction of a carbonyl to an amine or its reverse oxidation ( 18 reactions , G3 ) ; ( iv ) irreversible oxidation of carbonyls to un-activated carboxylic acids ( 5 reactions , G1 in the direction of oxidation ) ; and ( v ) irreversible reduction of hydroxycarbon to hydrocarbons ( 4 reactions , G4 ) . Only two central metabolic reactions ( marked in magenta background in Fig 5 ) oxidize hydrocarbons to hydroxycarbons ( G4 , in the direction of oxidation ) and require a reduction potential higher than that of NAD ( P ) : oxidation of succinate to fumarate and oxidation of dihydroorotate to orotate ( While formally being oxidation of hydrocarbon to hydroxycarbon , the oxidations of prephenate to 4-hydroxyphenylpyruvate and of arogenate to tyrosine present a special case since they create a highly stable aromatic ring and hence have enough energy to donate their electrons directly to NAD ( P ) ) . Similarly , the extended central metabolic network does not demand the low reduction potential required for the reduction of un-activated carboxylic acids ( G1 ) . The reduction potential range associated with NAD ( P ) therefore perfectly matches the vast majority of reversible redox reactions in extended central metabolism–i . e . , reduction of activated carboxylic acids and reduction of carbonyls ( orange , purple and blue distributions in Fig 3 ) –and can also support the common irreversible redox transformations of extended central metabolism–i . e . , reduction of hydroxycarbons and oxidation of carbonyls to un-activated carboxylic acids ( green and red distributions in Fig 3 ) . Cells typically rely on secondary redox carriers like quinones and ferredoxins ( Fig 3 , S1 Table ) , to support less common reactions , i . e . , oxidation of hydrocarbons and reduction of un-activated carboxylic acids . Why is the reduction potential of NAD ( P ) lower than the E’m of most carbonyls ( Fig 3 ) ? As biosynthesis of an NAD ( P ) derivative with higher reduction potential presents no major challenge [58] , why does this lower potential persist ? We suggest that this redox offset plays an important role in reducing the concentrations of cellular carbonyls by making their reduction to hydroxycarbons favorable . It is well known that carbonyls are reactive towards macromolecules , as they spontaneously cross-link proteins , inactivate enzymes and mutagenize DNA [59 , 60] . As the reduction potential of NAD ( P ) is lower than most carbonyls , the redox reactions in category G2 ( or G3 ) prefer the direction of reduction , thus ensuring that carbonyls are kept at lower concentrations than their corresponding hydroxycarbons ( or amines ) . Assuming a value of E′ = −330 mV for NAD ( P ) and taking the average E’m of the G2 reactions ( <E′m> ≅ −225 mV ) results in an estimated equilibrium concentration ratio [hydroxycarbon][carbonyl]=exp ( − ( E’[NAD ( P ) ]−<E’m> ) nFRT ) ≅3500 , thus ensuring very low levels of the carbonyl species . While we do not have many measurements to confirm this prediction , we note one central example: in E . coli , the concentration of oxaloacetate is 1–4 μM [61] , while the concentration of its conjugated hydroxyacid , malate , is 2–3 mM [52] . For ketoacids and open-ring sugars ( which are especially reactive due to the free carbonyl ) this effect is even more pronounced as both have especially high reduction potentials ( Fig 4 ) . Indeed , the reduction potential of ketoacids is so high that the reverse , oxidative reaction is usually supported by electron donors with a higher potential than NAD ( P ) , for example , quinones , flavins , and even O2 ( e . g . , lactate oxidase , glycolate oxidase ) . Interestingly , the reactions of category G2 that are supported by known enzymes in the KEGG database ( 75% of reactions in this category ) have significantly lower E’m than the remaining reactions , which are not known to be catalyzed by natural enzymes ( Δ <E′m> ≅ 20 mV , p < 0 . 005 ) . As such , we suggest that the G2 transformations that are known to be enzyme-catalyzed are mainly those that are amenable to redox coupling with NAD ( P ) ( Fig 4D ) . Within the subset of G2 transformations found in KEGG , those that use redox cofactors other than NAD ( P ) ( such as cytochromes , FAD , O2 , or quinones ) have higher E’m values ( Δ <E′m> ≅ 20 mV , not significant p = 0 . 03 ) than those that use NAD ( P ) ( Fig 4 ) . Finally , we note that the reduction potential of NADP and activated carboxylic acids ( activated G1 ) overlap almost completely , such that we would not expect NAD ( P ) to have a strong effect on the ratio between the concentrations of carbonyls and activated acids . This is to be expected as both carbonyls and activated carboxylic acids are reactive–e . g . , acetylphosphate and glycerate bisphosphate acetylates proteins spontaneously [62] and acyl-CoA’s S-acetylates cellular peptides non-enzymatically [63] . As such , there is no sense in driving the accumulation of carbonyls at the expense of activated carboxylic acids or vice-versa–neither approach would ameliorate non-specific toxicity .
In this work , we present a novel approach for predicting the thermodynamics of biochemical redox reactions . Our approach differs radically from group contribution methods , which rely on a large set of arbitrarily-defined functional groups , assume no energetic interactions between groups , and are restricted to metabolites that are decomposable into the groups spanned by the model . In contrast , quantum chemistry directly takes into account the electronic structure of metabolites in solution . Focusing on specific examples highlights the strengths of our quantum chemical approach as well as various weaknesses of GCM . For example , we find several reactions where the GCM predictions are obviously inaccurate as they are too high to be reasonable: 2-Hydroxy-5-methylquinone ⇔ 2 , 4 , 5-Trihydroxytoluene ( GCM: E′m = 543 mV , QC: E′o = −158 mV ) ; 2-Pyrone-4 , 6-dicarboxylate ⇔ 2-Hydroxy-2-hydropyrone-4 , 6-dicarboxylate ( GCM: E′m = 1406 mV , QC: E′o = −375 mV ) ; and Mevaldate ⇔ ( R ) -Mevalonate ( GCM: E′m = 132 mV , QC: E′o = −190 mV ) . Close inspection of the group matrix underlying these estimates reveals errors in the decomposition of the compounds . Failures in the GCM decomposition are likely due to the complexity of molecular representations in the standard INCHI format [64] and usually occur with aromatic and delocalized electrons . This reflects challenges inherent in group decomposition , which are avoided when using the quantum chemistry approach . A more illuminating example is that of 3-dehydroshikimate ⇔ shikimate ( shikimate dehydrogenase ) , the sole redox reaction in the shikimate pathway , converting erythrose 4-phosphate and PEP into chorismate . ( Chorismate is required for the biosynthesis of aromatic amino-acids , folates , quinones , and important secondary metabolites [65] ) . GCM predicts a value of E′m = −85 mV , which , if correct , indicates that the reduction of 3-dehydroshikimate with NAD ( P ) H is irreversible . On the other hand , quantum chemistry methods predict E′m = −268 mV , which corresponds to a 6 order-of-magnitude equilibrium concentration difference with respect to the GCM value . The quantum chemistry prediction thus implies reversibility of the oxidoreductase reaction with NAD ( P ) H . As oxidation of shikimate by NAD ( P ) has been shown to occur in-vivo in gram positive bacteria [66] , it is clear that the GCM prediction is wrong and that the quantum chemistry approach provides a more accurate assessment of the thermodynamic potential of this important biochemical reaction . Unlike previous efforts [67 , 68] , our quantum chemistry approach relies on a two-parameter calibration for each oxidoreductase reaction category , which reduces computational cost by avoiding the need to calculate vibrational enthalpies and entropies ( Supplementary Information ) . In future studies , improvements in accuracy could be achieved by exploring a larger space of quantum model chemistries , or—if more experimental data becomes available—calibrating using more sophisticated regression techniques , such as Gaussian Process regression [69] . Yet , as we have shown , the current procedure is sufficient to yield high coverage and accuracy at a reasonable computational cost . In contrast with GCM methods , our calibration parameters can be at least partially interpreted . One important contribution to the systematic bias in the raw quantum calculations ( i . e . the y-intercept in the linear regression ) comes from neglecting the vibrational component of the molecular enthalpy . Interpreting the slope parameter is more complex , yet examples in the literature show that it can be traced back to the choice of solvation [70] or—in the context of modeling quinone derivatives–to the basis set incompleteness and the shortcomings of the DFT exchange correlation functionals [71] . We note , however , that faster computational resources will eventually enable full ab initio prediction of hundreds of standard transformed redox potentials , rendering the two-parameter calibration and the use of empirical pKa values obsolete [15 , 72] . Importantly , the quantum chemical strategy is not subject to the inconsistencies that plague experimental databases . Experimental values are measured in a wide range of different conditions , including temperature , pH , ionic strength , buffers , and electrolytes . In many cases , the exact measurement conditions are not reported , making it practically impossible to account for these factors . Thus , even if we were to gain access to more experimental data , the lack of systematically applied conditions makes such resources problematic . In contrast , quantum chemical simulations can be performed in consistent , well-defined conditions . Why does the primary biological reduction potential range lie between -370 mV and -250 mV ? One possibility is a frozen evolutionary accident . In this view , NAD ( P ) was available early in evolution and was found useful in supporting multiple redox reactions; as such , it was fixed as the central redox carrier before the Last Universal Common Ancestor ( LUCA ) . While we cannot rule out this explanation , we suggest an alternative: that the primary reduction potential range represents a near optimal adaptation given biochemical constraints and selection pressures imposed throughout evolution . This idea is supported by the fact that most extant electron carriers already existed in LUCA [6] , and yet none have as extensive a role in metabolism as NAD ( P ) . Furthermore , derivatives of NAD ( P ) are simple to synthesize biochemically–e . g . deamino-NAD is a precursor of NAD–and can have considerably shifted reduction potentials [58] . Despite this , no organism has been found to rely on such derivatives . Finally , the deaza-flavin coenzyme F420 is a prominent electron carrier in the central metabolism of methanogens and other prokaryotes [73 , 74] , and has a reduction potential around -340 mV [75] , almost identical to that of NAD ( P ) . Hence , even organisms that partially replace NAD ( P ) use a carrier with a similar reduction potential . The enhanced resolution provided by quantum chemistry uncovers important patterns not accessible using traditional analyses . Exemplifying this , we found that the main cellular electron carrier , NAD ( P ) , is ‘tuned’ to reduce the concentration of reactive carbonyls , thereby keeping the cellular environment more chemically stable . Yet , this protection comes at a price: the oxidation of hydroxycarbons is thermodynamically challenging and often requires the use of electron carriers with higher reduction potential . A recent study demonstrates the physiological relevance of this thermodynamic barrier: the NAD-dependent 3-phosphoglycerate dehydrogenase–the first enzyme in the serine biosynthesis route–can sustain high flux in spite of its unfavorable thermodynamics only through coupling with the favorable reduction of 2-ketoglutarate [76] . Our analysis further supports the previous assertion that the TCA cycle has evolved in the reductive direction [53 , 77] . While all the other electron transfer reactions in the extended central metabolism belongs to oxidoreductase groups that can be supported by NAD ( P ) ( H ) , oxidation of succinate–a key TCA cycle reaction–cannot be carried by this electron carrier . As the reverse reaction , i . e . , fumarate reduction , can be support by NADH [78 , 79] , it is reasonable to speculate that the reaction first evolved in the reductive direction , and only later was adapted to work in the oxidative direction using an alternative cofactor . So long as sufficient experimental data is available to allow for calibration , our approach can be extended to other types of biochemical reactions . For example , understanding the thermodynamics of carboxylating and decarboxylating enzymes–the “biochemical gateways” connecting the inorganic and the organic world– could pave the way for the identification of highly efficient , thermodynamically favorable carbon fixation pathways based on non-standard but promising reaction chemistries [80 , 81] . In this way , high-resolution thermodynamic analyses may provide much needed insight for the engineering of microbes to address global challenges .
For each metabolite , we generated ten initial geometric conformations using ChemAxon’s calculator plugin ( Marvin 17 . 7 . 0 , 2017 , ChemAxon ) . Quantum chemistry calculations were performed using the Orca software package ( version 3 . 0 . 3 ) [86] . Geometry optimizations were carried out using DFT , with the B3LYP functional and Orca’s predefined DefBas-2 basis set ( see S3 Table for detailed basis set description ) . The COSMO implicit solvent model [87] was used , with the default parameter values of epsilon = 80 . 4 and refrac = 1 . 33 . DFT-D3 dispersion correction [88] using Becke-Johnson damping [89] was also included . Single point energy ( SPE ) calculations yield the value of the electronic energy EElectronic for each conformer at their optimized geometry . We used the optimized geometries obtained using DFT as inputs for SPE calculations ( see below and SI for details on the SPE model chemistry selected ) . Substrate and product conformers were sampled according to a Boltzmann distribution . By taking the difference of products’ and substrates’ EElectronic values , we obtain ΔEElectronic , which we treat as directly proportional to the standard reduction potential of the major species at pH 0: Eo ( MSatpH=0 ) ∼−ΔEElectronicnF The use of ΔEElectronic to approximate the reduction potential as opposed to ΔGro ( which includes rotational and vibrational enthalpies and entropies ) reduces computational cost and is motivated by the empirical observation that there is a strong correlation between ΔEElectronic and ΔGro for these redox systems ( S5 Fig , see SI for details ) . We note that we subtracted the energy of molecular hydrogen ( obtained with the same SPE model chemistry ) from ΔEElectronic in order to get redox the potentials relative to the standard hydrogen electrode . A similar approach has been used to model redox reactions in the context of organic redox flow batteries [28] . We use cheminformatic pKa estimates ( Marvin 17 . 7 . 0 , 2017 , ChemAxon ) , the extended Debye-Huckel equation and the Alberty-Legendre transform ( 16 , 17 ) to convert both the experimental standard redox potentials and the quantum chemical predictions of Eo ( MS at pH = 0 ) to the transformed redox potentials standardized to 1 mM , E′m ( pH = 7 , I = 0 . 25 ) . Then , independently for each redox category , we performed linear regressions between the E′m ( pH = 7 , I = 0 . 25 ) values and the available experimental redox potentials . The calibration via linear regression was implemented using the SciKit learn Python library . In order to optimize prediction accuracy , we ran geometry optimization and SPE calculations using a large diversity of model chemistries , generated by selecting one of ten possible DFT functionals , two wave function electronic structure methods , three possible basis sets , the option of adding implicit solvation , as well as a correction to account for dispersion interactions ( S1 Fig , and see SI for details ) . Optimizing for Pearson correlation coefficients r , we selected the following model chemistry to predict reactions without experimentally measured potentials: a DFT approach with the double-hybrid functional B2PLYP [32 , 33] , the DefBas-5 Orca basis set ( see S3 Table for detailed basis set description ) , COSMO implicit solvent [87] , and D3 dispersion correction [88] . To avoid overfitting , we trained the model chemistry optimization procedure on the experimental data for the G3 reaction category ( carbonyl to amine reduction ) , and validate its accuracy on the rest of the oxidoreductase reaction categories ( Table 1 and S3 Fig ) . Hybrid and double-hybrid DFT functionals have been shown to accurately capture the thermochemistry and noncovalent interactions of molecules when compared with coupled cluster results [90 , 91] . Therefore , we select this double-hybrid DFT approach covers the relevant physics of our problem while minimizing computational cost and maximizing predictive power . Although we explored a large set of DFT functionals , wave function methods , and basis sets , further improvements could be achieved by exploring a larger space of model chemistries , including the geometry optimization procedure , conformer generation method , as well as explicit solvation models [15] . For example , adapting a recent highly accurate method ( tested on four molecules ) based on the Linear Response Approximation ( LRA ) to the large scale prediction of E’m values would be an interesting direction [72] . We used the RDKit software tool ( http://www . rdkit . org ) , to obtain binary molecular fingerprints of each compound of interest . Because of the relatively small size of our training sets and in order to minimize overfitting , we used MACCS Key 166 fingerprints instead of other popular Morgan circular fingerprints [92] . We concatenated each redox half-reaction substrate/product fingerprint pair into a single reaction fingerprint [38] and used these as input training data for regularized linear regression . We then performed an independent regularized regression for each of the four different redox reaction categories . To obtain group contribution estimates of redox potentials , we use the group matrix and the group energies of Noor et al . [36] used in eQuilibrator [37] , an online thermodynamics calculator . We note that eQuilibrator uses the component contribution method ( CCM ) which combines group contribution energies with experimental reaction or formation Gibbs energies ( “reactant contributions” ) whenever these are available . That is , for reactions with available experimental data , eQuilibrator will return the experimental energies . Thus , for fair comparison against quantum chemistry we used the GCM code underneath eQuilibrator to obtain the group contribution estimates for all reactions in our test set . Just like the quantum chemical predictions , the GCM estimates were standardized to the E'm ( pH = 7 , I = 0 . 25 ) state . We design a strategy to detect reactions with potentially erroneous experimental values as listed in the NIST Thermodynamics of Enzyme-Catalyzed Reactions Database ( TECRDB ) [30] . We identify reactions whose predicted potential deviates from experiment by a similar amount for both the calibrated quantum chemistry and fingerprint-based modeling approaches . In order to make the errors associated to the two different modeling methods comparable , we normalize the prediction errors by computing their associated z-scores: ZErr= ( Err−μ ) σ . We set a threshold value for the z-score of Z = 2 , such that reactions with ZErr ( QC ) > 2 and ZErr ( fingerprints ) > 2 are assigned a high likelihood of having an erroneously tabulated experimental value in NIST-TECRDB . To generate a database of all possible redox reactions involving natural compounds , we use a decomposition of all metabolites into functional groups as per the group contribution method [36] . We find pairs of metabolites in the KEGG database with functional group vectors whose difference matches the reaction signature of any of the redox reaction categories of interest . For example , pairs of metabolites in the G1 category will have a group difference vector with a +1/-1 in the element corresponding to a carbonyl/carboxylic acid functional group respectively ( see SI for details ) . We note that every reaction generated by this strategy can be uniquely assigned to one of the four redox categories considered . Using this method we succeeded in generating a rough database of redox reactions . However , additional manual and semi-automated data cleansing was required to get the final version of the database ( see SI for further details ) . For example , use of the group difference vectors failed to account for the chirality of the metabolites , and in some instances stereochemistry was not maintained throughout the reaction . In order to solve this , we applied an additional filter , which used the conventions for assigning chirality ( R/S , L/D ) present in molecule names to match chirality between the substrate and product . Sugars proved to be especially problematic as those reactions did not maintain stereochemistry throughout; for these reactions , the above filtering method did not suffice , often keeping incorrect reactions such as L-Xylonate → L-Arabinose . For this , we used molecular naming conventions to eliminate the wrong reactions ( see SI for further details ) . We performed Welch’s unequal variance t-test to obtain the p-value for the null hypothesis that pairs of different reaction subcategories within group G2 have identical average E’m values ( Fig 4 ) . Welch’s t-test is an adaptation of Student’s t-test which does not assume equal variances . | Redox reactions define the energetic constraints within which life can exist . However , measurements of reduction potentials are scarce and unstandardized , and current prediction methods fall short of desired accuracy and coverage . Here , we harness quantum chemistry tools to enable the high-throughput prediction of reduction potentials with unparalleled accuracy . We calculate the reduction potentials of all redox pairs that can be generated using known biochemical compounds . This high-resolution dataset enables us to uncover global trends in metabolism , including the differences between and within oxidoreductase groups . We further demonstrate that the redox potential of NAD ( P ) optimally satisfies two constraints: reversibly reducing and oxidizing the vast majority of redox reactions in central metabolism while keeping the concentration of reactive carbonyl intermediates in check . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"quantum",
"chemistry",
"chemical",
"compounds",
"density",
"functional",
"theory",
"enzymes",
"enzymology",
"organic",
"compounds",
"carboxylic",
"acids",
"quantum",
"mechanics",
"thermodynamics",
"oxidation-reduction",
"reactions",
"proteins",
"acids",
"oxidoreductases",
"chemistry",
"oxidation",
"physics",
"biochemistry",
"electrochemistry",
"organic",
"chemistry",
"biology",
"and",
"life",
"sciences",
"chemical",
"reactions",
"physical",
"sciences",
"computational",
"chemistry"
] | 2018 | Quantum chemistry reveals thermodynamic principles of redox biochemistry |
We have carried out a comprehensive analysis of the determinants of human influenza A H3 hemagglutinin evolution . We consider three distinct predictors of evolutionary variation at individual sites: solvent accessibility ( as a proxy for protein fold stability and/or conservation ) , Immune Epitope Database ( IEDB ) epitope sites ( as a proxy for host immune bias ) , and proximity to the receptor-binding region ( as a proxy for one of the functions of hemagglutinin-to bind sialic acid ) . Individually , these quantities explain approximately 15% of the variation in site-wise dN/dS . In combination , solvent accessibility and proximity explain 32% of the variation in dN/dS; incorporating IEDB epitope sites into the model adds only an additional 2 percentage points . Thus , while solvent accessibility and proximity perform largely as independent predictors of evolutionary variation , they each overlap with the epitope-sites predictor . Furthermore , we find that the historical H3 epitope sites , which date back to the 1980s and 1990s , only partially overlap with the experimental sites from the IEDB , and display similar overlap in predictive power when combined with solvent accessibility and proximity . We also find that sites with dN/dS > 1 , i . e . , the sites most likely driving seasonal immune escape , are not correctly predicted by either historical or IEDB epitope sites , but only by proximity to the receptor-binding region . In summary , a simple geometric model of HA evolution outperforms a model based on epitope sites . These results suggest that either the available epitope sites do not accurately represent the true influenza antigenic sites or that host immune bias may be less important for influenza evolution than commonly thought .
The influenza virus causes one of the most common infections in the human population . The success of influenza is largely driven by the virus’s ability to rapidly adapt to its host and escape host immunity . The antibody response to the influenza virus is determined by the surface proteins hemagglutinin ( HA ) and neuraminidase ( NA ) . Among these two proteins , hemagglutinin , the viral protein responsible for receptor binding and uptake , is a major driver of host immune escape by the virus . Previous work on hemagglutinin evolution has shown that the protein evolves episodically [1–3] . During most seasons , hemagglutinin experiences mostly neutral drift around the center of an antigenic sequence cluster; in those seasons , it can be neutralized by similar though not identical antibodies , and all of the strains lie near each other in antigenic space [4–7] . After several seasons , the virus escapes its local sequence cluster to establish a new center in antigenic space [7–9] . There is a long tradition of research aimed at identifying important regions of the hemagglutinin protein , and by proxy , the sites that determine sequence-cluster transitions [4 , 6 , 10–21] . Initial attempts to identify and categorize important sites of H3 hemagglutinin were primarily sequence-based and focused on substitutions that took place between 1968 , the emergence of the Hong Kong H3N2 strain , and 1977 [10 , 11] . Those early studies used the contemporaneously solved protein crystal structure , a very small set of mouse monoclonal antibodies , and largely depended on chemical intuition to identify antigenically relevant amino-acid changes in the mature protein . Many of the sites identified in those studies reappeared nearly two decades later , in 1999 , as putative epitope sites with no additional citations linking them to actual immune data [4] . Those sites and their groupings are still considered the canonical immune epitope set today [3 , 16 , 22] . While the limitations of experimental techniques and of available sequence data in the early 1980’s made it necessary to form hypotheses based on chemical intuition , these limitations are starting to be overcome through recent advances in experimental immunological techniques and wide-spread sequencing of viral genomes . Therefore , it is time to revisit the question of whether or not our current understanding of the host immune response is reflected in the observed patterns of influenza hemagglutinin evolution . For example , at least one recent model has suggested that the hemagglutinin protein may evolve to modulate receptor-binding avidity rather than to modulate antibody-binding [23] . Moreoever , since the original epitope set was identified via sequence analysis , we do not even know whether bona-fide immune-epitope sites actually exist , i . e . , sites which represent a measurable bias in the host immune response . Most importantly , even if immune-epitope sites do exist and can be experimentally identified , it is possible that they do not experience more positive selection than other important sites in the protein . Some recent studies have begun to address these questions indirectly , via evolutionary analysis . For example , over the last two decades , virtually every major study on positive selection in hemagglutinin has found some but never all of the historical epitope sites to be under positive selection [3 , 16 , 18 , 19 , 23] . Furthermore , each of these studies has found a set of sites that are under positive selection but do not belong to any historical epitope . Finally , because every study identifies slightly different sites , there seems to be no broad agreement on which sites are under positive selection [12 , 16 , 18 , 19] . The sites found by disparate techniques are similar but they are never identical . To dissect the determinants of hemagglutinin evolution , we here linked several predictors , including relative solvent accessibility , the inverse distance from the receptor-binding region , and IEDB immune epitope data , to site-wise evolutionary rates calculated from all of the human H3N2 sequence data for the last 22 seasons ( 1991–2014 ) . We found that , individually , all these predictors explained approximately 15% of evolutionary rate variation . After controlling for biophysical constraints with relative solvent accessibility and function with distance to the receptor-binding region , the remaining predictive power of either IEDB or historical categories was relatively low . In addition , we found that current IEDB data does not reflect the historical epitope sites or their groups . Finally , by explicitly accounting for RSA , proximity , and host immune data , we found that we could predict nearly 35% of the evolutionary rate variation in hemagglutinin , nearly twice as much variation as could be explained by earlier models .
Our overarching goal in this study was to identify specific biophysical or biochemical properties of the mature protein that determine whether a given site will evolve rapidly or not . As a measure of evolutionary variation and selective pressure , we used the metric dN/dS . dN/dS can measure both the amount of purifying selection acting on a site ( when dN/dS ≪ 1 at that site ) and the amount of positive diversifying selection acting on a site ( when dN/dS ≳ 1 ) . For simplicity , we will refer to dN/dS as an evolutionary rate , even though technically it is a relative evolutionary rate or evolutionary-rate ratio . We built an alignment of 3854 full-length H3 sequences spanning 22 seasons , from 1991/92 to 2013/14 . We subsequently calculated dN/dS at each site , using a one-rate fixed-effects likelihood ( FEL ) model as implemented in the software HyPhy [24] . Several recent works have shown that site-specific evolutionary variation is partially predicted by a site’s solvent exposure and/or number of residue-residue contacts in the 3D structure [19 , 20 , 25–30] ( see Ref . [31] for a recent review ) . This relationship between protein structure and evolutionary conservation likely reflects the requirement for proper and stable protein folding: Mutations at buried sites or sites with many contacts are more likely to disrupt the protein’s conformation [30] or thermodynamic stability [32] . In addition , there may be functional constraints on site evolution . For example , regions in proteins involved in protein–protein interactions or enzymatic reactions are frequently more conserved than other regions [27 , 33 , 34] . However , these structural and functional constraints generally predict the amount of purifying selection expected at sites , and therefore they cannot identify sites under positive diversifying selection . Moreover , the short divergence time of viruses causes the systematic biophysical pressures that predict much of eukaryotic protein evolution to be much less dominant in viral evolution [28] . Thus , we set out to find a constraint on hemagglutinin evolution that was related to the protein’s role in viral binding and fusion . A few earlier studies had shown that sites near the sialic acid-binding region of hemagglutinin tend to evolve more rapidly than the average for the protein [4 , 20 , 21] . Furthermore , when mapping evolutionary rates onto the hemagglutinin structure , we noticed that the density of rapidly evolving sites seemed to increase somewhat towards the receptor-binding region ( Fig 1A ) . Therefore , as the primary function of hemagglutinin is to bind to sialic acid and induce influenza uptake , we reasoned that distance from the receptor-binding region of HA might serve as a predictor of functionally driven HA evolution . We calculated distances from the sialic acid-binding region ( defined as the distance from site 224 in HA ) , and correlated these distances with the evolutionary rates at all sites . We found that distance from the receptor-binding region was a strong predictor of evolutionary rate variation in hemagglutinin ( Pearson correlation r = 0 . 41 , P < 10−15 ) . Next , we wanted to verify that this correlation was representative of hemagglutinin evolution and not just an artifact of the specific site chosen as the reference point in the distance calculations . It would be possible , for example , that distances to several spatially separated reference sites all resulted in similarly strong correlations . We addressed this question systematically by making , in turn , each individual site in HA the reference site , calculating distances from that site to all other sites , and correlating these distances with evolutionary rate . We then mapped these correlations onto the structure of hemagglutinin , coloring each site according to the strength of the correlation we obtained when we used that site as reference in the distance calculation ( Fig 1B ) . We obtained a clean , gradient-like pattern: The correlations were highest when we calculated distances relative to sites near the receptor-binding site ( with the maximum correlation obtained for distances relative to site 224 ) , and they continuously declined and then turned negative the further we moved the reference site away from the apical region of hemagglutinin ( Fig 1B ) . This result was in stark contrast to the pattern we had previously observed when mapping evolutionary rate directly ( Fig 1A ) . In that earlier case , while there was a perceptible preference of faster evolving sites to fall near the receptor-binding site , the overall distribution of evolutionary rates along the structure looked mostly random to the naked eye . We thus found a geometrical , distance-based constraint on hemagglutinin evolution: Sites evolve faster the closer they lie toward the receptor-binding region . We also evaluated how proximity to the receptor-binding region performed as a predictor of dN/dS in comparison to the previously proposed structural predictors relative solvent accessibility ( RSA ) and weighted contact number ( WCN ) . We found that among these three quantities , proximity to the sialic acid-binding region was the strongest predictor , explaining 16% of the variation in dN/dS ( Pearson r = 0 . 41 , P < 10−15 , see also Fig 2 and S1 Fig ) . RSA and WCN explained 14% and 6% of the variation in dN/dS , respectively ( r = 0 . 37 , P < 10−15 and r = 0 . 25 , P = 7 × 10−9 ) . Proximity to the sialic acid-binding region and RSA were virtually uncorrelated ( r = 0 . 08 , P = 0 . 09 ) while RSA and WCN correlated strongly ( r = −0 . 64 , P < 10−15 ) . These results suggested that proximity to the sialic acid-binding region and RSA should be used jointly in a predictive model . Because hemagglutinin has , in addition to its function as a receptor-binding protein , a host of other intermediate functional states during the viral fusion process , we also tested the ability of structural metrics from the post-fusion state to predict hemagglutinin evolutionary rate [35] . We found no significant metric , either RSA or proximity , derived from the post-fusion state . ( Complete data and analysis scripts are available in the accompanying Github repository , see Methods for details . ) Another potential functional constraint on hemagglutinin evolution is a bias in the human immune system . This bias , generally referred to as antigenicity , describes the extent to which the human immune system does a better job attacking one region of a protein compared to another . Conventional wisdom states that functionally important sites in the protein that are targeted by antibodies will evolve more rapidly to facilitate immune escape . And indeed , our results from the previous subsection have shown that proximity to the receptor-binding region is a good predictor of evolutionary variation . However , if substitutions to avoid direct antibody binding are the primary cause of positive selection , then we would expect antigenic sites on hemaggalutinin to serve as a substantially better predictors of adaptation than proximity to the receptor-binding site alone . For influenza hemagglutinin H3 , there exists a list of canonical , historical epitope sites that are commonly considered to represent this bias [4] . However , these sites were not primarily defined based on actual immunological data , and they have not been re-validated since the late 1990s even though more experimental data is now available . ( See Discussion for details on the history of the historical epitope sites . ) Before we could generate a combined evolutionary model , we therefore considered it essential to validate the antigenic groups with available immunological data . As it turns out , the majority of antigenic data available did not agree with the historical epitope sites ( S1 Text ) . Therefore , we used both the historical epitope sites and a set of IEDB re-defined epitopes for further modeling . A detailed explanation of our re-grouping based on IEDB data is available in S1 Text . It is important to note that these groups are not intended to represent a new canonical set of hemagglutinin epitopes . Indeed , the data from which they were derived is limited and relatively poorly annotated . However , considering the magnitude of the difference between the historical epitopes and the available IEDB data we considered it imperative to include IEDB derived epitopes in our analysis . Thus , we considered both the historical epitope groups ( Bush 1999 ) and the IEDB derived epitopes 1–4 , defined in S1 Text . Because a site’s epitope status is a categorical variable , we calculated variance explained as the coefficient of determination ( R2 ) in a linear model with dN/dS as the response variable and epitope status as the predictor variable . We found that IEDB epitopes explained 15% of the variation in dN/dS , comparable to RSA and proximity . In comparison , the historical epitopes alone explained nearly 18% of the variation in dN/dS , outperforming all other individual predictor variables considered here ( Fig 2 and Table 1 ) . However , as discussed in S1 Text , the available IEDB data suggest that not all of the historical sites may be actual immune epitope sites . Therefore , we suspected that some of the predictive power of historical sites was due to these sites simply being solvent-exposed sites near the receptor-binding region . We similarly wondered to what extent the predictive power of the IEDB epitope sites was attributable to the same cause , since , in fact , both historical and IEDB epitope sites showed comparable enrichment in sites near the sialic acid-binding region and in solvent-exposed sites ( S2 Fig ) . Therefore , we analyzed how the variance explained increased as we combined epitope sites ( IEDB or historical ) with either RSA or proximity or both . We found that epitope status , under either definition ( IEDB/historical ) , led to increased predictive power of the model when combined with either RSA or proximity ( Fig 2 ) . However , a model consisting of just the two predictors RSA and proximity , not including any information about epitope status of any sites , performed even better than any of the other one- or two-predictor models , explaining 32% of the variation in dN/dS ( Fig 2 ) . Adding epitope status to this best-performing two-predictor model resulted in only minor improvement , from 32% to 34% variance explained in the case of IEDB epitopes and from 32% to 37% variance explained in the case of historical epitope sites ( Fig 2 and Table 1 ) . The geometrical constraints RSA and proximity explained more variance in dN/dS than did epitope sites , but were they also better at predicting sites of interest ? Because dN/dS can measure purifying as well as positive diversifying selection , the percent variance in dN/dS that a model explains may not necessarily accurately reflect how useful that model is in predicting specific sites , e . g . sites under positive selection . For example , one could imagine a scenario in which a model does exceptionally well on sites under purifying selection ( dN/dS ≪ 1 ) but fails entirely on sites under positive selection ( dN/dS > 1 ) . Such a model might explain a large proportion of variance but be considered less useful than a model that overall predicts less variation in dN/dS but accurately pinpoints site under positive selection . Therefore , we wondered whether epitope sites might do a poor job predicting background purifying selection but might still be useful in predicting sites with dN/dS > 1 . We found , to the contrary , that neither the historical nor the IEDB epitope sites could reliably predict sites with dN/dS > 1 , alone or in combination with RSA ( Fig 3A–3D ) . Proximity to the receptor-binding site , on the other hand , correctly predicted four sites with dN/dS > 1 , even in the absence of any other predictors . Notably , all models we considered here were robust to cross-validation . The cross-validated residual standard error was virtually unchanged from its non-cross-validated value in all cases ( Table 1 ) . Because proximity clearly identified four points with high dN/dS , we also verified that the proximity–dN/dS correlation was not caused just by these four points . We removed from our data set the four points that had both predicted and observed dN/dS > 1 , and found that a significant proximity–dN/dS correlation remained nonetheless ( r = 0 . 17 , p = 0 . 00001 ) . Finally , we compared the predictions from the geometrical model of hemagglutinin evolution to results from a recent study of antigenic cluster transitions; that study found seven sites near the receptor-binding region which were critical for cluster transitions according to hemagglutinin inhibition ( HI ) assays with ferret antisera [21] . The sites identified in Ref . [21] were 145 , 155 , 156 , 158 , 159 , 189 , and 193 . For comparison , our geometric model ( with predictors RSA and 1/Distance ) predicted none of these sites to be under positive selection . Sites predicted to have dN/dS > 1 were instead 96 , 137 , 138 , 143 , 222 , 223 , 225 , and 226 . Moreover , out of the seven sites from Ref . [21] , only one ( site 145 ) had an observed dN/dS significantly above 1 . By contrast , four of the eight sites predicted under the geometric model to have dN/dS > 1 did indeed have dN/dS significantly above 1 . Thus , the sites that determine the major antigenic changes in the virus did not at all overlap with the sites expected and observed to be under the greatest evolutionary pressure . When investigating the location of these sites in detail , we found that all of the sites we predicted to have dN/dS > 1 were located just basal to the receptor-binding site , whereas nearly all of the sites from [21] ( with the exception of 145 , the site with dN/dS > 1 ) were located on the apical side of the receptor-binding site ( Fig 4 ) . In summary , we have found that two simple geometric measures of a site’s location in the 3D protein structure , solvent exposure and proximity to the receptor-binding region , jointly outperformed , by a wide margin , any previously considered predictor of evolutionary variation in hemagglutinin , including immune epitope groups . In fact , the vast majority of the variation in evolutionary rate that was explained by the historical epitope sites was likely due to these sites simply being located near the receptor-binding region on the surface of the protein . However , historical epitope sites , in combination with solvent exposure and proximity , had some residual explanatory power beyond even a three-predictor model that combined the two geometric measures with IEDB immune-epitope data . We suspect that this residual explanatory power reflects the sequence-based origin of the historical epitope sites . To our knowledge , the historical epitope sites were at least partially identified by observed sequence variation , so that , to some extent , these sites are simply the sites that have been observed to evolve rapidly in hemagglutinin .
Efforts to define immune epitope sites in H3 hemagglutinin go back to the early 1980’s [10] . Initially , epitope sites were identified primarily by speculating about the chemical neutrality of amino acid substitutions between 1968 ( the year H3N2 emerged ) and 1977 , though some limited experimental data on neutralizing antibodies was also considered [10 , 11] . In 1981 , the initial four epitope groups were defined by non-neutrality ( amino-acid substitutions that the authors believed changed the chemical nature of the side chain ) and relative location , and given the names A through D [10] . Since that original study in 1981 , the names and general locations of H3 epitopes have remained largely unchanged [4 , 16] . The sites were slightly revised in 1987 by the same authors and an additional epitope named E was defined [11] . From that point forward until 1999 there were essentially no revisions to the codified epitope sites . In addition , while epitopes have since been redefined by adding or removing sites , no other epitope groups have been added [3 , 16 , 18]; epitopes are still named A–E . In 1999 , the epitopes were redefined by more than doubling the total number of sites and expanding all of the epitope groups [4] . At that time , the redefinition consisted almost entirely of adding sites; very few sites were eliminated from the epitope groups . Although this set of sites and their groupings remain by far the most cited epitope sites , it is not particularly clear what data justified this definition . Moreover , when the immune epitope database ( IEDB ) summarized the publicly available data for influenza in 2007 , it only included one IEDB B cell epitope in humans ( Table 2 in [36] ) . Although there were a substantial number of putative T cell epitopes in the database , a priori there is no reason to expect a T cell epitope to show preference to hemagglutinin as opposed to any other influenza protein; yet it is known that several other influenza proteins show almost no sites under positive selection . Moreover , it is known that the B cell response plays the biggest role is maintaining immunological memory to influenza , and thus it is the most important arm of the adaptive immune system for influenza to avoid . The historical H3 epitope sites have played a crucial role in molecular evolution research . Since 1987 , an enormous number of methods have been developed to analyze the molecular evolution of proteins , and specifically , to identify positive selection . The vast majority of these methods have either used hemagglutinin for testing , have used the epitopes for validation , or have at some point been applied to hemagglutinin . Most importantly , in all this work , the epitope definitions have been considered fixed . Most investigators simply conclude that their methods work as expected because they recover some portion of the epitope sites . Yet virtually all of these studies identify many sites that appear to be positively selected but are not part of the epitopes . Likewise , there is no single study that has ever found all of the epitope sites to be important . Even if the identified sites from all available studies were aggregated , we would likely not find every site among the historical epitopes in that aggregated set of sites . Given all of this research activity , it seems that the meaning of an immune epitope has been muddled . Strictly speaking , an immune epitope is a site to which the immune system reacts . There is no a priori reason why an immune epitope needs to be under positive selection , needs to be a site that has some number or chemical type of amino acid substitutions , or needs to be predictive of influenza whole-genome or hemagglutinin-specific sequence cluster transitions . Yet , from the beginning of the effort to define hemagglutinin immune epitopes , such features have been used to identify epitope sites , resulting in a set of sites that may not accurately reflect the sites against which the human immune system produces antibodies . Ironically , this methodological confusion has actually been largely beneficial to the field of hemagglutinin evolution . As our data indicate , if the field had been strict in its pursuit of immune epitopes sites , it would have been much harder to produce predictive models with those sites , in particular given that IEDB data on non-linear epitopes have been sparse until very recently . By contrast , the historical epitope sites have been used quite successfully in several predictive models of the episodic nature of influenza sequence evolution . In fact , in our analysis , historical epitopes displayed the highest amount of variance explained among all individual predictors ( Fig 2 ) . We argue here that the success of historical epitope sites likely stems from the fact that they were produced by disparate analyses each of which accounted for a different portion of the evolutionary pressures on hemagglutinin . Of course , it is important to realize that some of this success is likely the result of circular reasoning , since the sites themselves were identified at least partially from sequence analysis that included the clustered , episodic nature of influenza hemagglutinin sequence evolution . Despite the success of historical epitope groups , they only predict about 18% of the evolutionary rate-variation of hemagglutinin for the entire phylogenetic tree . Since many of these sites likely are not true immune epitopes ( and therefore not host dependent ) , one might ask which features of the historical epitope sites make them good predictors . We suspect that they perform well primarily because they are a collection of solvent-exposed sites near the sialic acid-binding region ( see S2 Fig ) . We had shown previously that sites within 8 Šof the sialic acid-binding site are enriched in sites under positive selection , compared to the rest of the protein [20] . A similar result was found in the original paper by Bush et al . [4] . However , the related metric of distance from the sialic acid-binding site has not previously been considered as a predictor of evolution in hemagglutinin . Furthermore , before 1999 , most researchers thought the opposite should be true; that receptor-binding sites should have depressed evolutionary rates [4] . Even today the field seems split on the matter [21] . As we have shown here , the inverse of the distance from sialic acid is a relatively strong quantitative predictor of hemagglutinin evolution; by itself this distance metric can account for 16% of evolutionary rate-variation . Moreover , by combining this one metric with another to control for solvent exposure , we can account for more than a third of the evolutionary rate variation in hemagglutinin . For reference , this number is larger than the variation one could predict by collecting and analyzing all of the hemagglutinin sequences that infect birds ( another group of animals with large numbers of natural influenza infections ) , and using those rates to predict human influenza hemagglutinin evolutionary rates [20] . In terms of re-grouping IEDB immune data , it is important to note that the IEDB has major limitations; not all existing ( not to mention all possible ) immunological data have been added . Further , the extent to which certain epitopes ( e . g . , stalk epitopes ) have been mapped may be more reflective of a bias in research interests among influenza researchers than a bias in the human immune system . Also , until recently , the ability to generate unbiased high-affinity antibodies to influenza has been limited [37 , 38] . Therefore , in our re-derivation of epitope groupings , we are certainly missing sites or may be incorrectly grouping the ones that we have . Our analysis of epitope sites will likely have to be redone as more data become available . However , we expect that as more non-linear data become available , they will broadly follow the trend observed in the linear epitope data , that is , the more antibodies are mapped , the more sites in the hemagglutinin protein appear in at least one mapping , until virtually every site in the entire hemagglutinin protein is represented . Under this scenario , the ability to predict evolution from immunological data would become worse , not better , as more data are accumulated . One additional caveat comes from any potential effect of glycosylation on influenza immune escape . Glycosylations on hemagglutinin can have a major effect on receptor and antibody binding [13] . In addition , the number of glycosylations in H3 hemagglutinin has increased since initial introduction of pandemic H3N2 in 1968 [13] . However , a priori there is no reason to believe that glycosylation will either increase or decrease dN/dS at individual sites or groups of sites; it could affect dN/dS in either direction , in particular if direct antibody escape is not the primary driver of hemagglutinin evolution . Moreover , there is no clear way to incorporate glycosylation into our regression model . In the future , investigating changing glycosylation patterns throughout the evolution of H3 hemagglutinin may yield important insights into influenza adaptation and immune escape . Why do geometric constraints ( solvent exposure and proximity to receptor-binding site ) do a good job predicting hemagglutinin evolutionary rates ? Hemagglutinin falls into a class of proteins known collectively as viral spike glycoproteins ( GP ) . In general , the function of these proteins is to bind a host receptor to initiate and carry out uptake or fusion with the host cell . Therefore , a priori one might expect that the receptor-binding region would be the most conserved part of the protein , since binding is required for viral entry . Yet , in hemagglutinin sites near the binding region are the most variable in the entire protein . There are at least two possible models that might explain this observation . First , conventional wisdom says that in terms of host immune evasion , antibodies that bind near the receptor-binding region may be the most inhibitory , and hence mutations in this region the most effective in allowing immune escape . Viral spike GPs have a surface that is both critical for viral survival and is sufficiently long lived that a host immune response is easily generated against it . There are likely many other viral protein surfaces that are comparatively less important or sufficiently short lived during a conformational change that antibody neutralization is impractical . Thus , the virions that survive to the next generation are those with substantial variation at the surface or surfaces with high fitness consequences and a long half-life in vivo . Evolutionary variation at surfaces with low or no fitness consequences , or at short-lived surfaces , should behave mostly like neutral variation and hence appear as random noise , not producing a consistent signal of positive selection . Second , according to the avidity modulation model of Hensley et al . [23] , it is possible that antibody inhibition is not overcome by escaping the antibody directly . Considering the fact that neither historical nor IEDB immune epitopes vastly out-performed our simple distance metric , we think that our results support a model which does not expect an evolutionary bias based on antibody binding sites . However , it remains a possibility that the historical epitopes and current IEDB data are simply wrong about which sites and groups of sites the human immune system attacks . Either way , our work highlights the need for a paradigm shift in the field . We also need to consider that actual epitope sites , i . e . , sites toward which the immune system has a bias , may not be that important for the evolution of viruses . An epitope is simply a part of a viral protein to which the immune system reacts . Therefore , it represents a host-centered biological bias . The virus may experience stronger selection at regions with high fitness consequences but that generate a relatively moderate host response compared to other sites with low fitness consequences that generate a relatively strong host response . Moreover , there is little reason to believe that influenza must escape an antibody by directly reducing the binding of that antibody . There are other possible scenarios for immune evasion , e . g . avidity modulation as stated above . Thus , we expect that the geometric constraints we have identified here will be more useful in future modeling work than the IEDB epitope groups we have defined . Moreover , we expect that similar geometrical constraints will exist in other viral spike glycoproteins , and in particular in other hemagglutinin variants . By contrast to the clear geometric constraints we observed for the pre-fusion structure , we found no comparable result for the post-fusion structure . There are perhaps several good reasons to expect this result . First , the transition state is likely very short-lived , such that the human immune system is not able to generate antibodies against it . Second , due to the short-lived functional nature of the transition state , there is likely relatively little selection for folding stability . Therefore , for the post-fusion structure we do not expect to observe the RSA–rate correlation that exists in the pre-fusion structure and in most other proteins . Third , models describing the transition from the pre-fusion to the post-fusion state show that the HA1 chain dissociates from the HA2 chain [39] . Subsequently , the HA2 chain carries out virtually all of the fusogenic functions . Thus , the HA1 chain is likely the functional unit in the first step of entry and the HA2 chain is likely the functional unit in the second . However , there is almost no rapid evolution happening in the HA2 chain , i . e . , the HA2 chain does not seem to experience any positive diversifying selection . Remarkably , the sites we found that experienced the most positive selection showed minimal overlap with the sites found to be minimally sufficient for explaining the major antigenic transitions in H3N2 , as determined by HI assays with ferret antisera [21] . While both groups of sites lie near the sialic-acid binding region , the vast majority of positively selected sites are located basally to sialic acid whereas sites identified by HI assays lie predominantly on the apical side ( Fig 4 ) . This finding suggests that HI assays and positive selection analyses reflect distinct biological mechanisms . For example , HI assays might not accurately reflect selection pressures in vivo . Such a result would suggest that influenza is not under pressure to directly escape antibody binding . Alternatively , it is possible that the standard manner for obtaining ferret antisera simply may not represent a good proxy for the cyclical nature of human influenza infections [40] . Indeed , recent evidence suggests that , at least for the pandemic H1N1 strain , cyclical infections can shift the antibody response toward the receptor-binding region [41] . In future work , disentangling the different mechanisms reflected by HI assays and by positive-selection analyses will likely be crucial for improved prediction of HA evolution and of optimal vaccine strains .
All of the data we analyzed were taken from the Influenza Research Database ( IRD ) [42] . The IRD provides IEDB immune data curated from the data available in the Immune Epitope Database [43] . We used sequences that had been collected since the 1991–1992 influenza season . Any season before the 1991–1992 season had an insufficient number of sequences to contribute much to the selection analysis . The sequences were filtered to remove redundant sequences and laboratory strains . The sequences were then aligned with MAFFT [44] . Since it is known that there have been no insertions or deletions since the introduction of the H3N2 strain , we imposed a strict opening penalty and removed any sequences that had intragenic gaps . In addition , we manually curated the entire set to remove any sequence that obviously did not align to the vast majority of the set; in total the final step only removed about 10 sequences from the final set of 3854 sequences . For the subsequent evolutionary rate calculations , we built a tree with FastTree 2 . 0 [45] . To compute evolutionary rates , we used a fixed effects likelihood ( FEL ) approach with the MG94 substitution model [24 , 46 , 47] . We used the FEL provided with the HyPhy package [24] . For the full setup see the linked GitHub repository ( https://github . com/wilkelab/influenza_HA_evolution ) . As is the case for all FEL models , an independent evolutionary rate is fit to each site using only the data from that column of the alignment . Because our data set consisted of nearly 4000 sequences , almost every site in our alignment had a statistically significant posterior probability of being either positively or negatively selected after adjusting via the false discovery rate ( FDR ) method . As shown in Fig 3 , all evolutionary rates fall into a range between dN/dS = 0 and dN/dS = 4 . We computed RSA values as described previously [28] . Briefly , we used DSSP [48] to compute the solvent accessibility of each amino acid in the hemagglutinin protein . Then , we used the maximum solvent accessibilities [49] for each amino acid to normalized the solvent accessibilities to relative values between 0 and 1 . We found that RSA calculated in the trimeric state produced better predictions than RSA calculated in the monomeric state . Thus , we used multimeric RSA in all models in this study . Both multimeric and monomeric RSA are included in the supplementary data . To create the structural heat map of correlations shown in Fig 1B , we first needed to calculate the correlations between evolutionary rates and pairwise distances , calculated in turn for each location in the protein structure as the reference point for the distance calculations . Conceptually , we can think of this analysis as overlaying a grid on the entire protein structure , where we first calculate the distance to various grid points from every Cα in the entire protein , and then compute the correlation between the set of distances to the sites on the grid and the evolutionary rate at those sites . In practice , we calculated the distance from each Cα to every other Cα . We then colored each residue by the correlation obtained between evolutionary rates and all distances to its Cα . All statistical analyses were performed using R [50] . We built the linear models with both the lm ( ) and glm ( ) functions . For cross validation , we used the cv . glm ( ) function within the boot package . Residual standard error values were computed by taking the square root of the delta value from cv . glm ( ) . With the exception of graph visualizations , all figures in this manuscript were created using ggplot2 [51] . A complete data set including evolutionary rates , epitope assignments , RSA , and proximity to the receptor-binding site is available as S1 Dataset . Raw data and analysis scripts are available at https://github . com/wilkelab/influenza_HA_evolution . In the repository , we have included all human H3 sequences from the 1991–1992 season to present combined into a single alignment . We have cleaned the combined data to only include sequences with canonical bases , non-repetitive sequences , and we have hand filtered the data to ensure all included sequences align appropriately to the 566 known amino acid sites . In addition , we have built a tree and visually verified that there were no outlying sequences on the tree for the combined set . The site-wise numbering for the H3 hemagglutinin protein reflects the numbering of the mature protein; this numbering scheme requires the removal of the first 16 amino acids in the full-length gene . Thus , for protein numbering purposes , site number 1 is actually the 17th codon in full-length gene numbering . The complete length of the H3 hemagglutinin gene is 566 sites while the total length of the protein is 550 sites . It is important to point out that the mature H3 protein has two chains ( HA1 and HA2 ) that are produced by cutting the presursor ( HA0 ) protein between sites 329 and 330 in protein numbering . In addition , as a result of cloning and experimental diffraction limitations , most ( or likely all ) hemagglutinin structures do not include some portion of the first or last few amino acids of either chain of the mature protein , and crystallographers always remove the C-terminal transmembrane span from HA2 . For example , the structure we used ( PDBID: 4FNK ) in this study does not include the first 8 amino acids of HA1 , the last 3 amino acids of HA1 , or the last 48 amino acids of HA2 . As a result , HA1 includes sites 9–326 and HA2 includes sites 330–502 . The complete data table in the project repository lists the gene sequence from one of the three original H3N2 ( Hong Kong flu ) hemagglutinin ( A/Aichi/2/1968 ) , the gene numbering , the protein numbering , the numbering of one H3N2 crystal structure , historical immune epitope sites from 1981 , 1987 and 1999 , and every calculated parameter used ( and many others than were not used ) in this study . In general , the most common epitope definitions in use today are those employed by Bush et . al 1999 [4] . Throughout this work , we refer to the Bush et . al 1999 epitopes as the“historical epitope sites” . | The influenza virus is one of the most rapidly evolving human viruses . Every year , it accumulates mutations that allow it to evade the host immune response of previously infected individuals . Which sites in the virus’ genome allow this immune escape and the manner of escape is not entirely understood , but conventional wisdom states that specific “immune epitope sites” in the protein hemagglutinin are preferentially attacked by host antibodies and that these sites mutate to directly avoid host recognition; as a result , these sites are commonly targeted by vaccine development efforts . Here , we combine influenza hemagglutinin sequence data , protein structural information , IEDB immune epitope data , and historical epitopes to demonstrate that neither the historical epitope groups nor epitopes based on IEDB data are crucial for predicting the rate of influenza evolution . Instead , we find that a simple geometrical model works best: sites that are closest to the location where the virus binds the human receptor and are exposed to solvent are the primary drivers of hemagglutinin evolution . There are two possible explanations for this result . First , the existing historical and IEDB epitope sites may not be the real antigenic sites in hemagglutinin . Second , alternatively , hemagglutinin antigenicity may not be the primary driver of influenza evolution . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Geometric Constraints Dominate the Antigenic Evolution of Influenza H3N2 Hemagglutinin |
Phenotypic variation between individuals of a species is often under quantitative genetic control . Genomic analysis of gene expression polymorphisms between individuals is rapidly gaining popularity as a way to query the underlying mechanistic causes of variation between individuals . However , there is little direct evidence of a linkage between global gene expression polymorphisms and phenotypic consequences . In this report , we have mapped quantitative trait loci ( QTLs ) –controlling glucosinolate content in a population of 403 Arabidopsis Bay × Sha recombinant inbred lines , 211 of which were previously used to identify expression QTLs controlling the transcript levels of biosynthetic genes . In a comparative study , we have directly tested two plant biosynthetic pathways for association between polymorphisms controlling biosynthetic gene transcripts and the resulting metabolites within the Arabidopsis Bay × Sha recombinant inbred line population . In this analysis , all loci controlling expression variation also affected the accumulation of the resulting metabolites . In addition , epistasis was detected more frequently for metabolic traits compared to transcript traits , even when both traits showed similar distributions . An analysis of candidate genes for QTL-controlling networks of transcripts and metabolites suggested that the controlling factors are a mix of enzymes and regulatory factors . This analysis showed that regulatory connections can feedback from metabolism to transcripts . Surprisingly , the most likely major regulator of both transcript level for nearly the entire pathway and aliphatic glucosinolate accumulation is variation in the last enzyme in the biosynthetic pathway , AOP2 . This suggests that natural variation in transcripts may significantly impact phenotypic variation , but that natural variation in metabolites or their enzymatic loci can feed back to affect the transcripts .
A longstanding goal in genetics is to unravel the molecular and genetic bases of complex traits such as disease resistance , growth , and development . While phenotypic variation in natural populations is largely quantitative and polygenic , understanding this variation is complicated by the interaction of environmental and genetic factors [1 , 2] . Quantitative trait mapping , the most common approach to analyze complex traits , measures the association of genetic markers with phenotypic variation , delineating quantitative trait loci ( QTLs ) [1 , 3] . Advances in statistical models , improvements in marker technology , and expanding genomic resources have lead to increasingly refined QTL maps for a wide array of traits , ranging from development and morphology to metabolism and disease resistance [4–10] . In spite of these considerable efforts , the molecular basis of many quantitative traits remains unknown . Recently , our understanding of quantitative traits has been enhanced by genomic approaches that use microarray technology to measure global transcript levels in mapping populations and map expression QTL ( eQTL ) [11–13] . Whole-genome eQTL analysis in yeast , mice , and humans has revealed that gene expression traits are highly heritable , and can have surprisingly complex underlying genetic architecture [13–15] . Recently , similar global analysis of gene expression was conducted in two independent A . thaliana recombinant inbred line ( RIL ) mapping populations [16 , 17] . These studies revealed large numbers of both cis- and trans-acting eQTL , with evidence of nonadditive genetic variation and transgressive segregation , consistent with results from animal systems . In addition , network eQTL analysis in the Bay × Sha RIL population showed that transcript variation was controlled by variation in specific biological networks including both biosynthetic and signal transduction pathways [18] . These studies present a detailed picture of variation in gene expression and its underlying genetic architecture , but the relationship between transcript levels and the resultant phenotypic variation remains poorly understood . Testing the connection between eQTLs and downstream phenotypic variation requires a phenotype with detailed molecular and quantitative genetic information . Metabolic phenotypes are ideal for these studies , because these traits are highly variable and can be accurately measured using high-throughput techniques [5 , 19 , 20] . Knowledge of biochemical pathways enables comparison between the transcript level of a biosynthetic gene and downstream metabolic phenotypes . This engenders detailed hypotheses about the basis of metabolic variation that incorporate biochemical relationships , flux concepts , and transcriptional regulation [21] . Derived traits generated from the raw metabolite accumulation data can provide unique insights into the metabolic network [22 , 23] . These derived traits can include the sum of related metabolites , providing information about the whole pathway , or the ratio of two metabolites related as precursor and product can be used to query variation in a specific enzymatic process . We test our ability to link eQTLs with phenotypic variation using two secondary metabolite pathways responsible for the synthesis of aliphatic and indolic glucosinolates within A . thaliana . These metabolites play an important role in plant defense against herbivory , and have chemopreventive activity in the human diet . An improved understanding of the genetic basis of glucosinolate variation thus affects evolution and ecology as well as nutrition and agriculture [24 , 25] . Glucosinolates are synthesized by a well-studied biosynthetic pathway [24 , 26 , 27] , with known transcription factors [28 , 29] and cloned QTLs controlling structural diversity and content within Arabidopsis [19 , 30] ( Figure 1 ) . Aliphatic and indolic glucosinolates , derived from elongated methionine derivatives and tryptophan , respectively , are synthesized and subsequently modified by two independent yet parallel pathways ( Figure 1A ) . These biosynthetic pathways possess distinct enzymes and divergent regulation [27] . Production of aliphatic glucosinolates is controlled by three cloned QTLs controlling specific biosynthetic enzymes: GSL . Elong , GSL . ALK , and GSL . OX ( GSL = GlucoSinoLate; Figure 1B and 1C ) [31–33] . Additional QTLs have been identified which , according to current knowledge , are not associated with known biosynthetic genes [30 , 34] . As such , the aliphatic and indolic glucosinolate metabolic pathways provide a useful model system to link phenotypic QTLs with eQTLs . To compare phenotypic QTLs and eQTLs , we measured the accumulation of aliphatic and indolic glucosinolates in the Bay × Sha RIL population . In addition to 14 and 11 metabolic QTL for the indolic and aliphatic metabolites , respectively , several epistatic interactions were detected . Using the same seeds and developmental stages as the previous eQTL mapping in the Bay × Sha RIL population allowed us to compare metabolite QTL and eQTL locations [17] . Comparing metabolite QTLs with network eQTLs controlling the expression of aliphatic and indolic glucosinolate biosynthetic pathways indicates that all network eQTLs colocate with metabolic QTLs , but the inverse statement is not true , as some QTLs are detected only for metabolite traits . We obtained evidence that variation in biosynthetic enzymes and possibly transcription factors can control natural variation for transcript levels of glucosinolate biosynthetic genes . The heritability of metabolic traits was on average lower than that for transcript levels , suggesting that metabolite accumulation may be more susceptible to environmental factors . This detailed picture of glucosinolate accumulation and modification shows that eQTLs can be associated with changes in the resulting phenotype , allowing us to generate testable mechanistic hypotheses regarding the interplay between expression variation and downstream phenotypic variation .
We measured glucosinolate production by the A . thaliana accessions , Bayreuth ( Bay-0 ) and Shahdara ( Sha ) , the parents of the Bay × Sha RIL population . The Bay-0 and Sha accessions differed in both the quantity of glucosinolates accumulated and the specific structures synthesized , verifying that the Bay-0 × Sha RIL population is potentially informative for analyzing the relationship of eQTLs to metabolic variation . The glucosinolate profile of Sha is similar to that previously published for the Cape Verdi Islands ( Cvi-1 ) accession , which forms predominantly three carbon ( C3 ) and four carbon ( C4 ) alkenyl glucosinolates , with high total aliphatic glucosinolate content ( Tables 1 , S1 , and S2 ) [19 , 30] . In contrast , Bay-0 resembles Landsberg erecta ( Ler ) , which contains mostly C3 hydroxy glucosinolates , with lower total aliphatic glucosinolate content ( Table 1 ) [19 , 30] . The parental accessions also differed in partitioning of indolic glucosinolates into different structures , with Bay-0 producing significantly more 4-methoxy-indol-3-ylmethyl glucosinolate ( Table 1 ) . We measured the average glucosinolate content within the Bay-0 × Sha RILs and compared the trait distribution among 403 RILs to the Bay-0 and Sha parental means ( Table 1 ) . Some RILs accumulated two aliphatic glucosinolates ( 4-methylsulfinylbutyl [4-MSO] and 4-methylthiobutyl [4-MT] ) that are not found in the parental accessions . Transgressive segregation for this biosynthetic capability was previously observed in the Ler x Cvi RIL population and shown to result from epistasis between the GSL . AOP and GSL . Elong loci [6 , 30–32 , 35–37] . In the Sha parent , the AOP2 enzyme fully converts all 4-MSO into but-3-enyl glucosinolate , preventing the detectable accumulation of 4-MSO within Sha . In Bay-0 , 4-MSO does not accumulate due to the Bay-0 allele at Elong , preventing the formation of 4C glucosinolates . Plants containing the GSL . AOPOHP ( AOP3 ) allele from the Bay-0 parent and the GSL . ElongC4 ( MAM1 ) allele from Sha accumulate 4-MSO and 4-MT because the AOP3 enzyme expressed by the GSL . AOPOHP allele from Bay-0 can not convert the 4-MSO precursor to hydroxyl glucosinolates ( Figure 1 ) [31] . In addition to transgressive segregation for biosynthetic potential , there is transgressive segregation for glucosinolate levels . For this population , the transgressive segregation for the quantity of glucosinolates produced is almost entirely negative , as the RIL population includes numerous lines producing less total aliphatic glucosinolate than either parental accession , but no lines that accumulate average levels higher than the Sha parent ( Table 1 ) . This is especially striking for total indolic glucosinolate content , where all of the RILs were significantly lower than both parents . Because the Bay-0 and Sha parents were grown concurrently with the RILs , this is not due to environmental effects . This observation of negative transgressive segregation contrasts with the Ler × Cvi RIL population , where both positive and negative transgressive segregation was observed , with RILs producing both greater and lesser quantities of glucosinolates than either parental accession [30] . Given that the GSL . AOP and GSL . ELONG loci are in common between both populations , this difference in transgressive segregation suggests the influence of additional QTLs that are not variable in both populations . To compare the underlying genetics controlling metabolite variation with variation in gene expression , we estimated the heritability of individual glucosinolate metabolic traits and total glucosinolate content traits , as well as transcript levels for glucosinolate biosynthetic genes ( Figure 2 and Tables S1 and S2 ) . Heritability estimates for glucosinolate content traits were significantly lower than the heritability of RMA ( robust multichip analysis ) estimated transcript levels of their respective biosynthetic genes . This was true for both indolic and aliphatic glucosinolates ( Figure 2 ) . Differences in heritability of metabolite and expression traits could arise from differences in population size used for these estimates ( 403 lines for metabolites and 211 lines for transcript levels ) . However , recalculating heritabilities for the glucosinolate metabolites using the same 211 RILs measured in the eQTL analysis did not significantly change the values ( unpublished data ) . Analysis of aliphatic glucosinolate variation among the RILs identified 11 QTLs that control either aliphatic glucosinolate content or the partitioning of aliphatic glucosinolates into particular structures ( Figure 3 ) . The QTLs affecting the largest number of aliphatic glucosinolate traits and causing the largest phenotypic differences were the previously identified GSL . AOP and GSL . Elong loci ( Tables S4 and S5 and Figure 1 ) . Polymorphism at these QTLs altered aliphatic glucosinolate content as well as derived ratio and summation traits . The GSL . ALIPH . II . 42 and GSL . ALIPH . V . 66 QTLs altered individual glucosinolate content and summations but did not have as dramatic an affect on the glucosinolate ratios ( Figures 3 and 4 ) . In contrast , GSL . ALIPH . I . 0 , GSL . ALIPH . III . 10 , and GSL . ALIPH . IV . 48 QTLs were more specific to the derived summation and ratio traits than to the raw glucosinolate metabolite accumulation ( Figure 3 and Tables S4 and S5 ) . For example , the GSL . ALIPH . IV . 48 QTL influences only 3C aliphatic glucosinolate accumulation , and the GSL . ALIPH . I . 0 QTL controls ratio traits involving 4-MT and 8C glucosinolate ( Figure 4 and Tables S4 and S5 ) . These QTLs highlight the ability of derived traits to provide unique insights into metabolic variation . Mapping eQTLs controlling transcript levels for the known aliphatic glucosinolate biosynthetic genes identified eight eQTL clusters that all coincided with aliphatic glucosinolate metabolite QTLs ( GSL . ALIPH . I . 0 , GSL . ALIPH . I . 73 , GSL . ALIPH . II . 15 , GSL . ALIPH . II . 42 , GSL . ALIPH . III . 10 , GSL . AOP , GSL . Elong , and GSL . ALIPH . V . 66; Figures 3 and 4 and Table S6 ) . Two of these eQTLs , GSL . AOP and GSL . Elong , are known to be cis-eQTLs controlling the expression of four biosynthetic genes [31] . However , these loci appear to modify in trans the transcript levels for a broad set of aliphatic glucosinolate biosynthetic genes . Conducting a network expression analysis using an updated gene list for the aliphatic biosynthetic pathway showed that six of these eQTL clusters ( including GSL . AOP and GSL . Elong ) were also detected using the network mean z-score for the aliphatic glucosinolate biosynthetic gene network . The network mean z-score is a derived trait obtained by averaging across the aliphatic glucosinolate transcripts within a RIL ( Table S7 ) [18] . As the aliphatic glucosinolate transcripts are believed to participate in a highly coregulated network , this derived trait is potentially informative regarding network control [18 , 38 , 39] . These six network eQTLs appear to control transcript levels of the pathway in trans , but whether trans functionality occurs via metabolite feedback or transcriptional mechanisms remains to be elucidated ( Figure 4 ) . The eQTLs detected in the network analysis predominantly controlled biosynthetic enzymes acting early in the aliphatic glucosinolate pathway—in the elongation and core biosynthetic stages—and not the secondary modification stages ( Figures 4 and S1 ) . Of the other three metabolite QTLs that did not show a network eQTL , the GSL . ALIPH . I . 20 QTL overlapped with a cis-eQTL for UGT74B1 , and the other two did not coincide with loci affecting the expression of any known biosynthetic genes . Variation at the side-chain–modifying GSL . AOP and GSL . Elong loci controlled the accumulation of most metabolites and transcripts ( Figures 3 and 4 ) . Neither locus had been previously identified as impacting transcript accumulation for the whole glucosinolate biosynthetic network . Both GSL . AOP and GSL . Elong are controlled by cis-eQTLs leading to differential enzyme expression . As in Cvi , the Sha GSL . AOP locus expresses the AOP2 enzyme , leading to alkenyl glucosinolate production , higher glucosinolate content , and elevated transcript levels for most aliphatic glucosinolate genes ( Figure 4 and 5 ) . We used Col-0 ( which is null for AOP2 and AOP3 ) transformed with a functional AOP2 gene from Brassica oleracea to test if the presence of functional AOP2 transcript can affect both metabolite and transcript levels [40] . AOP2 from B . oleracea conducts the same reaction as AOP2 from Sha and is also associated with elevated aliphatic glucosinolate content in Brassica , allowing us to test the conservation of this locus across the two species [40 , 41] . Introduction of a transcript encoding functional AOP2 results in the production of alkenyl glucosinolates and a statistically significant doubling of total aliphatic glucosinolate content , as is the case with the presence of a functional AOP2 transcript contributed by the Sha allele at the GSL . AOP QTL ( Figure 5 ) . The introduction of the AOP2 transcript also leads to induction of 17 of 22 aliphatic glucosinolate biosynthetic genes and three of seven regulatory genes ( Figure 5 ) . This supports the hypothesis that the Sha allele at the GSL . AOP QTL controls metabolite and transcript levels for aliphatic glucosinolates due to increased expression of the AOP2 gene . This suggests the presence of a previously unrecognized regulatory effect of AOP2 , whereby it controls transcript accumulation for most biosynthetic genes potentially through transcription factors . While we could not detect any micro-RNA signatures within the AOP2 transcript or gene , it remains to be shown whether the metabolite and transcript effect is due to the enzymatic activity of AOP2 or some other regulatory signature within the AOP2 transcript . The association of both GSL . AOP and GSL . Elong with eQTLs for the majority of aliphatic glucosinolate biosynthetic genes and metabolites ( Figure 4 ) suggests a regulatory interplay between the metabolites directly synthesized by these enzymes and transcript levels for the aliphatic glucosinolate biosynthetic genes . We tested the identified QTLs for pairwise epistatic interactions controlling metabolite accumulation , partitioning , or transcript levels . This analysis identified at least one pairwise epistatic QTL interaction for all metabolites , with variation in at least half of the metabolites controlled by interactions between GSL . AOP , GSL . Elong , and GSL . V . 66 ( Figure 4 ) . The most common pairwise interaction was detected between GSL . AOP and GSL . Elong , controlling 7 of 9 aliphatic glucosinolate metabolites ( Figure 4 ) . In contrast to the metabolites , most expression traits did not identify epistatic eQTL interactions . The few transcripts traits that identified epistatic eQTL interactions encode biosynthetic enzymes functioning in the early steps of the elongation cycle: MAM1 , BCAT4 , and an Aconitase . This suggests that genes in the elongation cycle may be regulated differently from the rest of the aliphatic glucosinolate pathway genes ( Figures 1 , 4 , and S1 ) . To investigate the nature of the epistatic interaction between GSL . AOP and GSL . Elong , we calculated mean phenotypic values for the RILs containing each of the allelic combinations at these two loci ( Figure 6 ) . GSL . AOP × GSL . Elong interaction had a negative epistatic effect on the total content of both aliphatic and indolic glucosinolates , with lines possessing the nonparental allelic combination of GSL . AOPOHP from Bay-0 and GSL . ElongC4 from Sha exhibiting significantly lower glucosinolate content than either parent ( Figure 6 ) . Lines possessing Sha alleles at both loci had the highest glucosinolate accumulation . The GSL . AOP and GSL . Elong loci also control the network expression mean z-score for the aliphatic biosynthetic gene network , but did not exhibit a pairwise epistasis for this trait ( Figure 4 ) . In contrast , the genes in the methionine elongation cycle did identify an epistatic interaction between GSL . AOP and GSL . Elong ( Figure 6; BCAT4 is shown as an example ) . This lack of an epistatic effect on aliphatic glucosinolate network expression is not likely a statistical artifact , as the pattern of the group means shows a striking difference between aliphatic glucosinolate accumulation and network expression of the aliphatic biosynthetic genes ( Figure 6 ) . Substitution of the Bay-0 GSL . ElongC3 allele for the Sha allele in a background containing the Sha GSL . AOPAlk allele leads to increased accumulation of aliphatic glucosinolate biosynthetic transcripts but lower aliphatic glucosinolate content ( Figure 6 ) . This suggests that these two loci regulate both transcript and metabolite accumulation via distinct mechanisms . We analyzed the indolic glucosinolate pathway as a second test of our ability to link eQTLs altering transcript levels for biosynthetic genes with metabolite accumulation QTLs . A total of 13 QTLs were identified that control the accumulation of indolic glucosinolates , their partitioning into particular structures , or both ( Figure 7 and Table S8 ) . These loci affect one or more of the indolic glucosinolate traits in this population , with the directionality of allelic effects mixed , so that Bay-0 alleles at some loci increase trait values while the Bay-0 alleles at other loci decrease the metabolite trait values ( Table S8 ) . Interestingly , the GSL . INDOLIC . IV . 8 and GSL . INDOLIC . V . 20 QTLs map to the same genomic locations as the previously known GSL . AOP and GSL . ELONG loci , which control aliphatic glucosinolate variation . Transgenic analysis confirmed that simulating the Sha GSL . AOP allele by introducing the AOP2 gene into a null background increases total indolic glucosinolate content by about 30% ( p = 0 . 035 ) . This shows that variation at GSL . AOP affects indolic glucosinolate metabolism and that there is cross-talk between the pathways for indolic and aliphatic glucosinolate production . Testing the identified indolic metabolite QTLs for pairwise epistatic interactions identified numerous epistatic interactions affecting indolic glucosinolate accumulation . Notably , the GSL . INDOLIC . II . 15 QTL was detected as epistatically interacting with five other QTLs ( GSL . INDOLIC . III . 7 , GSL . INDOLIC . III3 . 60 , GSL . INDOLIC . IV4 . 36 , GSL . INDOLIC . V . 45 , and GSL . INDOLIC . V . 59; Table S8 ) . These epistatic interactions affect the partitioning of the indolic glucosinolates into two distinct methoxylated derivatives without significantly altering the total indolic glucosinolate content . GSL . INDOLIC . II . 15 might encode or regulate enzymes responsible for this methoxylation ( Table S8 ) . A regulation hypothesis is supported by the fact that the GSL . INDOLIC . II . 15 region contains a massive trans-acting eQTL that influences transcript levels for more than 5 , 000 genes [17] . Analysis of eQTLs controlling transcript levels of individual indolic glucosinolate biosynthetic genes identified four eQTL clusters ( GSL . INDOLIC . III . 60 , GSL . INDOLIC . IV . 8 , GSL . INDOLIC . V . 45 , and GSL . INDOLIC . V . 59; Figure 7 and Tables S6 and S7 ) . Because the indolic glucosinolate biosynthetic genes are also believed to be coregulated , we estimated pathway expression mean z-value to map network QTLs . This identified the same four loci , plus three additional QTLs affecting expression of the indolic gene network ( GSL . INDOLIC . II . 15 , GSL . INDOLIC . III . 7 , and GSL . INDOLIC . V . 5 ) . All three of these network-specific eQTLs colocalized with metabolite QTLs , supporting the ability of the z-scale network approach to derive biological information . All seven eQTLs colocalize with loci that control either the accumulation or partitioning of the indolic glucosinolates produced by these genes . This suggests that the eQTLs controlling transcript levels for the indolic glucosinolate biosynthetic genes also affect variation in their metabolite products . Again , it remains to be tested whether the QTLs primarily affect transcript levels , causing downstream metabolite effects , or if the polymorphisms first affect the metabolites , influencing transcript levels through some form of feedback . There are also indolic metabolite QTLs , such as GSL . INDOLIC . IV . 36 , with no detectable impact on gene expression traits .
The three most consistently detected QTLs for aliphatic glucosinolate content and transcript variation were GSL . AOP , GSL . Elong , and GSL . V . 66 . Two of these network QTLs , AOP and Elong , contain cis-eQTLs controlling variation in transcript levels for enzymes located at the beginning and end of the aliphatic glucosinolate pathway . Neither AOP nor Elong has previously been associated with regulation of other glucosinolate biosynthetic or regulatory genes . This suggests that the accumulation of aliphatic glucosinolates is regulated by multiple mechanisms functioning such that enzyme variation can feed back to alter transcript accumulation ( Figure 1 ) . The GSL . AOP QTL is likely controlling glucosinolate content and expression through natural variation in the AOP2 gene ( Figure 5 ) . Altering the expression of the AOP2 transcript , the enzyme at the end of the aliphatic glucosinolate pathway might pull carbon into aliphatic glucosinolate production ( Figures 4–6 ) . In addition , introducing the transcript for AOP2 increases transcript level for the aliphatic biosynthetic pathway , suggesting that regulation also occurs by control of gene expression and potentially metabolic fluxes through the beginning of the aliphatic glucosinolate pathway ( Figures 5 and 6 ) . It is possible that a metabolite produced by or used by the AOP2 enzyme may have the capacity to regulate transcript accumulation for the rest of the biosynthetic network through feedback . Direct regulation of transcript accumulation by metabolites has been noted for a variety of riboswitches [42–45] . For the GSL . Elong network QTLs , the comparison of eQTLs to metabolite QTLs suggests that variation in transcript levels for at least two biosynthetic genes ( MAM1 and MAM3 ) at GSL . Elong directly causes metabolic variation , and that this biosynthetic variation may alter transcript regulation for nearly the entire biosynthetic pathway . Specifically , lines that contain 4C glucosinolates have lower transcript levels for almost all biosynthetic genes . One possibility is that a metabolic intermediate produced in the GSL . Elong4C background negatively regulates gene expression . Alternatively , both the GSL . AOP and GSL . Elong loci could possess tightly linked second loci that cause the observed transcriptional polymorphisms and also interact epistatically . Testing the interaction between biosynthetic loci and transcript levels will require careful genetic manipulation with full transcriptome analysis . Comparing the remaining network eQTLs to candidate glucosinolate transcriptional regulators shows that the GSL . ALIPH . V . 66 QTL overlaps the physical position of the MYB28 transcription factor , which has a large-effect cis-eQTL [17 , 39] . Likewise , the GSL . INDOLIC . V . 45 and GSL . INDOLIC . V . 59 QTLs overlap the physical position of the indolic transcription factors , ATR1 and ATR2 , and both genes have cis-eQTLs [17 , 29 , 46] . This suggests that these three QTLs may be explained by variation in the expression of these known regulatory genes . In addition , the GSL . ALIPH . I . 0 and GSL . ALIPH . III . 10 QTLs overlap with the physical position for the aliphatic glucosinolate transcriptional regulators AtDOF1 . 1 and IQD1 , respectively [28 , 47] . Neither gene exhibits a cis-eQTL , suggesting that if these genes are responsible for GSL . ALIPH . I . 0 and GSL . ALIPH . III . 10 QTLs , it is potentially due to an activity polymorphism . Alternatively , small changes in transcript levels for these transcription factors might lead to large changes in network regulation ( Figure S1 ) . The GSL . ALIPH . II . 15 QTL , a major network eQTL for transcript levels for both aliphatic and indolic glucosinolate pathways , does not overlap any known biosynthetic or regulatory genes . This locus does colocalize with a massive eQTL cluster controlling the expression of several thousand genes [17] . This suggests that this region may be highly pleiotropic and its effects on glucosinolate content may be indirect . These results suggest that eQTLs can control metabolite production through a variety of direct and indirect regulatory mechanisms . While most eQTLs co-located with QTL clusters controlling several metabolites , there were also instances in which a statistically significant association between expression and metabolic phenotypes was limited to one or few metabolic traits . For example , SGT74B1 catalyzes glycosylation of the characteristic glucosinolate backbone structure , and its expression is controlled by a single eQTL at GSL . ALIPH . I20 [48] . This large-effect cis-eQTL maps to a 100-kb interval that includes the physical position of SGT74B1 . While we might predict that variation in the expression of SGT74B1 would influence production of multiple aliphatic glucosinolates , this locus was only identified as controlling one metabolic trait , the accumulation of but-3-enyl glucosinolate . This gene is not specific to the synthesis of but-3-enyl glucosinolate , as previous work has demonstrated that SGT74B1 has a broad biochemical capacity to glucosylate glucosinolates [48] . Instead , the lack of detected effects in this study for this eQTL on the accumulation of other glucosinolates is likely because genotypes accumulating but-3-enyl glucosinolate also exhibit the highest level of total aliphatic glucosinolates within this population ( Table 1 ) . This suggests that the SGT74B1 expression polymorphism is only limiting when flux across the biosynthetic pathway is maximized . Consequently , unexpected factors directly related to biochemical pathway connectivity and flux can interfere with our ability to directly associate eQTLs for biosynthetic genes with specific metabolite QTLs . As such , eQTLs are not predictive in all contexts . This comparative analysis of eQTLs to metabolic QTLs has provided novel insights , including the identification of the enzymes AOP2 and MAM1 as well as the transcription factor MYB28 as potential regulators of transcript accumulation for the complete aliphatic glucosinolate biosynthetic pathway . In addition , differential expression of the ATR1 and ATR2 transcription factors have been implicated as the underlying cause of QTLs controlling the indolic glucosinolate pathway . Data presented in this study validate the potential of the enzyme AOP2 to control aliphatic glucosinolate gene expression . The other regulatory roles hypothesized above remain to be verified . Identification and validation of the molecular causes of the 17 different QTLs identified in this study will require a complex mixture of experiments ranging from transgenic complementation to validate the gene , promoter swaps to validate the difference in gene expression , and recombination mapping to more precisely identify the causal polymorphism [49] . For glucosinolates and their biosynthetic genes , we observed significant differences between the estimated heritability of metabolite accumulation and transcript levels , respectively ( Figure 2 ) . Expression traits had consistently higher broad sense heritability than the accumulation of individual metabolites or metabolite summation traits . Because genetically identical individuals were used for both experiments , there is no difference in the amount of genotype variation available to transcript and metabolite traits . Thus , environmental inputs may affect metabolic traits more strongly than transcripts . The mechanistic link between sequence polymorphism and variation in transcript levels may have fewer intervening processes than a causal sequence polymorphism and metabolic variation . The presence of additional regulatory processes , both metabolic and post-transcriptional , may allow greater environmental heterogeneity effects on metabolite accumulation . It is also possible that the results obtained for the glucosinolate pathway are not indicative of typical transcript or metabolite heritability . There is evidence that glucosinolate production is under diverse selection pressures , favoring high levels of plasticity in glucosinolate accumulation mediated by environmental stimuli such as nutrient and water availability and wounding [50–52] . This would require that metabolite traits be influenced by subtle environmental heterogeneity , leading to reduced estimates of their heritability . It is therefore important to expand this analysis to other metabolic pathways to determine the extent to which these conclusions are generalizable . We also observed that metabolic traits identified significantly more epistasic interactions than the corresponding transcripts . Regulatory processes that occur between transcript accumulation and metabolite accumulation , e . g . , metabolic feedback , post-transcriptional regulation , and enzyme activity regulation , may increase regulatory interactions between loci , leading to metabolite traits showing higher levels of epistasis . The constant adjustment of metabolite flux through complex networks may also enhance the potential for epistasis . This finding may be specific to the glucosinolate system . A broader metabolomics analysis in comparison to eQTLs for all known enzymatic loci will be required to test whether these differences in heritability and epistasis are a general feature of transcriptomic and metabolomic networks . If this is the case , a detailed modeling approach may contribute to understanding differences in genetic architecture between metabolic and transcript networks . In this report , we show that it is possible to relate natural variation at the transcript and metabolite levels for two glucosinolate biosynthetic networks . Furthermore , this analysis shows that the comparison of eQTLs to metabolite QTLs within an a priori–defined framework can identify complex regulatory mechanisms whereby variation in enzymes or metabolites may feed back to alter transcript accumulation . For aliphatic glucosinolates , the beginning and end of the biosynthetic pathway interact to control the whole pathway . These feedback associations can lead to the rapid generation of new hypotheses about the regulation of biosynthetic networks , but also show that the de novo reconstruction of biosynthetic relationships from metabolite data will require great care . In all cases , variation in gene expression also affected the resultant metabolites , although extrapolating the effects of gene expression on metabolism requires caution due to interplay of biochemical mechanisms . Combining different genomics datasets will greatly expand our ability to understand both the regulation of metabolism and the relationship between transcription and metabolism .
The Bay × Sha population of 403 A . thaliana RILs [53] was used to map QTLs controlling individual and total glucosinolate content for both the aliphatic and indolic glucosinolates . The QTL on Chromosome V for total content of aliphatic glucosinolates in the August 2005 experiment is also presented elsewhere for clarity ( Sonderby , Hansen , Halkier , and Kliebenstein , unpublished data ) . Further , 211 of these lines have been analyzed for variation in gene expression and used to map QTLs controlling transcript levels [17 , 18] , allowing comparison of QTLs controlling metabolite accumulation with transcript levels for the underlying biosynthetic genes . Seeds were imbibed and cold-stratified at 4 °C for 3 d to break dormancy . Two complete plantings were grown simultaneously in neighboring growth chambers to provide independent biological replicates for each experiment . The full experiment was replicated three times between March of 2004 and August of 2005 , providing six glucosinolate measurements for most lines , totaling nearly 2 , 600 measurements . The replicates were labeled May 2004 , May 2005 , and August 2005 . For the May 2005 and August 2005 experiments , plants were grown in flats with 36 cells per flat , and maintained under short-day conditions in controlled environment growth chambers . For the May 2004 experiment , plants were grown in flats with 96 cells per flat , and maintained under short-day conditions in controlled environment growth chambers . Using the same growth chambers , similar growth conditions , and assaying glucosinolate content at the same developmental stage analyzed in the eQTL mapping experiment maximizes our ability to compare the metabolic QTL results with eQTLs for the biosynthetic genes [17 , 18] . At 35 days after germination , a fully expanded mature leaf was harvested , digitally photographed , and analyzed for glucosinolate content as described below at the same age as the plants used for eQTL analysis [17 , 18] . The glucosinolate content of excised leaves was measured using a previously described high-throughput analytical system [30 , 54] . Briefly , one leaf was removed from each plant , digitally photographed , and placed in a 96-well microtiter plate with 500 μL of 90% methanol and one 3 . 8-mm stainless steel ball-bearing . Tissue was homogenized for 5 min in a paint shaker , centrifuged , and the supernatant was then transferred to a 96-well filter plate with 50 μL of DEAE sephadex . The sephadex-bound glucosinolates were eluted by incubation with sulfatase . Individual desulfo-glucosinolates within each sample were separated and detected by high-performance liquid chromatography ( HPLC ) –diode-array detection , and identified and quantified by comparison to purified standards . Tissue area for each leaf was digitally measured using Image J with scale objects included in each digital image [55] . The glucosinolate traits are reported per square centimeter of leaf area . There was no significant variation detected for leaf density within these lines ( unpublished data ) . In addition to the content of individual glucosinolates , we developed a series of summation and ratio traits based on prior knowledge of the glucosinolate pathways ( Table S1 ) [56] . For instance , the content of 3-MT , 3-MSO , 3-OHP , and allyl glucosinolates were summed ( sum3C ) to provide an estimate of the content of 3C aliphatic glucosinolates within these lines ( Figure 1C and Table S1 ) . This enables the detection of QTLs that specifically alter 3C glucosinolate accumulation irrespective of specific side-chain modification . The ratio traits were created to measure the efficiency of partitioning a class of glucosinolates into particular structures . For example , the ratio allyl glucosinolate to total 3C aliphatic glucosinolates ( all_r3 ) allows discrimination of the efficiency of production of 3C alkenyl glucosinolates independent of the accumulation of 3C glucosinolates ( Figure 1C and Table S1 ) . These ratios and summation traits allow us to isolate the effects of variation at individual steps of glucosinolate biosynthesis from variation affecting the rest of the biosynthetic pathway [56] . For each glucosinolate trait , we determined the average value per RIL per experiment for QTL mapping . Because there was no significant difference in the variance of the traits between the experiments , we also calculated the average value per RIL across all three experiments for all traits ( Table S2 ) . Heritability of each glucosinolate trait was estimated using the general linear model procedure within SAS ( http://www . sas . com ) where broad sense heritability was defined as σ2g/σ2p ( Table S1 ) , where σ2g is the estimated genetic variance for the metabolite among different genotypes in this sample of RILs , and σ2p is the estimated phenotypic variance for the metabolite [2] . We used previously published biochemical and coexpression data to identify all known or predicted genes encoding glucosinolate biosynthetic enzymes [27 , 31 , 38 , 57 , 58] . For these purposes , the indolic and aliphatic glucosinolate pathways are considered to be independent biosynthetic processes . This appears to reflect the biological reality , as the two pathways use different genes and amino acid precursors [25 , 27] . Gene families were separated into genes involved in aliphatic or indolic glucosinolate pathways based on biochemical or phenotypic data where possible [59–61] . Where this was not possible , coexpression with the known indolic or aliphatic glucosinolate genes was combined with published biochemistry to separate gene family members into their respective pathways [38] . This generated a list of genes involved in aliphatic and indolic glucosinolate biosynthesis ( Table S3 ) . Heritability , eQTL position , eQTL effect , and transcript levels for individual transcripts were obtained using the RMA estimated expression values from the previously published analysis of gene expression in the Bay × Sha population [17] . To conduct network expression analysis , transcript levels for each biosynthetic gene within each RIL were standardized as z-scores . This is done by taking the transcript level for a gene within a RIL , subtracting the mean transcript level for that gene among the RILs , then dividing the resulting value by the standard deviation for that transcript among the RILs [18] . The mean z-score for the aliphatic and indolic glucosinolate biosynthetic genes was calculated within each RIL for each replicate [18] . This pathway mean z-score per RIL per replicate was then used to estimate heritability of the aliphatic and indolic glucosinolate pathway gene expression as described for the metabolites . The mean z-score per RIL across replicates was calculated and used to map QTLs controlling transcript levels of the aliphatic and indolic glucosinolate biosynthetic pathways ( Table S4 ) . Because these global transcription studies were conducted in the same mapping population grown under the similar conditions and in the same growth chambers , it was possible to compare the expression and metabolite data . The Bay × Sha RIL population has been previously genotyped [53] , and additional markers were obtained from the expression QTL analysis [62] as well as markers specific for the GSL . AOP and GSL . Elong loci [6 , 30] . To maximize our ability to detect all possible QTLs , we used the averages from each experiment , May 2004 , May 2005 , and August 2005 , as well as the average across all experiments to conduct four QTL mapping tests . This was done for all individual glucosinolate traits , summation traits , and ratio traits ( Table S1 ) . The four averages for each trait were independently used for QTL mapping within Windows QTL Cartographer v2 . 5 ( http://statgen . ncsu . edu/qtlcart/WQTLCart . htm ) [63–65] . Composite interval mapping was implemented using Zmap ( Model 6 within Windows QTL Cartographer v2 . 5 ) with a 10-cM window and an interval mapping increment of 2 cM . Forward regression was used to identify five cofactors per quantitative trait . The declaration of statistically significant QTLs is based on permutation-derived empirical thresholds using 1 , 000 permutations for each trait mapped [66 , 67] . The Eqtl module of QTL Cartographer was used to automatically identify the location of each significant QTL for each trait from each experiment and the whole experiment average ( Tables S5–S7 ) [65] . Composite interval mapping with permutations to assign significance using the underlying trait distribution is fairly robust at handling normal or near-normal traits as found for most of our traits [68] . In addition , all data from the three different experiments were used in the multi-trait composite interval algorithm within QTL Cartographer v2 . 5 . QTL clusters were identified by using the QTL summation approach , where the position of each QTL for each trait for each experiment is indicated by a 1 , and the number of traits controlled by a QTL at a given position is totaled [18] . This summation was conducted using four groups: Group I , all aliphatic glucosinolate metabolite traits; Group II , all eQTLs for aliphatic glucosinolate biosynthetic genes; Group III , all indolic glucosinolate metabolite traits; and Group IV , all eQTLs for indolic glucosinolate biosynthetic genes . These QTL clusters identified a set of defined genetic positions that were then named respective to their position and whether they affected aliphatic or indolic glucosinolate content ( Tables S5–S7 ) . The QTLs at the previously characterized and cloned AOP and Elong loci were named as such [30–32 , 69] . To further validate each QTL identified and query for potential epistasis , we conducted an analysis of variance ( ANOVA ) using all experiments . For each trait , the markers most closely associated with each significant main-effect QTL for that trait were used as main effect cofactors . In addition , experiment ( May 2004 , May 2005 , and August 2005 ) was used as a main effect cofactor . An automated SAS script was then developed to directly test all main effects as well as all possible pairwise interactions , including experiment × marker ( QTL ) and marker ( QTL ) × marker ( QTL ) interactions . Significance values were corrected for multiple testing within a model using false discovery rate ( < 0 . 05 ) in the automated script . The script returned all significance values as well as QTL main-effect estimates in terms of allelic substitution values ( Tables S5–S7 ) . No significant three-way interactions were identified . Two independent homozygous lines containing functionally expressed AOP2 transcript from B . oleracea expressed from a 35S promoter were obtained in the Col-0 background that is null for AOP2 and AOP3 [31 , 40] . These two lines were grown in a randomized block design with wild-type Col-0 and tested by HPLC for glucosinolate content and by ATH1 Affymetrix microarrays ( http://www . affymetrix . com ) for altered transcript levels [17 , 18] . For total aliphatic glucosinolate content , six individual plants per line were measured and ANOVA used to test for altered glucosinolate accumulation . The complete experiment was conducted twice . As there was no difference between the independent transgenic lines , this factor was removed from the model . From each experiment , two independent RNA samples from Col-0 and two independent RNA samples from Col-0::AOP2 were obtained and hybridized with ATH1 Affymetrix microarrays as described [17 , 18] . Transcript levels for the genes involved in aliphatic glucosinolate biosynthesis ( Table S3 ) were obtained and used in a targeted ANOVA testing the effect of the AOP2 transgene . p-values were tested for significance against a false discovery rate of 0 . 05 using this subset of genes [70] .
The microarray dataset used in this study has been deposited at European Bioinformatics Institute ArrayExpress ( http://www . ebi . ac . uk/arrayexpress ) under numbers E-TABM-126 and E-TABM-224 . All gene identifiers are listed in Table S3 . | Natural genetic variation and the resulting phenotypic variation between individuals within a species have been of longstanding interest in wide-ranging fields . However , the molecular underpinnings of this phenotypic variation are relatively uncharted . Recently , genomics methodologies have been applied to understanding natural genetic variation in global gene expression . This , however , did not resolve the connection between variation in gene expression and the resulting physiological phenotype . We used two metabolic pathways within the model plant Arabidopsis to show that it is possible to connect genomic analysis of genetic variation to the resulting phenotype . This analysis showed that the connections between gene expression and metabolite variation were complex . Finally , the major regulators of gene expression variation for these pathways are two biosynthetic enzymes rather than traditional transcription factors . This analysis provides insights into how to connect transcriptomic and metabolomic datasets using natural genetic variation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"biochemistry",
"plant",
"biology",
"arabidopsis",
"computational",
"biology",
"plants",
"genetics",
"and",
"genomics"
] | 2007 | Linking Metabolic QTLs with Network and cis-eQTLs Controlling Biosynthetic Pathways |
Trypanosoma cruzi , the protozoan causative of Chagas disease , is classified into six main Discrete Typing Units ( DTUs ) : TcI-TcVI . This parasite has around 105 copies of the minicircle hypervariable region ( mHVR ) in their kinetoplastic DNA ( kDNA ) . The genetic diversity of the mHVR is virtually unknown . However , cross-hybridization assays using mHVRs showed hybridization only between isolates belonging to the same genetic group . Nowadays there is no methodologic approach with a good sensibility , specificity and reproducibility for direct typing on biological samples . Due to its high copy number and apparently high diversity , mHVR becomes a good target for typing . Around 22 million reads , obtained by amplicon sequencing of the mHVR , were analyzed for nine strains belonging to six T . cruzi DTUs . The number and diversity of mHVR clusters was variable among DTUs and even within a DTU . However , strains of the same DTU shared more mHVR clusters than strains of different DTUs and clustered together . In addition , hybrid DTUs ( TcV and TcVI ) shared similar percentages ( 1 . 9–3 . 4% ) of mHVR clusters with their parentals ( TcII and TcIII ) . Conversely , just 0 . 2% of clusters were shared between TcII and TcIII suggesting biparental inheritance of the kDNA in hybrids . Sequencing at low depth ( 20 , 000–40 , 000 reads ) also revealed 95% of the mHVR clusters for each of the analyzed strains . Finally , the method revealed good correlation in cluster identity and abundance between different replications of the experiment ( r = 0 . 999 ) . Our work sheds light on the sequence diversity of mHVRs at intra and inter-DTU level . The mHVR amplicon sequencing workflow described here is a reproducible technique , that allows multiplexed analysis of hundreds of strains and results promissory for direct typing on biological samples in a future . In addition , such approach may help to gain knowledge on the mechanisms of the minicircle evolution and phylogenetic relationships among strains .
The protozoan parasite Trypanosoma cruzi ( Kinetoplastea: Trypanosomatidae ) is the causative agent of Chagas disease . This parasite infects millions of people throughout its distribution in Latin America . Chagas disease can display a broad pathological spectrum , including potentially fatal cardiological and gastrointestinal dysfunctions [1] . T . cruzi is a monophyletic taxon showing a remarkable genetic heterogeneity , with at least six phylogenetic lineages formally recognised as Discrete Typing Units ( DTUs ) , TcI–TcVI [2 , 3]; and a seventh lineage , named TcBat [4–6] . The genetic diversity of T . cruzi was firstly revealed by Multilocus Enzyme Electrophoresis [7 , 8] and posteriorly by very diverse techniques including Multilocus Sequence Typing ( MLST ) [9–12] , microsatellite typing ( MLMT ) [13–18] , target-specific PCR [19–21] , PCR-RFLP [22 , 23] , PCR-DNA blotting with hybridization assays [24–26] , and recently by amplicon deep sequencing [27 , 28] . The different approaches have their own advantages and disadvantages and bring out the genetic diversity of T . cruzi at different levels . Approaches that allow direct typing from biological samples ( blood , tissues , etc . ) , avoiding parasite culture , are more suitable for clinical and epidemiological studies . However , nowadays there is no methodologic approach with a good sensibility , specificity and reproducibility for direct typing on biological samples . Because there is usually a low number of parasites in infected tissues or blood samples , genetic markers with high number of copies are required to achieve good sensitivity of detection [29] . In this regard , T . cruzi , as all the kinetoplastids , has a unique and large mitochondrion which contains a complex network of DNA , the kinetoplastic DNA ( kDNA ) . The kDNA represents approximately 20–25% of the total cellular DNA in T . cruzi and consists of two kind of circular DNA molecules: maxicircles and minicircles . Maxicircles contain mitochondrial genes characteristic of other eukaryotes [30] . Minicircles are present in tens of thousands of copies [31] . Each of them is organized into four highly conserved regions located 90° apart each other , and an equal number of hypervariable regions ( mHVRs ) interspersed between the conserved regions [32] . The highly conserved regions of minicircles have been widely used as targets for molecular detection of T . cruzi DNA . The used primers show a good sensitivity and specificity [29] and amplify a region of about 330 bp that totally include the mHVRs present between conserved regions . This amplified region has been used in hybridization assays ( mHVR probes ) and DTU-specific hybridization was observed only between isolates belonging to the same genetic group [25 , 26 , 33–35] . This specificity observed in hybridization assays suggests the presence of DTU specific sequences and even genotype-specific sequences ( i . e . sequences showing specificity at intra-DTU level ) . However , technical limitations that existed until a few years ago for sequencing these highly variable kDNA regions , prevented the identification of the sequences in which the specificity relies . Some attempts were made by cloning and sequencing some mHVRs [36 , 37] but the limited number of studied sequences were not enough to obtain a complete picture of the genetic diversity of these sequences . Thus , the observed hybridization patterns between mHVRs continue being a black box system and the sequence diversity of T . cruzi mHVRs virtually unknown . Beyond the potential utility for strain typing , studying mHVR diversity is also interesting because these sequences are involved in functions that are only known in kinetoplastids and in no other eukaryotic organism . mHVRs code for short RNAs called guide RNAs ( gRNAs ) . gRNAs are involved on edition of several mitochondrially-coded mRNAs . This edition varies from addition of some Us to building almost the full open reading frame of the mRNA [38 , 39] . In this sense , gRNAs can be inferred from sequences of the mitochondrial mRNAs and diversity on edition among strains can be addressed [40] . In addition , studying mHVR diversity can shed light on how such sequences evolve and how they are inherited . Here , we propose an amplicon deep sequencing approach that allows an accurate knowledge of the sequence diversity of the hypervariable region of kDNA minicircles of T . cruzi and opens the possibility of functional and evolutionary studies . This approach can be also used as a typing method for hundreds of samples at time .
DNA from nine cloned T . cruzi strains belonging to the six main DTUs was examined in this study ( Table 1 ) . All the strains were typified by using an optimized Multilocus Sequence Typing scheme based on four gene fragments ( HMCOAR , GPI , TcMPX and RHO1 ) according to Diosque et al . [7] , in order to confirm DTU for each strain . In order to amplify the minicircles hypervariable region , kDNA specific primers 121 ( 5’-ACACTCTTTCCCTACACGACGCTCTTCCGATCTAAATAATGTACGGG ( T/G ) GAGATGCATGA-3’ ) and 122 ( 5’-GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTGGTTCGATTGGGGTTGGTGTAATATA-3’ ) were modified by adding an oligo adapter to be used in an Illumina platform . The mHVR libraries were generated by a one-step PCR performed in 5 μl reaction volumes containing 5 ng of DNA , 250 nM of each primer , 2 μM of barcode primers , 5 U of Fast Start High Fidelity Enzyme Blend ( Roche ) , 0 . 50 μl of 10X buffer ( supplied with the Fast Start High Fidelity Enzyme Blend ) , 25 nM of MgCl2 ( Roche ) , 0 . 25 μl of DMSO ( Roche ) , 10 mM of PCR grade nucleotide mix ( Roche ) . The PCR reaction was carried out on a Veriti Thermal Cycler ( Life Technologies ) and ran as follow: an initial denaturation step ( 10 min at 95°C ) , 10 cycles ( 95°C for 15 seconds , 60°C 30 seconds , 72°C 1 min ) , 2 cycles ( 95°C for 15 seconds , 80°C 30 seconds , 60°C 30 seconds , 72°C 1 min ) , 8 cycles ( 95°C for 15 seconds , 60°C 30 seconds , 72°C 1 min ) , 2 cycles ( 95°C for 15 seconds , 80°C 30 seconds , 60°C 30 seconds , 72°C 1 min ) , 8 cycles ( 95°C for 15 seconds , 60°C 30 seconds , 72°C 1 min ) and 5 cycles ( 95°C for 15 seconds , 80°C 30 seconds , 60°C 30 seconds , 72°C 1 min ) . Amplicons were then purified using the magnetic beads Agencourt AMPure XP-PCR Purification ( Beckman Genomics , USA ) . The concentration of the purified amplicons was controlled using Qubit Fluorometer 2 . 0 ( Invitrogen , USA ) . All libraries were validated using the Fragment Analyzer system ( Advanced Analytical Technologies , USA ) . The average size of the mHVR amplicons was ~480bp . All samples were then pooled and prepared according to the manufacturer's recommendations ( Illumina Protocols: Sequencing Library Preparation ) and sequenced on an Illumina MiSeq using a 500 cycle v2 kit ( Illumina , San Diego , USA ) to produce amplicons of approximately ~480 bp in length ( 250 bp paired-end reads ) .
The raw data set has been deposited in the NCBI SRA database ( BioProject ID: PRJNA514922 ) .
A total of 22 , 092 , 382 paired reads were obtained by amplicon sequencing of the mHVR from nine strains belonging to six DTUs . A total of 14 , 766 , 753 sequences were retained ( an average of ≈1 . 4 million of sequences per strain ) after trimming low quality ends , merging paired reads ( forward and reverse ) , elimination of chimeric reads and filtering by base quality ( S1 Table ) . Surviving sequences were clustered according to different identity thresholds ( 85% , 90% and 95% ) ( Table 2 , S2 and S3 Tables ) . The number of mHVR clusters for each strain was very similar using different thresholds ( with differences less than 10% in all comparisons between 85% and 95% thresholds ) . However , clustering at 85% threshold returned few more mHVR clusters than clustering at 90% and 95% identity ( See Table 2 , S2 and S3 Tables ) . In addition , most clusters were highly divergent among them ( S1 Fig ) . At any threshold , the number of mHVR clusters was variable among strains and DTUs ( Table 2 ) , ranging from 71 ( Mncl2 –TcV ) to 373 ( X109/2 –TcIII ) clusters . Additionally , strong intra-DTU variations in the number of clusters were observed in strains of TcI and TcII ( Table 2 ) . Finally , rarefactions of each dataset discarded that these differences among strains are the effect of different sequencing depths ( Table 2 , S2 and S3 Tables ) . Strains belonging to TcIV , TcV and TcVI showed some dominant clusters containing a high proportion of reads ( i . e . the cluster size ) ( Fig 1 ) . The sum of the six most abundant clusters in TcIV , TcV and TcVI represent in all cases more than 50% of the clustered sequences ( 80 . 9% and 69 . 1% in the TcV strains LL014R1 and MNcl2 , respectively; 58 . 7% in TcIV strain CANIIIcl1; and 52 . 5% in the TcVI strain LL015P68R0cl4 ) . Even more , in LL014R1 and MNcl2 ( TcV strains ) the most abundant cluster represented the 29 . 7% and 17 . 8% of the total mHVR , respectively . Instead , none of the clusters present in TcI , TcII and TcIII strains represented more than 5 . 2% . This higher diversity in TcI , TcII and TcIII is also revealed by a higher Simpson diversity index than other DTUs ( Table 2 ) . Moreover , intra-DTU differences in mHVR cluster diversity were observed in TcII . Particularly , Tu18cl93 had relatively less cluster diversity than Esmeraldo ( Table 2 and Fig 1 ) . As expected , shared mHVR clusters were mostly observed in strains belonging to the same DTU . However , the percentage of shared clusters was highly variable depending on DTU . TcV strains ( LL014R1 and MNcl2 ) showed the higher proportion of shared clusters ( 97 . 3%; 72/74 ) . However , we observed strong differences in the cluster sizes ( Fig 2C ) although a positive correlation was detected ( correlation coefficient , r = 0 . 75 ) and some shared clusters were highly abundant in both strains ( Fig 2C ) . TcI strains ( PalDa20cl3 and TEV55cl1 ) shared 17 . 5% ( 83/475 ) , and TcII strains ( Tu18cl93 and Esmeraldo ) shared 7 . 1% ( 33/466 ) . Conversely , when we look for shared mHVR clusters between strains belonging to different DTUs , we detected none or few shared clusters ( Fig 2D–2I and S2 Fig ) . The Bray-Curtis dissimilarity between strains was calculated using mHVR clusters conformed at the different identity thresholds ( 85% , 90% and 95% ) . Such dissimilarities were used to analyze principal coordinates ( PCoA ) and to build UPGMA trees ( Fig 3 and S3 Fig ) . Strains from the same DTU clustered together ( Fig 3 ) despite the high dissimilarities between strains belonging to the same DTU ( Fig 3C ) . These high dissimilarities between strains belonging to the same DTU determine that the three first axis in the PCoA explain just 49 . 1% of the variance . TcV strains clustered distant from other DTUs . TcIII and TcIV strains clustered near to each other . Interestingly , TcVI strain was placed between TcII and TcIII in the PCoA . Moreover , TcVI was clustered with TcII in the UPGMA tree ( Fig 3C ) . Such results are not in agreement with the hypothesis of uniparental inheritance of the minicircles in the hybrid TcVI , which comes from hybridization between TcII and TcIII . Consequently , we analyzed shared clusters between TcII , TcIII and the hybrids DTUs ( TcV and TcVI ) in order to analyze the hypotheses of uniparental or biparental inheritance of minicircles . We used a 90% identity threshold in order to be more confident about the identity by descendance of the clusters . We observed that TcV and TcVI share 11/530 and 19/559 mHVR clusters with TcII , respectively . Likewise , TcV and TcVI shared 12/429 and 9/469 mHVR clusters with TcIII , respectively ( Fig 4 ) . Instead , TcII and TcIII share only 2 mHVR clusters between them out of a total number of clusters of 842 combining TcII and TcIII . These results suggest that minicircle inheritance is biparental . In addition , TcV and TcVI shared more mHVR clusters with their parental DTUs than between them ( Fig 4 ) which is concordant with the hypothesis of independent origins of TcV and TcVI . In order to test if parallel amplicon sequencing would be useful for simultaneous typing of hundreds of strains , we first evaluated rarefaction curves . In general , the minimum number of reads required to detect at least 95% of the observed clusters was 20 , 000 filtered reads . The only exception was MNcl2 , which required 40 , 000 filtered reads . Increasing the number of reads per sample beyond 20 , 000 slightly increased the number of detected mHVR clusters ( Fig 5A ) . In addition , we evaluated the minimum number of reads required to observe the right DTU assignment described in Fig 2 . As few as 10 , 000 reads were enough to accurate clustering of the strains ( Fig 5B and 5C ) at 100% of the rarefactions . Amplicon sequencing of the mHVR could be useful to identify intra-DTU clusters , particularly in TcV or TcVI where strains may have the same composition of mHVR clusters but with high differences in abundance of each one . In order to develop future methods to assign strains to intra-DTU clusters is pre-requisite that amplicon sequencing can be reproducible to determine mHVR cluster abundance . Consequently , we assessed reproducibility by sequencing and comparing two independent PCR reactions of the mHVR in LL015P68R0cl4 strain ( TcVI ) . High correlation in cluster abundances in different PCRs of the same sample was observed ( r = 0 . 999 for the three different identity thresholds ) ( Fig 5D ) .
Here , we made a deep amplicon sequencing of the hypervariable region of kDNA minicircles in the six main lineages ( DTUs ) of T . cruzi . To the best of our knowledge , this is the first time that these kDNA regions were sequenced at millions of reads of depth . Our results shed light on different and very interesting aspects of these intriguing DNA sequences . We accurately show the level of sequence diversity of mHVR within strains , between strains belonging to the same DTU , and between strains belonging to different DTUs . Although it was already known that mHVR were highly diverse [36] , the magnitude of this diversity at the intra- and inter-DTU level has not been demonstrated with the high precision provided by an NGS approach , as we made here . We propose a method for typing/elucidating intra-specific diversity of T . cruzi based on the deep sequencing of the hypervariable region of kDNA minicircles . The idea is based on the outdated but highly sensitive method of mHVR probes [25 , 26 , 35 , 46–48] . Such probes are useful to detect T . cruzi diversity in biological samples . However , this methodology has the disadvantages of being technically cumbersome , relying on visual interpretation of bands and requiring representative strains of the diversity of T . cruzi in every assay ( used as probes ) . The deep amplicon sequencing approach proposed here is reproducible and based on objective sequence data which can be stored in online databases . Also , the method is multiplexable for hundreds of samples at time and it would be directly applied to biological samples as the mHVR probes . The method may be potentially useful to address epidemiological questions about associations between intra-specific diversity and variability in clinical manifestations of the chronic disease or the different rates of congenital transmission in different endemic areas . Such questions have been unsuccessfully addressed using molecular markers with low resolution and/or low sensitivity on biological samples . We determined that around 20 , 000 filtered reads are enough to reveal most mHVR diversity in a strain and theoretically allowing for running hundreds of samples in a single run of a MiSeq with costs similar or lower than MLST . However , a wider set of strains belonging to the six main lineages must be studied . In addition , new bioinformatic methods of analysis will be required for a direct application of the method to biological samples . In order to develop such typing method , we preliminarily analyzed and compared the diversity of mHVR sequences in reference strains of six DTUs and at millions of reads of sequencing depth . We observed that strains of the same DTU share more mHVR clusters than strains of different DTUs . However , unprecedented high differences in mHVR cluster composition was observed for strains of the same DTU with less than 20% of shared mHVR clusters in TcI and TcII . Instead , almost all mHVR clusters were shared between different TcV strains . In addition , the patterns of DTU specificity observed by using mHVR probes may be explained in TcV and TcVI by the presence of some shared and abundant clusters . Instead , considering the higher diversity and low abundance of clusters in TcI , TcII and TcIII , the global pattern of sequences is probably the responsible of specificity in the hybridization assays involving these DTUs . Interestingly , our data revealed that diversity of mHVR sequences was variable even within a DTU . This was particularly evident in TcII , where the number of mHVR clusters in Esmeraldo strain doubled that of Tu18cl93 . Such differences may be caused by long times in culture as it has been observed for other trypanosomatids [40 , 49] . However , both strains were isolated in the eighties and although it is possible that they had different times in culture , such times would be not very different ( i . e . not in the order of decades ) . According to this , we suppose that the observed difference in mHVR diversity between the two TcII strains is not due to long time in culture . In support of the hypothesis of no influence of the time in culture , we observed no differences in mHVR diversity between the two TcV strains examined , despite they have very different times of isolation and maintenance mode in the laboratory . One of them was isolated in the 1980s and subjected to long periods of maintenance in culture ( Mncl2 ) ; and the other TcV strain ( LL014R1 ) was isolated in 2008 and maintained in triatomine-mouse passages . Our results also shed some light on the evolutionary mechanism determining the large genetic distances in mHVR sequences among strains and DTUs . The focus should be first placed on TcV strains which are identical according to MLST and which shared most mHVR clusters . Despite this , they strongly varied in relative frequencies of mHVR clusters . Such variations cannot be attributed to simple stochasticity of the PCR amplification because we observed good correlation between different PCR reactions from the same sample ( Fig 5D ) . Consequently , it is probable that minicircle diversity is mainly driven by genetic drift . We propose that when two strains diverge , the frequencies of mHVR cluster varies stochastically , some clusters increasing their relative frequency and other decreasing it . The next step can be seen in strains of TcI which are more genetically distant than the TcV ones . Such TcI strains show clusters with high abundance in one strain and with very low ( or null ) abundance in the other one ( look at most clusters located on the axes in Fig 2 ) . Therefore , some clusters will be lost if such lost is not deleterious ( i . e . replaced by a different mHVR class that codes a gRNA editing the same mRNA fragment ) . Thus , strains would diverge by variations in frequency of the mHVR classes faster than by changes in their sequences . These variations in the frequency of mHVR classes probably are not under selective pressure . mHVR frequency variations are apparently allowed because the effective edition of the mRNA is not dependent on the abundance of a minicircle [50 , 51] . Variations in the frequency of mHVR classes have been also inferred for T . brucei and Leishmania [52] and by a theoretical study assuming random or partially random segregation of minicircles [53] . With the purpose of developing in the future DTU specific PCRs , we analyzed if different DTUs share common mHVR clusters . Telleria et al . [36] did not detected shared sequences between DTUs probably because the low sequencing depth . With a different approach , Velazquez et al . [37] detected that most abundant mHVR classes in CL-Brener ( TcVI ) were also present in other DTUs but in a considerably lower frequency . We detected shared mHVRs between different DTUs but we did not detect any sequence shared by the six DTUs . Interestingly , we observed shared clusters between TcVI and TcIII ( 2 . 1% ) . This is expected considering that TcIII is a parental DTU of the hybrid TcVI and maxicircle sequences of TcIII are closely related to the TcVI ones [54–58] . However , the TcVI strain also shared 2 . 5% of mHVR clusters with Esmeraldo strain ( belonging to TcII , the other parental DTU of TcVI ) . Something similar is observed for the also hybrid DTU TcV ( Fig 3 ) . Instead , only 2 mHVR clusters were shared between TcII and TcIII strains ( 0 . 2% ) . This clearly suggests that although maxicircles have apparently uniparental inheritance in TcV and TcVI , minicircles were probably inherited from both parentals and some of them persisted for 60 , 000 years since hybridization [59] . Biparental inheritance of minicircles and maxicircles has been proposed for Trypanosoma brucei hybrids [60–62] . In this parasite , it has been observed that maxicircle and minicircle inheritance is biparental in hybrids . However , maxicircles ( 20–50 copies ) are homogenized by genetic drift resulting in the loss of whole maxicircles of one parental in few generations . However , minicircles have much more copies and they resist the fixation effect of genetic drift for more time . Consequently , maxicircle inheritance is biparental and just seems to be uniparental due to genetic drift . As consequence of the biparental inheritance of minicircles , it has been proposed that such inheritance may help to preserve mHVR diversity in T . brucei preventing the effect of the drift , and even that T . brucei requires genetic exchange to prevent the deleterious effect of loss of essential minicircle classes [53] . Nevertheless , genetic exchange has remained elusive to be detected in T . cruzi . Experimental hybrids obtained by Gaunt and coworkers showed that maxicircles are from one parental but minicircles were not analyzed [63] and kDNA inheritance was still not addressed in more recent experimental hybrids [64] . In addition , the frequency of genetic exchange may be variable among different DTUs . TcV and TcVI ( which display a clearly clonal genetic structure at population level ) [9 , 10 , 12 , 57] have very low mHVR diversity . Instead , TcI , TcII and TcIII , for which genetic exchange has been proposed in the nature [11 , 13 , 15 , 65] , have higher mHVR diversity . Moreover , our data may help elucidate the origin of hybrid DTUs . It has been proposed that TcV and TcVI are the result of a single hybridization event between TcII and TcIII and both DTUs diverged posteriorly [66 , 67] . However , the alternative hypothesis ( two independent hybridization ) gain weight in the last years . Particularly , Multilocus Microsatellite Typing ( MLMT ) and Multilocus Sequence Typing ( MLST ) analyses favored the two independent hybridizations hypothesis [57 , 59] . Considering biparental inheritance , and assuming a single hybridization event , the two hybrid DTUs ( TcV and TcVI ) should share more mHVR classes between them than with the parentals . However , our analyses show the contrary with very few classes shared between TcV and TcVI ( Fig 4 ) . This result supports independent hybridizations for the origin of TcV and TcVI . Alternatively , because both DTUs would have lost many mHVR clusters , the high divergence among them may have been caused by simple stochasticity , although is less likely . Interestingly , if minicircle are biparentally inherited it is expected that they will behave like the nuclear genes . So , it is expected that nuclear phylogenies will be similar to the mHVR phylogeny and both discordant to maxicircle phylogeny in cases of hybridization or introgression . However , some hypotheses about events that occurred very distant in time ( e . g . mitochondrial introgression in the origin of TcIII [57–58] ) might not be addressed by mHVR-based phylogenies because the almost null number of shared mHVR clusters between some DTUs . Concluding , massive amplicon sequencing of the mHVR is reproducible and suitable for typing hundreds of T . cruzi strains at time because few thousands of reads are required per sample . However , some drawbacks still need solution . The main problem in biological samples are mixed infections of different genotypes or DTUs which are very frequent [48] . However , such problem can be overpassed by developing new bioinformatic methods comparing mHVR composition of a sample against a reference mHVR database which should collect information about the diversity in the DTUs of T . cruzi . In addition , the develop of an online database where mHVR representative sequences are stored is needed . We are currently working on such items . In addition , some rare events of mitochondrial introgression observed in natural populations of T . cruzi lead to discordant typing between nuclear and maxicircle markers [16 , 68 , 69] . However , it is unknown the effect of mitochondrial introgression on minicircles . In this sense , a Multilocus deep Sequence Typing ( MLdST ) may be good alternative and a second step . The deep sequencing of amplicons of the mHVR plus satDNA ( a 195 bp sequence with 105 sequences per genome ) [70] may help elucidate such rare events and may increase sensitivity for typing on biological samples . | Chagas disease is an important public health problem in Latin America showing a wide diversity of clinical manifestations and epidemiological patterns . It is caused by the parasite Trypanosoma cruzi . This parasite is genetically diverse and classified into six main lineages . However , the relationship between intra-specific genetic diversity and clinical or epidemiological features is not clear , mainly because low sensitivity for direct typing on biological samples . For this reason , genetic markers with high copy number are required to achieve sensitivity . Here , we deep sequenced and analyzed a DNA region present in the large mitochondria of the parasite ( named as mHVR , 105 copies per parasite ) from strains belonging to the six main lineages in order to analyze mHVR diversity and to evaluate its usefulness for typing . Despite the high sequence diversity , strains of the same lineage shared more sequences than strains of different lineages . Curiously , hybrid lineages shared mHVR sequences with both parents suggesting that mHVR ( and DNA minicircles from the mitochondria ) are inherited from both parentals . The mHVR amplicon sequencing workflow proposed here is reproducible and , potentially , it would be useful for typing hundreds of biological samples at time . It also provides a valuable approach to perform evolutionary and functional studies . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Accession",
"numbers",
"Results",
"Discussion"
] | [
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"phylogenetics",
"data",
"management",
"protozoans",
"mitochondria",
"molecular",
"biology",
"techniques",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"genetic",
"epidemiology",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"kinetoplasts",
"artificial",
"gene",
"amplification",
"and",
"extension",
"bioinformatics",
"epidemiology",
"biological",
"databases",
"evolutionary",
"systematics",
"molecular",
"biology",
"trypanosoma",
"cruzi",
"biochemistry",
"trypanosoma",
"eukaryota",
"sequence",
"databases",
"cell",
"biology",
"polymerase",
"chain",
"reaction",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"evolutionary",
"biology",
"organisms"
] | 2019 | Elucidating diversity in the class composition of the minicircle hypervariable region of Trypanosoma cruzi: New perspectives on typing and kDNA inheritance |
Cell fusion in genetically identical Neurospora crassa germlings and in hyphae is a highly regulated process involving the activation of a conserved MAP kinase cascade that includes NRC-1 , MEK-2 and MAK-2 . During chemotrophic growth in germlings , the MAP kinase cascade members localize to conidial anastomosis tube ( CAT ) tips every ∼8 minutes , perfectly out of phase with another protein that is recruited to the tip: SOFT , a recently identified scaffold for the MAK-1 MAP kinase pathway in Sordaria macrospora . How the MAK-2 oscillation process is initiated , maintained and what proteins regulate the MAP kinase cascade is currently unclear . A global phosphoproteomics approach using an allele of mak-2 ( mak-2Q100G ) that can be specifically inhibited by the ATP analog 1NM-PP1 was utilized to identify MAK-2 kinase targets in germlings that were potentially involved in this process . One such putative target was HAM-5 , a protein of unknown biochemical function . Previously , Δham-5 mutants were shown to be deficient for hyphal fusion . Here we show that HAM-5-GFP co-localized with NRC-1 , MEK-2 and MAK-2 and oscillated with identical dynamics from the cytoplasm to CAT tips during chemotropic interactions . In the Δmak-2 strain , HAM-5-GFP localized to punctate complexes that did not oscillate , but still localized to the germling tip , suggesting that MAK-2 activity influences HAM-5 function/localization . However , MAK-2-GFP showed cytoplasmic and nuclear localization in a Δham-5 strain and did not localize to puncta . Via co-immunoprecipitation experiments , HAM-5 was shown to physically interact with NRC-1 , MEK-2 and MAK-2 , suggesting that it functions as a scaffold/transport hub for the MAP kinase cascade members for oscillation and chemotropic interactions during germling and hyphal fusion in N . crassa . The identification of HAM-5 as a scaffold-like protein will help to link the activation of MAK-2 cascade to upstream factors and proteins involved in this intriguing process of fungal communication .
Fusion between genetically identical cells occurs in many different organisms and plays pivotal roles in different developmental processes , such as myoblast fusion during muscle formation , macrophage fusion involved in tissue remodeling and fusion of trophoblasts during placental development [1] , [2] . Cell fusion is also important for the formation of the interconnected mycelial network that is the hallmark of filamentous fungal growth [3] , [4] , [5] . In addition to hyphal fusion , fusion can also occur between genetically identical germinating asexual spores ( conidia ) of filamentous fungi [6] , [7] , [8] . Both hyphal and germling fusion are integral to the formation of an interconnected hyphal network and impart fitness benefits , as well as mediating genetic mixing and the sharing of resources [3] , [4] , [9] , [10] , [11] , [12] , [13] . In the filamentous ascomycete fungus , Neurospora crassa , germinated conidia ( germlings ) in close proximity fuse via specialized conidial anastomosis tubes ( CATs ) that form at germ tube tips or between conidia [14] . In a fungal colony , hyphal fusion is observed in the central parts of the colony , in contrast to the peripheral parts where hyphae avoid each other . A large number of genes have been identified in N . crassa that are important for the process of sensing , chemotropic interactions and CAT fusion and have contributed to an understanding of this complex developmental system in filamentous ascomycete fungi [7] , [15] , [16] , [17] . An essential part of chemotropic interactions in N . crassa is the oscillatory recruitment of three kinases of a MAPK cascade ( NRC-1 , MEK-2 and MAK-2 ) and of a protein of unknown function , SOFT ( SO ) , to CAT tips [18] , [19] . In strains carrying loss-of-function mutations in these genes , oscillatory recruitment of NRC-1/MEK-2/MAK-2 or SO , chemotropic interactions and fusion do not occur [8] , [18] , [19] , [20] . It was proposed that the alternating oscillation of MAK-2 and SO to CAT tips may function to establish two distinct physiological states in interacting germlings to enable chemotropism to persist , avoid self-stimulation and assure a rapid and efficient cell fusion [19] , [21] . Recently , it has been shown that an ortholog of SOFT in the related filamentous ascomycete species , Sordaria macrospora , PRO40 , is a scaffold protein for the cell wall integrity MAP kinase pathway ( MAK-1 ) [22]; Δmak-1 mutants in both N . crassa and S . macrospora are fusion mutants [22] , [23] . Previously , it was shown that kinase activity of MAK-2 is required to maintain oscillatory recruitment of both MAK-2 and SO; addition of ATP-analog 1NM-PP1 to a mutant containing an inhibitable MAK-2 protein encoded by a mak-2Q100G allele , disrupted the oscillation of both MAK-2 and SO in communicating germlings and stalled chemotropic interactions and the fusion process [19] . The use of mak-2Q100G strain also contributed to the identification of downstream genes whose expression levels depend on functional MAK-2 [24] . However , MAK-2 kinase targets involved in the oscillation process have not been identified and the characterization of such potential targets could help unravel molecular mechanisms associated with this highly regulated and complex process . In recent years , highly sensitive liquid chromatography-mass spectrometry ( LC-MS ) based quantitative phosphoproteomic techniques have contributed to our understanding of kinase pathway function in eukaryotic cells [25] , [26] , [27] , [28] . To identify MAK-2 kinase targets in N . crassa , we took a global approach by identifying phosphopeptides in mak-2Q100G germlings , treated or not with 1NM-PP1 . From the phosphoproteomic screen , a number of candidate MAK-2 target proteins were identified , one of which encodes a protein previously identified as being essential for germling/hyphal fusion , HAM-5 [23] , [29] . We show that HAM-5 oscillates during chemotropic interactions with components of the MAK-2 pathway , physically interacts with NRC-1 , MEK-2 and MAK-2 and was required for localization of MAK-2 and MEK-2 to puncta . Our data supports the hypothesis that HAM-5 functions as a scaffold protein by binding to and co-localizing to CAT tips with all three kinases in the MAK-2 cascade during chemotropic growth in germlings as well as in hyphae undergoing fusion events . These studies shed new light on the mechanisms of oscillation during communication and chemotropic interactions between genetically identical cells and which may be important for function of this conserved MAPK pathway in other filamentous ascomycete fungi .
To better understand the role of MAK-2 during chemotropic interactions , we set out to identify putative kinase targets using a global quantitative phosphoproteomics approach using a strain carrying an inhibitable mak-2Q100G allele . Altering a specific amino acid in the ATP binding site ( glutamine for glycine ) renders MAK-2 sensitive to inhibition to the ATP analogue 1NM-PP1 , but does not affect MAK-2 kinase activity in the absence of inhibitor [19] . In the absence of the inhibitor 1NM-PP1 , strains and germlings containing the mak-2Q100G allele ( his3::mak-2Q100G; Δmak-2 ) showed wild-type growth and fused normally , while in the presence of inhibitor , the mak-2Q100G strain showed a mutant mak-2 phenotype; chemotropic interactions and fusion were not observed , consistent with inactivation of MAK-2 kinase activity [19] . For identifying MAK-2-dependent phosphorylation events , 4 . 5-hr old mak-2Q100G germlings were treated with DMSO ( two samples and two biological replicates ) or 1NM-PP1 ( dissolved in DMSO; two samples and two biological replicates ) for 10 min . Proteins were extracted , digested , and enriched phosphopeptides identified and quantified using isobaric peptide tags for relative and absolute quantification ( iTRAQ ) -based LC-MS/MS analyses . Although the same peptides across experimental conditions are labeled with different iTRAQ reagents and indistinguishable by mass , different masses will be generated in the tandem MS by releasing the reporter ions for the 4-plex iTRAQ method ( Figure S1 ) . We performed the full experiment twice , resulting in a total of eight samples , which were analyzed for phosphopeptide identity and abundance . From these experiments a total of 3200 unique phosphopeptides were identified . These 3200 unique phosphopeptides originated from 1164 proteins ( Dataset S1 ) . A small percentage of the identified peptides have multiple phosphorylation sites ( 12% , Figure 1B ) , as compared to peptides where only a single phosphorylation site was identified ( 88% , Figure 1B ) . Phosphorylation sites were predominantly identified on serine residues ( 75% ) , and to a lesser extent on threonine ( 22% ) and tyrosine ( 3% ) residues ( Figure 1C ) . FunCat analysis [30] showed the set of identified phosphorylated proteins in germlings originated from a wide spectrum of functional categories ( Figure 1A ) , including , not unexpectedly , proteins involved in metabolism , energy , cell cycle and DNA processing , protein synthesis and transcription . However , proteins within the functional categories of cellular communication/signal transduction , cell defense , and interaction with the environment were also identified , suggesting germlings are poised to respond to variations in their environment ( Dataset S1 ) . Of the 3200 phosphopeptides , 96 unique phosphopeptides from 67 proteins were>1 . 5 times less abundant ( p<0 . 05 ) in at least one replica experiment in the 1NM-PP1 treated mak-2Q100G germlings relative to the DMSO-treated control cells ( Table S1 ) . Functional category analyses [30] of this set of proteins showed enrichment for genes involved in metabolism , energy , cell cycle and DNA processing , transcription , protein fate , regulation of metabolism and protein function , cellular communication/signal transduction , interaction with the environment , cell fate and cell type differentiation ( p<0 . 01 ) ( Figure 1A; Dataset S1 ) . Three proteins in this group had previously been shown to be required for hyphal fusion , including HAM-9 , HAM-11 and MAK-1 [23] , [24] , [31] , [32] . In addition to the MAP kinase involved in the osmotic response signaling ( OS2 ) [32] , CUT-1 , which is implicated in the osmotic stress response [33] , [34] as well as the transcription factor that is a target of the OS-2 pathway ( ASL-1/ATF-1 , NCU01345 ) [35] were identified ( Table 1; Table S1 ) . Of these 3200 unique phosphopeptides , 33 phosphopeptides ( from 27 proteins ) were identified with an increased abundance of at least 1 . 5 fold ( p<0 . 05 ) in the 1NM-PP1 treated germlings in at least one replica experiment . Functional category analyses [30] of this set of 27 proteins showed over-representation for genes/proteins involved in phosphate metabolism ( perhaps as a response to 1NM-PP1 exposure ) , protein synthesis , protein fate and protein with binding function or cofactor requirements ( Dataset S1 ) . To identify potential direct MAK-2 targets , we inspected the 96 phosphopeptides that showed reduced abundance in the 1NM-PP1 treated mak-2Q100G germlings for MAPK consensus phosphorylation sites ( P-X-S*/T*-P ) [36] , [37] . Nine proteins were identified , of which five were annotated as hypothetical proteins . One of the proteins was ASL-1 , the transcription factor that is a target of the OS-2 pathway and which shows an ascospore-lethal phenotype [35]; Δmak-2 mutants also show an ascospore-lethal phenotype [8] , [38] . The remaining proteins included a predicted 6-phosphofructo-2-kinase ( NCU01728 ) , a predicted trehalose phosphatase ( NCU05041 ) and a protein previously reported to be required for hyphal fusion , HAM-5 [23] , [29] ( Table 1 ) . In addition to a predicted MAPK phosphorylation site , a predicted MAPK docking motif ( R/K-R/K- ( X ) 1-5-I/L-X-I/L ) [39] was predicted in NCU07868 ( hypothetical protein ) and HAM-5 ( Table 1 ) . Of the nine genes encoding potential MAK-2 phosphorylation targets , five showed a reduction in expression levels in either a Δpp-1 mutant ( transcription factor that is a target of the MAK-2 pathway ) or in a mak-2Q100G germlings treated with 1NM-PP1 [24] , including ham-5 ( Table 1 ) . To determine whether the putative direct or indirect MAK-2 phosphorylation targets were involved in germling fusion , strains carrying individual deletions of 8 out of the 9 genes whose proteins contained a MAPK consensus phosphorylation site and showed decreased abundance in the 1NM-PP1-treated mak-2Q100G germlings ( including strains carrying deletions in all five hypothetical proteins ) ( Table 1 ) , plus an additional 31 deletion mutants selected from a subset of proteins that showed decreased abundance in 1NM-PP1-treated mak-2Q100G germlings , but which lacked a predicted MAPK consensus phosphorylation site ( Table S1 ) , were evaluated for the ability to undergo chemotropic interactions and cell fusion . Of these , Δham-5 in the first set , and Δham-9 , Δham-11 , and Δmak-1 strains in the second set , were fusion defective . To further characterize putative MAK-2 targets that are required for germling fusion , we GFP-tagged proteins where localization during chemotropic interactions/cell fusion had not been determined . This effort included HAM-5 . The ham-5 mutant was identified in a forward screen for mutants that failed to form a heterokaryon [29] and encodes a large protein of 1686 amino acids ( aa ) with seven putative WD40 repeats at the N-terminus ( aa 14-313 , grey boxes; Figure 2A ) . At the C-terminus , an unstructured region of low complexity was identified ( shaded white boxes; Figure 2A ) that contains stretches of predominately proline and glutamine residues . Two coiled coil domains were also predicted in HAM-5 ( aa 1168-1190 and 1257-1286 , red boxes; Figure 2A ) . In total , 16 phosphorylation sites were identified , located mainly at the middle section of HAM-5 ( Figure 2A ) . Three phosphorylation sites were identified from our phosphoproteomics analysis , one of which was a putative MAPK site at amino acid residue 506 ( serine ) ( Figure 2A; Table 1 ) . Thirteen additional HAM-5 phosphopeptide sites were identified in a recent phosphoproteomics study of N . crassa hyphal cultures exposed to different carbon sources [40] . A putative MAPK docking site was also identified between residues 1128-1136 ( RRKPPALDL ) in the C-terminus of HAM-5 ( yellow bar; Figure 2A ) . In conidia , germlings and in hyphae , HAM-5-GFP fluorescence was observed in small , cytoplasmically localized puncta ( white arrows , Figure S2 ) . In mature hyphae , localization to septa was also observed ( red arrow , Figure S2A ) . During chemotropic interactions between germlings , HAM-5-GFP driven by the tef-1 promoter was observed as small puncta in interacting germlings and also at the tip ( Figure 2B ) . Importantly , HAM-5 oscillated to CAT tips in germling pairs during chemotropic interactions , with localization dynamics similar to MAK-2 or SO: HAM-5-GFP appeared at the CAT tip of one germling , but was absent from the CAT tip in the partner germling , while 4 min later , HAM-5-GFP was present at the CAT tip of the second partner germling , but was absent from the CAT tip of the first germling ( Movie S1 ) . During the fusion process , a HAM-5-GFP signal was also observed in puncta that localized either to the cell periphery or to points close to nuclear compartments , which were devoid of HAM-5-GFP ( Figure 2B , asterisk ) . When HAM-5-GFP localized to the CAT tips , the number of cytoplasmically localized puncta in the partner germling was reduced and the GFP signal was more dispersed in the cytoplasm ( Figure 2B ) . HAM-5-GFP was detected at the site of germling contact and remained there until the cytoplasm of the two germlings mixed ( Figure 2C ) . To assess whether the HAM-5-GFP-tagged versions were biologically functional , we crossed wild type ( WT ) strains carrying ham-5-gfp driven by either the native , tef-1 or ccg-1 promoter into the Δham-5 mutant . Progeny bearing both ham-5-gfp and Δham-5 showed a fusion phenotype and frequency similar to WT , indicating that ham-5-gfp driven either by the native , ccg-1 or tef-1 promoter was functional . Localization of HAM-5-GFP driven by either ccg-1 or tef-1 in the Δham-5 strain was similar to HAM-5-GFP in the WT strain , and showed localization to puncta and oscillation to CAT tips during chemotropic interactions . ham-5-gfp driven by its native promoter in the Δham-5 strain also showed localization to puncta ( Figure S3 ) . However , due to low expression levels of HAM-5-GFP in this strain , localization during chemotropic interactions could not be fully assessed . In the following sections , HAM-5-GFP is shown driven by the tef-1 promoter unless stated otherwise . A heterokaryon of a strain carrying ham-5-gfp driven by tef-1 and a strain carrying histone H1-dsRED ( a nuclear marker ) , showed non-overlapping fluorescence ( Figure S2 ) , indicating that HAM-5-GFP was excluded from the nucleus . This localization pattern is similar to SO-GFP [19] , but different than MAK-2-GFP , which localizes to the cytoplasm and to nuclei ( Figure S2 ) [19] . HAM-5 is predicted to be a highly phosphorylated protein , with 16 predicted phosphorylation sites ( Figure 2A ) . To assess whether HAM-5 is a phosphorylation target of MAK-2 , as indicated by the phosphoproteomics data , we introduced ham-5-GFP into a Δmak-2 strain and determined its phosphorylation status during germling fusion . HAM-5-GFP was immunoprecipitated from WT ( ham-5-gfp ) and Δmak-2 ( ham-5-gfp ) 5 hr-old germlings using anti-GFP antibodies and assayed for phosphorylation status using Western blot analysis with anti-phosphoserine/threonine antibodies that specifically recognize phosphoserine or phosphothreonine sites followed by a proline residue ( MAPK phosphorylation sites ) . The results showed that HAM-5-GFP from both WT and Δmak-2 cells was specifically phosphorylated ( Figure 2D ) . The HAM-5-GFP protein band from WT germlings showed a slight smear upwards and was slightly larger than that observed in Δmak-2 ( ham-5-gfp ) germlings ( Figure 2D ) . These data support the phosphoproteomics data , which suggested a MAK-2-dependent modification of HAM-5 during germling fusion ( Table 1 ) . However , it is clear that HAM-5-GFP was also phosphorylated in the Δmak-2 mutant , suggested possible additional regulatory inputs into HAM-5 via phosphorylation by other proteins during conidial germination/germling fusion . To assess whether the identified phosphosite in HAM-5 ( serine 506 ) was required for HAM-5 function during chemotropic interactions and germling fusion , strains carrying site-directed mutations whereby serine 506 was altered to an alanine ( phosphorylation impaired ) or a glutamate ( phosphorylation mimic ) residue were evaluated: chemotropic behavior and cell fusion in germlings carrying the ham-5S506A or ham-5S506E mutations driven by the ccg-1 promoter were indistinguishable from wild type germlings and fully complemented the growth phenotype of the Δham-5 strain ( Figure S4A ) . To assess whether mutations in the predicted MAPK docking site affected HAM-5 function , a strain was constructed where the first three amino acids of the MAPK docking site were changed to alanine ( RRKPPALDL to AAAPPALDL ) . However , germlings bearing this ham-5 allele ( ham-5RRK1128AAA ) also showed a similar communication and fusion phenotype to WT germlings resulting in similar growth phenotype to WT and complemented fully the growth phenotype of the Δham-5 strain ( Figure S4A ) . Thus , the predicted MAPK docking site in HAM-5 might not be functional or additional MAPK docking sites are present making this site redundant for function . We predicted that HAM-5-GFP would show altered localization when introduced into Δmak-2 germlings , due to the inability of these cells to undergo chemotropic interactions . However , HAM-5-GFP localized to puncta in Δmak-2; ham-5-gfp germlings , as observed in wild type cells ( Figure 2E ) . However , no oscillation of HAM-5-GFP puncta to cell tips was observed , consistent with the lack of fusion and chemotropic interactions in Δmak-2 germlings . HAM-5-GFP puncta were localized randomly within the cell , with some puncta close to the nuclear periphery and membrane , but cells also contained at least one HAM-5-GFP puncta localized to the cell tip ( Figure 2E ) . The stable , tip-anchored localization of HAM-5-GFP ( Figure 2E , arrows ) was not observed in WT germlings in the absence of chemotropic interactions ( in isolated germlings ) , and was unique to Δmak-2; ham-5-gfp cells . The observation that HAM-5-GFP showed oscillation during germling fusion , but still localized to puncta in Δmak-2 germlings , suggested that either HAM-5 interacted with SO or with other proteins in the MAK-2 signal transduction pathway . Previously , it was shown that the components of the MAK-2 pathway , the MAPKKK ( NRC-1 ) and the MAPKK ( MEK-2 ) , oscillate with MAK-2 during chemotropic interactions in germlings [18] . To determine if HAM-5 oscillated with the components of the MAK-2 complex or with SO , we used heterokaryons of strains carrying ham-5-gfp and either mCherry-tagged mak-2 , mek-2 , nrc-1 or so alleles and examined co-localization of these proteins during chemotropic interactions in germlings . As shown in Figure 3 , HAM-5-GFP + MAK-2-mCherry , or HAM-5-GFP + MEK-2-mCherry or HAM-5-GFP + NRC-1-mCherry were co-recruited to the CAT tips during chemotropic interactions . In homokaryotic strains carrying both HAM-5-GFP and MAK-2-mCherry , the dynamics of the two proteins were identical and showed simultaneous oscillatory recruitment to CAT tips ( Movie S2 ) . HAM-5-GFP + MAK-2-mCherry and HAM-5-GFP + MEK-2-mCherry also co-localized to cytoplasmic puncta during chemotropic interactions ( Figure 3 ) . As observed in [18] , the NRC-1-mCherry signal was weak , and recruitment of HAM-5-GFP + NRC-1-mCherry to cytoplasmic puncta could not be assessed . In contrast to MAK-2/MEK-2/NRC-1 , SO-mCherry + HAM-5-GFP always appeared at opposite CAT tips ( Figure 3D ) and with exactly opposite dynamics during chemotropic interactions ( Movie S3 ) . The localization of HAM-5 during chemotropic interactions suggested that HAM-5 physically interacts with MAK-2 , MEK-2 and/or NRC-1 . To test this hypothesis , we performed co-immunoprecipitation experiments using strains carrying HAM-5-GFP + MAK-2-mCherry , HAM-5-GFP + MEK-2-mCherry or HAM-5-GFP + NRC-1-mCherry . As controls , we used strains carrying MAK-2-mCherry , MEK-2-mCherry or NRC-1-mCherry-tagged proteins with GFP driven by the ccg-1 promoter; GFP in these strains showed only cytoplasmic localization and was never observed in puncta . A specific interaction between HAM-5-GFP and MEK-2-mCherry and HAM-5-GFP and NRC-1-mCherry was detected when HAM-5-GFP was immunoprecipitated using anti-GFP antibodies from 5 hr-old germlings and subsequently re-probed using anti-mCherry antibodies ( Figure 3E ) , consistent with the co-localization of these proteins observed by confocal microscopy ( Figure 3A–C ) . No interaction between the mCherry-tagged proteins and cytoplasmic GFP was observed ( Figure 3E; Figure S4C ) . An interaction between MAK-2-mCherry and HAM-5 could not be assessed , as MAK-2-mCherry showed non-specific binding . However , using phospho-specific anti-P42/P44 ( Erk1/Erk2 ) antibodies that recognize phosphorylated MAK-2 [8] , an interaction between HAM-5-GFP and MAK-2 was detected via co-immunoprecipitation ( Figure 4D ) . By contrast , no interaction was detected between HAM-5-GFP and SO-mCherry ( Figure 3E ) . HAM-5 is a large protein with predicted protein-protein interaction domains including seven WD40 repeats , which are predicted to form β-propeller structures that have been implicated in coordinating protein assemblages in other systems [41] . To test the hypothesis that the WD40 domain is involved in HAM-5-MAK/MEK-2/NRC-1 interactions , we constructed two mutant ham-5 alleles: one in which the WD40 motifs ( aa 67-348 ) were removed ( HAM-5Δ67-348 ) and one in which only the first 351 aa including the WD40 motifs of HAM-5 were retained ( HAM-51-351 ) . Both alleles were tagged with gfp , and function and localization were assessed in both WT and Δham-5 mutant strains . The ham-51-351 construct failed to complement the growth or fusion defects of the Δham-5 mutant ( Figure S4A ) . When observed microscopically , the localization of the HAM-51-351-GFP in Δham-5 germlings was cytoplasmic and nuclear; no puncta were observed ( Figure 4B; Figure S5A ) . This result is in contrast to the full length HAM-5-GFP , which localized to puncta and was excluded from the nucleus ( Figure 2; Figure S2 ) . However , in a WT background , a portion of the HAM-51-351-GFP localized to puncta and showed oscillation during chemotropic interactions between germlings ( Figure 4A ) . These observations suggest that in WT germlings , HAM-51-351-GFP may bind the native untagged HAM-5 , resulting in localization to puncta when the complex oscillates to CAT tips during chemotropic interactions . To determine whether HAM-51-351-GFP ( in a Δham-5 mutant ) physically interacted with MAK-2 or MEK-2 , we performed co-immunoprecipitation experiments using either anti-mCherry antibodies ( for MEK-2-mCherry ) or anti-p42/44 antibodies for MAK-2 [8] . HAM-51-351 specifically immunoprecipitated phosphorylated MAK-2 but not MEK-2-mCherry ( Figure 4C , D; Figure S4D ) . The ham-5Δ67-348 construct also failed to complement the growth or fusion defects of the Δham-5 mutant ( Figure S4A ) , consistent with an essential role for the HAM-5 WD40 domain . Cellular fluorescence of HAM-5Δ67-348-GFP in germlings or hyphae was not observed and less protein was produced than other GFP-tagged proteins ( Figure S4B ) , suggesting that the WD40 domain is required for HAM-5-GFP stability . However , co-immunoprecipitation experiments revealed a specific interaction between HAM-5Δ67-348 and MEK-2-mCherry , but not with MAK-2 ( Figure 4C; Figure S4F ) . These biochemical interaction studies indicated that the WD40 domain of HAM-5 is important for interactions with MAK-2 and the C-terminus is important for interactions with MEK-2 . The predicted MAPK docking site , which was not required for function by mutational analyses ( see above ) , is located in the C-terminus of HAM-5 , indicating that this site is not essential for MAK-2-HAM-5 interactions . Scaffold proteins such as Ste5 in Saccharomyces cerevisiae [42] , which assembles the pheromone response MAPK pathway , regulates spatial functionality of this pathway via nuclear/plasma membrane shuttling of Fus3 during mating . We hypothesized that HAM-5 was also required for the assembly of the MAK-2 cascade members in complexes and subsequent recruitment of these complexes to their correct cellular location during chemotropic interactions ( e . g . the CAT tip ) . To test this hypothesis , we introduced MAK-2-GFP or MEK-2-mCherry into a Δham-5 strain . In contrast to WT and Δmak-2 germlings where HAM-5-GFP was localized to puncta ( Figure 2 ) , MAK-2-GFP showed cytoplasmic and nuclear localization in the Δham-5 mutant; no puncta were observed ( Figure 5A , B ) . In Δham-5 hyphae , MEK-2-mCherry localized to the cytoplasm and to septa , but puncta were not observed as in WT hyphae ( Figure 5C , D; Movie S4 ) . We then tested whether HAM-5 was required to establish a stable interaction between MEK-2 and MAK-2 . When MEK-2-mCherry was immunoprecipitated from WT germlings , phosphorylated MAK-2 was also detected , while in Δham-5 germlings , co-immunoprecipitation of phosphorylated MAK-2 with MEK-2-mCherry was not detectable ( Figure 5E; Figure S4E ) . SO-GFP was also cytoplasmically localized in Δham-5 ( so-gfp ) germlings ( Figure S5 ) , a localization pattern identical to that observed in WT ( so-gfp ) germlings not undergoing chemotropic interactions . Phosphorylation of MAK-2 by the upstream kinase , MEK-2 , is required for fusion [18] , but is not fully dependent on functional HAM-5 . In a Δham-5 mutant , phosphorylated MAK-2 is still detectable in hyphal preparations [29] . We confirmed this result in germlings , where phosphorylated MAK-2 was also observed in the Δham-5 samples ( Figure 6A ) . MAK-2 phosphorylation is also observed in two other fusion mutant strains: Δham-7 and Δham-11 [24] , [43] ( Figure 6A ) . To investigate whether the localization of MAK-2 complexes to puncta was dependent on HAM-5 or the ability to undergo chemotropic interactions , we expressed HAM-5-GFP and MAK-2-mCherry in Δham-7 and Δham-11 cells . Although Δham-7 and Δham-11 germlings are unable to undergo chemotropic interactions and cell fusion [23] , [24] , [43] , co-localization of HAM-5-GFP and MAK-2-mCherry to puncta was still observed ( Figure 6B-D ) . Interestingly , as seen in the Δmak-2; ham-5-gfp strain , tip-anchored HAM-5-GFP and MAK-2-mCherry were present in Δham-7 and Δham-11 germlings ( Figure 6 , arrows ) . Most mutants affected in germling fusion are also deficient in hyphal fusion [7] , although localization of MAK-2 or SO during hyphal interactions has not been previously reported . Whether hyphal fusion is similarly coordinated as germling fusion is unknown , as different avoidance and fusion signals may be present at the periphery and older parts of a colony [44] . Another difference between germlings and hyphae is the presence of cytoplasmic flow in hyphae [10] that may influence the oscillation of proteins to sites of fusion . We therefore evaluated the localization of HAM-5-GFP during hyphal fusion in a mature colony . As shown in Figure 7 , oscillation of HAM-5-GFP to the tips of hyphae undergoing chemotropic interactions was observed with dynamics very similar to that during germling fusion ( where MAK-2 has a cycling time of ∼8 min at a single hyphal tip ) . Similar to germlings , MAK-2-mCherry also showed co-localization with HAM-5-GFP and oscillated with identical dynamics . MAK-2-mCherry was also observed in nuclei , while HAM-5-GFP was excluded ( Figure S6A and Movie S5 ) . In addition to localization to sites at fusion tips of hyphae , puncta containing HAM-5-GFP and MAK-2-mCherry within adjoining hyphal compartments also showed oscillation ( Figure 7 and Figure S6 ) . Interestingly , upon membrane merger ( t = 22 min ) , oscillation in both fusion hyphae was completely coordinated for an additional 30 minutes ( Figure 7C , D , Figure S6 and Movie S5 ) . We further assessed how far oscillation of HAM-5-GFP puncta extended in hyphal compartments that surrounded a fusion point . In filamentous fungi like N . crassa , hyphal compartments are delineated by septa , but septa contain a pore through which organelles , including nuclei , can move [45] . We observed coordinated oscillation of HAM-5 in over six hyphal compartments that were ∼100 µm from the point of fusion for a total distance of ∼200 µm ( Figure S7 ) . Hyphal compartments that showed different or no oscillation of HAM-5 distant from the point of fusion were delineated by septa ( Movie S6 ) . These data show that the oscillation of HAM-5-GFP in the hyphal network was coordinated over large distances surrounding a fusion point , but could be restricted by septa . Cytoplasmic flow , septal plugging and fusion events in nearby hyphae may also affect the oscillation of HAM-5 in compartments surrounding hyphal fusion points .
In this study , we show that HAM-5 functions as a scaffold protein for the MAK-2 MAP kinase complex and is required for oscillation of this complex during chemotropic interactions during germling and hyphal fusion in N . crassa . Our findings are complemented by the accompanying study of Dettmann et al . , [46] that show physical interaction between HAM-5 and MAK-2/MEK-2/NRC-1 via mass spectrometry and yeast two hybrid , assessing both indirect and direct physical interactions of HAM-5 with the MAK-2 kinase complex . In other filamentous ascomycete species , mutations in nrc-1 , mek-2 , mak-2 orthologs results in strains unable to undergo vegetative cell fusion [47] , [48] , [49] , [50] as well as defects in growth , reproduction , virulence and host colonization phenotypes , indicating expanded functions for this MAPK pathway in filamentous fungi as compared to yeast [51] , [52] , [53] , [54] , [55] , [56] . ham-5 is highly conserved in the genomes of filamentous ascomycete species [51] . We predict that these ham-5 homologs will function as a scaffold in these species for mak-2/mek-2/nrc-1 orthologs , and which may be important for mediating growth , reproduction and virulence functions of this important and conserved signal transduction pathway . The MAK-2 MAPK signal transduction pathway ( MAK-2 , MEK-2 and NRC-1 ) in filamentous fungi is orthologous to the pheromone response pathway in S . cerevisiae ( Fus3 , Ste7 and Ste11 ) . Previously , it was shown that a FUS3/KSS1 ortholog in the filamentous ascomycete species Magnaporthe grisea , ( PMK1 ) as well as its ortholog in Aspergillus nidulans ( mpkB ) complements the pheromone response/mating defect of a S . cerevisiae fus3Δ/kss1Δ mutant [49] , [57] . In S . cerevisiae , the Ste5 scaffold protein allosterically facilitates Ste7 phosphorylation of N . crassa MAK-2 ( called N . cra mpkB ) [58] and A . nidulans MpkB [57] , indicating conservation of regulation of these kinases by allosteric motifs within Ste5 . However , although components of these two MAPK pathways are highly homologous in fungi , an ortholog of STE5 is absent in the genomes of filamentous ascomycete species . Future experiments comparing the function of these non-homologous scaffold proteins ( STE5 and HAM-5 ) in regulating conserved signal transduction pathways will reveal how selection and evolution has shaped convergent evolution of these processes . In S . cerevisiae , Gβγ is involved in the recruitment of Ste5 to the plasma membrane upon pheromone exposure , thereby recruiting the Fus3 MAPK cascade to the membrane . Membrane binding of Ste5 likely concentrates the bound MAP kinases spatially , promoting amplification of the signal [59] , [60] . In N . crassa , how HAM-5 and MAK-2 are recruited to the membrane is still elusive , but is dissimilar from Ste5 since the Gβγ ortholog in N . crassa is not involved in germling fusion [61] , [62] . Other upstream factors shared between S . cerevisiae and N . crassa might regulate the activation of the MAPK pathway . One is STE50 , a component in yeast that helps to activate Ste11 . In the accompanying article , Dettmann et al . , [46] identified a role for N . crassa STE-50 as an activator of NRC-1; Δste-50 mutants were fusion deficient . Three other proteins acting upstream of NRC-1 and STE-50 were also identified: the MAP4 kinase STE-20 , the small GTPase RAS-2/SMCO-7 and the capping protein of the adenylate cyclase ( AC ) complex , CAP-1/NCU08008 . Strains carrying a deletion of any of these three genes still showed residual germling fusion , suggesting multiple and redundant inputs into the MAK-2 signal transduction pathway . In S . cerevisiae , Bem1 interacts with Ste20 , Ste5 and actin [63] . In N . crassa , strains carrying either a deletion of bem-1 , encoding a predicted scaffold for NADPH oxidase ( NOX ) , or its regulator ( NOXR ) , are germling/hyphal fusion deficient [23] . Activated RAC-1 also localizes to CAT tips during chemotropic interactions and may also function as an upstream activator [64]; Δrac-1 mutants are also are germling fusion defective [23] . During chemotropic interactions , HAM-5/MAK-2 complex assembles in puncta at the CAT tip and in the cytoplasm , followed by disassembly of HAM-5/MAK-2 complex from puncta , not just at the CAT tip , but from puncta observed throughout the germling and fusion hyphae , a cycle that repeats itself during chemotropic interactions every ∼8 min . In the absence of HAM-5 , localization of MAK-2 kinase complex to puncta was impaired , while in Δmak-2 mutants , HAM-5 puncta were still observed . These observations indicate that MAK-2 kinase activity is essential for disassembly of the HAM-5/MAK-2 complex ( Figure 8 ) during chemotropic interactions . The formation of HAM-5/MAK-2 complexes in puncta was not disrupted in other fusion mutants , such as Δham-7 and Δham-11; cortical localization of the HAM-5/MAK-2 complexes were observed at cell tips , although oscillation was not . These data suggest that the HAM-5/MAK-2/MEK-2/NRC-1 complexes are poised to signal for chemotropic interactions , but that the absence of HAM-7 or HAM-11 disrupts signaling that results in disassembly of the HAM-5/MAK-2 complexes , both within the cell and at the cell cortex . Few downstream targets of MAK-2 have been identified in filamentous fungi . The PP-1 protein , a transcription factor similar to Ste12 from yeast , is a likely downstream factor that is required for the activation of genes that play a role during the cell fusion and membrane merger [24] , [46] , [65] . Another target of MAK-2 is MOB-3 , a protein of the STRIPAK complex involved in cell fusion that assures correct nuclear localization of MAK-1 [66] . Among the phosphorylated proteins in addition to HAM-5 identified in this study are members of the osmosensing ( OS-2 , CUT-1 and ASL-1 ) and cell wall integrity pathways ( MAK-1 ) and other proteins of unknown functions but which are required for fusion ( HAM-9 and HAM-11 ) ( Table S1 ) . The accompanying study [46] also identified MAK-1 , OS-2 and CUT-1 in MAK-2 complexes via mass spectrometry . Both studies also identified other proteins that interacted with MAK-2/MEK-2/NRC-1 complex [46] or as potential phosphorylation targets of MAK-2 ( this study; Table S1 ) , including a glucokinase ( NCU00575 ) , SUC ( pyruvate decarboxylase ) , an aminotransferase ( NCU03500 ) , a trehalose-phosphatase ( NCU05041 ) , CAMK-4 calcium/calmodulin-dependent kinase-4 , and four hypothetical proteins ( NCU00627 , NCU00935 , NCU006247 and NCU08330 ) [46] . The identification of MAK-2 phosphorylation targets provides new clues to the interconnectivity of signaling pathways in N . crassa; these two combined datasets will be a rich resource for further studies on the MAPK pathway function in fungi . It is challenging to identify kinase targets that are often present in low abundance and have low phosphorylation stoichiometry , from a complex whole cell lysate with limited sample size . The multiplexed iTRAQ quantitation strategy , high specificity of phosphopeptide enrichment and high resolution nano-flow LC separation coupled to MS , as used in this study , have together contributed to the success of identifying and quantifying thousands of phosphopeptides from a small size sample ( ∼200 µg protein per sample condition ) . Our phosphoproteomics dataset from 5 hr old germlings provides information on stage-specific phosphorylation events on over ∼1100 proteins ( Dataset S1 ) . This dataset can be further compared to a recently published study on ∼3500 phosphorylated proteins identified under hyphal conditions and different carbon sources [40] . Both of these studies provide rich datasets for the filamentous fungal research community to interrogate the identity and function of phosphorylation sites on a large fraction of proteins in the N . crassa proteome . For example , three phosphorylation sites on HAM-5 from germlings were identified from this study , but an additional 13 HAM-5 phosphopeptides were identified in a sample from a 20 hr-old hyphal culture [40] . For chemotropic interactions , further studies on the additional phosphorylation sites and further dissection of the protein domains in the C-terminus of HAM-5 may explain how this scaffold protein itself is recruited to puncta , and may reveal additional binding partners for HAM-5 . Such studies will elucidate the molecular mechanism and function of oscillation of the HAM-5/MAK-2/MEK-2/NRC-1 complex during chemotropic interactions . Understanding the molecular basis of germling/hyphal fusion in filamentous fungi provides a window into fungal language and communication and provides a paradigm for self-signaling mechanisms in multicellular eukaryotic species .
Deletion strains used to screen for fusion mutants and strains constructed for this study are listed in Table S2 . Strains were grown on Vogel's minimal medium ( VMM ) [67] ( with supplements as required ) and were crossed on Westergaard's medium [68] . Transformations and other N . crassa molecular techniques were performed as described [69] or using protocols available at the Neurospora home page at the FGSC ( http://www . fgsc . net/Neurospora/NeurosporaProtocolGuide . htm ) . To construct the ham-5 alleles , PCR was performed with the restriction enzyme linkers included in the primer region . We amplified ham-5 alleles using primers 1-5: ham-5FXbaI tttttctagaATGTCGGTCCCCGGACACA; ham-5RpacI aaaattaattaaGATCATCTCACTATGATGCAAC; ham-5 WD40onlyR tttttaattaaGTTAGCAGGATGTTGAACGTTG; ham-5RWD40 tttatgcatatttaaaTCATGGTGGCAGCATACAATC; ham-5FWD40 tttatttaaatCCTGCTAACATGTTACCTCC; and cloned the fragments into pCR-Blunt vector ( Invitrogen ) . For constructing the point mutations at the predicted phosphorylation site we used fusion PCR strategy with primers CGTCATCGGGGGCGCGCGGCACCTCATGGCG and GGTGCCGCGCGCCCCCGATGACGCGAAAGTTGT ( S → A ) and CGTCATCGGGCTCGCGCGGCACCTCATGGCG and GGTGCCGCGCGAGCCCGATGACGCGAAAGTTGT ( S → E ) together with the primers TTTGATGCATCACAATGCTGACC and TTAAGGGCCGAATTCTTCGC . The mutated constructs were ligated into the NsiI and EcoRI sites of the HAM-5 gene . For constructing the mutation at the predicted docking site , we used primers ggtcgctgacaaactcgaat and tttgccggctgcCTCGGAACTGCGCGCGCGG that has the restriction site NaeI in the linker and TTTCGGCCAGCATCATGAGA and tttagcgctCCTCCAGCACTCGACCTTCGC that has the restriction site AfeI in the linker . The two respective products were digested with NsiI and NaeI and AfeI and Tth111I , respectively and ligated using a three-point ligation in the vector with HAM-5 cut open with NsiI and Tth111I . We sequenced and digested the constructs from the pCR-Blunt vector with the appropriate restriction enzymes . The different fragments were ligated into plasmid pMF272 ( AY598428 ) [70] , [71] . For the MAK-2-mCherry , SO-mCherry , MEK-2-mCherry and NRC-1-mCherry strains , plasmid TSL84C was used . To generate plasmid pTSL84C , sGFP , from plasmid pMF272 , was removed by digestion with PacI and EcoRI restriction enzymes and replaced by a version of mCherry that is codon-optimized for N . crassa and includes a C-terminal 6x-His-tag ( mCherryNc-6xHis ) . Plasmid pMFP26 [72] contains untagged codon-optimized mCherryNc and was used as a template for PCR using forward primer OTS177 ( AAATTAATTAACGTGAGCAAGGGCGAGGAGGATAAC ) and reverse primer OTS202 ( AAAGAATTCCTAGTGGTGGTGGTGGTGGTGGCTGCCCTTGTACAGCTCGTCCATGCCGCCG ) , which contained information for the 6xHis tag and a stop codon . Plasmid derivatives with the tef-1 promoter instead of ccg-1 promoter were obtained by swapping the ccg-1 promoter for the tef-1 promoter using the restriction enzymes NotI and XbaI . A tandem construct of tef-1-ham-5-gfp and tef-1-mak-2-mCherry was created by digesting the tef-1-mak-2-mCherry pMF272 plasmid with restriction enzymes PspOMI and BstBI and the tef-1-ham-5-gfp construct with NotI and BstBI . The latter fragment was ligated into the tef-1-mak-2-mCherry pMF272 plasmid to create tef-1-mak-2-mCherry and tef-1-ham-5-gfp . All constructs were transformed into the WT his-3 strain with selection for His+ prototrophy . Homokaryotic strain was obtained via microconidial purification [73] . A strain bearing cytoplasmic GFP was obtained by transformation of the empty pMF272 plasmid into the WT his-3 strain . All micrographs with HAM-5-GFP are with strains bearing the tef-1-ham-5-gfp constructs unless stated otherwise . Deletion strains were obtained from the FGSC [74] that were generated as part of the N . crassa functional genomics project [69] , [75] . For each deletion strain , both the mating type A and mating type a strains were analyzed , if available . To assess the ability of fusion between conidia of a deletion strain as compared to wild type , slant tubes containing the strains were grown for 4–6 days or until significant conidiation occurred . Conidia were harvested by vortexing the slant tube with 2 ml ddH2O and subsequently filtered by pouring over cheesecloth to remove hyphal fragments . Conidia were diluted to a concentration of 3 . 3×107 conidia/ml . For each sample , 300 µl of spore suspension was spread on a 9 cm solid VMM plate . The plates were dried in a fume hood for 20-30 minutes and incubated for 3–4 hours at 30°C . Squares of 1 cm were excised and observed with a Zeiss Axioskop 2 using a 40× Plan-Neofluor oil immersion objective . The ability of germlings to communicate was determined by evaluating whether germlings displayed homing behavior when germinated conidia were within ∼15 um of each other . Samples for protein extraction were grown for 4 . 5 h at 30°C and 200 rpm . Shaking was stopped and samples were grown for 20 minutes longer to encourage cell-cell interaction . The mak-2Q100G strain was treated with 10 µM 1NM-PP1 final concentration in DMSO or with DMSO alone for 10 minutes . Cells were harvested by filtration and frozen before protein extraction . Protein was extracted using the Trizol procedure ( according to manufacture's protocols ) and was kept in a solution of 6 M guanidine , 50 mM ammonium bicarbonate at pH 7 . 4 . 200 µg for each sample was used for guanidine digestion: 1 ) pH was adjusted to pH 7 . 4 with the iTRAQ resuspending buffer ( 500 mM ) , 2 ) 1 hr reduction using 5 mM DTT at 56°C , 3 ) 1 h alkylation using 10 mM iodoacetamide at RT in the dark , 4 ) samples were diluted 10 fold with 25 mM NH4HCO3 , pH 7 . 8 , 5 ) 2 mM CaCl2 and trypsin was added at a 1∶50 ( trypsin-to-protein ) ratio and incubated for 3 hr at 37°C with gentle shaking , 6 ) trypsin was added again in the same ratio and samples were incubated over-night at 37°C with gentle shaking , 7 ) subsequently , a standard C18 Solid phase extraction ( SPE ) was performed with 80% ACN and no TFA , 8 ) samples were dried in a Speed-Vac and a bicinchoninic acid ( BCA ) assay was performed . For 4-plex iTRAQ labeling , 100 µg lyophilized sample per iTRAQ label was used: 1 ) samples were reconstituted with 30 . 0 µL of dissolution buffer ( 500 mM triethylammonium bicarbonate ) , sonicated and vortexed to resuspend the peptides , 2 ) sample pH was checked ( ∼pH 8 . 5 ) , 3 ) each vial of iTRAQ reagent ( 114 , 115 , 116 and 117 ) was brought to room temperature and 70 µL of ethanol was added to each iTRAQ reagent vial , 4 ) samples were vortexed for 1 min and spun down , 5 ) each labeled reaction mix was added to one separate sample , 6 ) samples were vortexed , spun down and incubated for 1 hr at RT . Samples were subsequently hydrolyzed by adding 300 uL of 0 . 05% TFA ( 3 times the volume ) , vortexed , spun down and incubated at room temperature for another 30 min , 7 ) samples were concentrated to 40 µL using a Speed-Vac , 8 ) samples were pooled into a fresh 2 mL silanized tube and concentrated to ∼100 µL using a Speed-Vac before a desalting ( SPE C18 ) step was performed . For phosphopeptide enrichment , magnetic Ni-NTA-agarose beads were obtained from Qiagen ( Valencia , CA Part N#36111 ) : 1 ) 50 µL of the 5% suspension metal ion activated NTA was used for 100 ug peptides , 2 ) beads were first prepared by washing 3× with nano-pure water ( 800 . 0 µL of water per 1 . 0 mL of bead suspension ) , 3 ) beads were then treated with 100 mM EDTA , pH 8 . 0 ( 800 . 0 µL of 100 mM EDTA per 1 . 0 mL of bead suspension ) for 30 min with end-over-end rotation , 4 ) EDTA solution was removed , and beads were washed 3× with nano-pure water ( at the same ratio ) . Subsequently , the beads were treated with 10 mM aqueous metal ion solution ( 800 . 0 uL of 100 mM FeCl3 per 1 . 0 mL of bead suspension ) for 30 min with end-over-end rotation , 5 ) after removing excess metal ions , beads were washed 3× with water ( at the same ratio ) , and resuspended in 1∶1∶1 acetonitrile/methanol/0 . 01% acetic acid for aliquoting into microcentrifuge tubes , 6 ) peptide samples were resuspended in 200 . 0 µl wash/resuspension buffer ( 80% acetonitrile , 0 . 1% TFA ) , 7 ) beads were washed with of 80% acetonitrile with 0 . 1% TFA ( 200 µL of 80% acetonitrile per 50 . 0 µL of beads ) and precipitated using the magnetic stand; the supernatant was discarded . The resuspended samples ( 100 µg peptides in 200 µl of 80% acetonitrile , 0 . 1% TFA ) were added to the activated beads and incubated for 30 min with end-over-end rotation , 8 ) beads were precipitated using the magnetic stand and washed for 1 min with 80% acetonitrile , 0 . 1% TFA . This step was repeated three more times , 9 ) phosphopeptides were eluted from the beads using an appropriate amount of elution buffer ( 50 . 0 µL of the elution buffer per every 50 . 0 uL of beads/100 . 0 ug of peptides ) after incubating for 5 min , 10 ) the samples were then acidified to pH 4 . 0 by concentrating the samples down to 5−10 µl in a Speed-Vac and reconstituted in 30 µL with 0 . 1% TFA . All peptide samples were analyzed using an automated home-built constant flow nano LC system ( Agilent ) coupled to an LTQ Orbitrap Velos mass spectrometer ( Thermo Fisher Scientific ) operating in data-dependent mode [76] . Electrospray emitters were custom made using either 360 µm o . d . ×20 µm i . d . chemically etched fused silica . The nano LC system for phosphoproteomics analysis has an online 4-cm×360 µm o . d . ×150 µm i . d . C18 SPE column ( 5- µm Jupiter C18 , Phenomenex , Torrence , CA ) to desalt each phosphopeptide sample ( 20 µL ) , which is connected to a home-made 60-cm×360 µm o . d . ×50 µm i . d . capillary column ( 3- µm Jupiter C18 , Phenomenex , Torrence , CA ) . Mobile phase flow rate was 100 nL/min and consisted of 0 . 1 M acetic acid in water and 0 . 1 M acetic acid in 70∶30 ( v/v ) acetonitrile:water . For each sample , three technical replicates of LC-MS analyses were performed as shown in Figure S1 . These included ( i ) an LC gradient of 300 min with the LTQ Orbitrap Velos mass spectrometer acquiring higher-energy collisional dissociation ( HCD ) scans; ( ii ) an LC gradient of 300 min with the LTQ Orbitrap Velos mass spectrometer acquiring alternating collision-induced dissociation ( CID ) , ETD ( electron transfer dissociation ) , and higher-energy collisional dissociation ( HCD ) scans; ( iii ) an LC gradient of 180 min with the LTQ Orbitrap Velos mass spectrometer acquiring alternating collision-induced dissociation ( CID ) , ETD ( electron transfer dissociation ) , and higher-energy collisional dissociation ( HCD ) scans . For peptide identification , MS/MS spectra were searched against a decoy Neurospora protein sequence database using SEQUEST [77] . Search parameters included: trypsin enzyme specificity with a maximum of two missed cleavages , +/- 50 ppm precursor mass tolerance , +/- 0 . 05 Da product mass tolerance , and carbamidomethylation of cysteines and iTRAQ labeling of lysines and peptide N-termini as fixed modifications . Allowed variable modifications were phosphorylation of serine , threonine or tyrosine residues . MSGF spectra probability value [78] was also calculated for peptides identified from SEQUEST search . Measured mass accuracy and MSGF spectra probability were used to filter identified peptides to <1% false discovery rate ( FDR ) at spectrum level . iTRAQ reporter ions were extracted using the MASIC software [79] within 10 ppm mass tolerance of each expected iTRAQ reporter ion from each MS/MS spectrum . The sum of the individual iTRAQ reporter ion values from all MS/MS spectra for a given peptide was used for calculating their relative abundance across different conditions . To correct any systematic error due to pipetting , data were normalized by the median of iTRAQ reporter ion of the individual sample . Fold change of each phosphopeptide was calculated by dividing the data points from the two different conditions ( i . e . control and 1NM-PP1-treated ) and transformed into Log2 scale . Statistical analyses were performed using Students t-test . Only data with changes exceeding 1 . 5× greater in control versus 1NM-PP1-treated ( p<0 . 05 ) were considered differential . The strain used to cross so-gfp into a Δham-5 strain was AF-SoT8 and the strain used to cross mak-2-gfp into a Δham-5 strain was AF-M512 ( Table S2 ) . Oscillation studies performed with HAM-5-GFP and mCherry tagged strains were prepared as described above with modifications from [20] . Images were taken using a Leica SD6000 microscope with a 100×1 . 4 NA oil-immersion objective equipped with a Yokogawa CSU-X1 spinning disk head and a 488-nm or 561-nm laser controlled by Metamorph software . Multiple pairs of interacting germlings were analyzed per experiment and representative pairs are shown for each strain . The ImageJ software was used for image analysis . Harvested conidia ( 1×106/ml ) were inoculated in 100 ml VMM in flasks and incubated for 2 . 5 hrs at 30°C with shaking at 200 rpm , then an additional 2 . 5 hrs at 30°C without shaking . Germlings from 3 flasks were harvested by vacuum filtration over a nitrocellulose membrane and frozen in liquid nitrogen . Protein extraction from ground mycelium was performed using 1 ml lysis buffer described in [8] containing complete protease inhibitors , phosphatase inhibitor and Triton X-100 . 20 µl supernatant was used for western blotting and the remaining fraction was used for immunoprecipitation using Protein G Dynabeads ( Invitrogen ) , according to manufacturer's instructions , with the following exceptions: mouse or rabbit anti-GFP antibody ( Roche or Life Technologies , respectively ) or rabbit anti-mCherry antibody ( Bio-vision ) was covalently bound to the beads using BS3 ( Sulfo-DSS , Fisher scientific ) or DMP ( dimethylpimelimidate ) according to manufacturer's instructions . Supernatant samples were incubated with the beads overnight at 4°C . Beads were washed with standard PBS for three times before protein was removed from the beads by heating at 70°C for 10 min in 1× loading buffer , and samples were run on a 4–12% Nu-Page Bis-Tris GelGel ( NOVEX , Life Technologies ) . Protein gels were subjected to Western blot analysis using standard methods . Samples for the MAK-1 and MAK-2 phosphorylation western blots were treated similarly , except , after protein extraction with 1 ml lysis buffer , 25 µl of protein sample was directly loaded on a 7% NuPage Bis-Tris GelGel ( NOVEX , Life Technologies ) protein gel . Gels were subjected to Western blot analysis using standard methods and detection of phosphorylated MAK-1 and MAK-2 was carried out using anti-phospho p44/42 MAP kinase antibodies ( 1∶3000 dilution ) ( PhosphoPlus antibody kit; Cell Signaling Technology ) as described [8] . Detection of phosphorylated HAM-5-GFP was performed using anti-phosphothreonine-proline/phosphoserine-proline antibodies ( Abcam ) . | Cell fusion between genetically identical cells of the fungus Neurospora crassa occurs when germinating asexual cells ( conidia ) sense each other's proximity and redirect their growth . Chemotropic growth is dependent upon the assembly of a MAPK cascade ( NRC-1/MEK-2/MAK-2 ) at the cell cortex ( conidial anastomosis tubes; CATs ) , followed by disassembly over an ∼8 min cycle . A second protein required for fusion , SO , also assembles and disassembles at CAT tips during chemotropic growth , but with perfectly opposite dynamics to the MAK-2 complex . This process of germling chemotropism , oscillation and cell fusion is regulated by many genes and is poorly understood . Via a phosphoproteomics approach , we identify HAM-5 , which functions as a scaffold for the MAK-2 signal transduction complex . HAM-5 is required for assembly/disassembly and oscillation of the MAK-2 complex during chemotropic growth . Our data supports a model whereby regulated modification of HAM-5 controls the disassembly of the MAK-2 MAPK complex and is essential for modulating the tempo of oscillation during chemotropic interactions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"fungal",
"biochemistry",
"signal",
"transduction",
"mycology",
"mitogenic",
"signaling",
"fungi",
"cell",
"biology",
"neurospora",
"crassa",
"fungal",
"genetics",
"genetics",
"biology",
"and",
"life",
"sciences",
"molecular",
"genetics",
"molecular",
"cell",
"biology",
"cell",
"signaling",
"molecular",
"biology",
"neurospora",
"organisms"
] | 2014 | HAM-5 Functions As a MAP Kinase Scaffold during Cell Fusion in Neurospora crassa |
Protein S-palmitoylation , a hydrophobic post-translational modification , is performed by protein acyltransferases that have a common DHHC Cys-rich domain ( DHHC proteins ) , and provides a regulatory switch for protein membrane association . In this work , we analyzed the presence of DHHC proteins in the protozoa parasite Giardia lamblia and the function of the reversible S-palmitoylation of proteins during parasite differentiation into cyst . Two specific events were observed: encysting cells displayed a larger amount of palmitoylated proteins , and parasites treated with palmitoylation inhibitors produced a reduced number of mature cysts . With bioinformatics tools , we found nine DHHC proteins , potential protein acyltransferases , in the Giardia proteome . These proteins displayed a conserved structure when compared to different organisms and are distributed in different monophyletic clades . Although all Giardia DHHC proteins were found to be present in trophozoites and encysting cells , these proteins showed a different intracellular localization in trophozoites and seemed to be differently involved in the encystation process when they were overexpressed . dhhc transgenic parasites showed a different pattern of cyst wall protein expression and yielded different amounts of mature cysts when they were induced to encyst . Our findings disclosed some important issues regarding the role of DHHC proteins and palmitoylation during Giardia encystation .
The flagellated protozoan parasite Giardia lamblia is a major cause of non-viral/non-bacterial diarrheal disease worldwide . This parasite can cause asymptomatic colonization or acute or chronic diarrheal illness and malabsorption [1] . Infection begins with the ingestion of Giardia in its cyst form which , after exposure to gastric acid in the host stomach and proteases in the duodenum , gives rise to trophozoites . The inverse process is called encystation and begins when the trophozoites migrate to the lower part of the small intestine where they receive signals that trigger synthesis of the components of the cyst wall . The encystation process is tightly regulated but the exact mechanism that controls this process is still obscure . Expression of the three Cyst Wall Proteins ( CWP ) and the glycopolymer biosynthetic enzymes , is largely upregulated . In addition , several other proteins , whose roles in encystation are yet to be discovered , are upregulated at the transcriptional level [2] , [3] . Various protein posttranslational modifications ( PTM ) have been implicated in the development of encystation , such as phosphorylation [4] and deacetylation [5] , among others [6] , [7] , [8] . There is also some evidence of the role of PTM in gene regulation for the control of this process [9] . Protein S-palmitoylation ( hereafter referred to as palmitoylation ) , the post-translational addition of palmitic acid ( 16∶0 ) to cysteine residues of proteins , is a PTM essential for proper membrane trafficking to defined intracellular membranes or membrane sub-domains , protein stability , protein turnover , and vesicle fusion [10] , [11] , [12] . Unlike the other lipid modifications , palmitoylation is potentially reversible , providing a regulatory switch for membrane association [13] , [14] . Palmitoylation is catalyzed by a family of protein acyltransferases ( PATs ) , which transfer a palmitoyl moiety derived from palmitoyl-CoA to a free thiol of a substrate protein to create a labile thioester linkage [15] , [16] . The discovery of these enzymes came through studies in yeast that identified the PATs Erf2 and Akr1 , which are active against Ras and casein kinase , respectively [17] , [16] . These enzymes are polytopic integral membrane proteins which share the conserved Asp-His-His-Cys ( DHHC ) - cysteine-rich domain ( CRD ) . The general membrane topology predictions indicate that the core structure of a PAT is four transmembrane domains ( TMDs ) , with the N- and C- terminus in the cytoplasm [18] . The signature feature DHHC-CRD , which is indispensable for palmitoylating activity , is located in the cytoplasmic loop between the second and third TMDs [19] . There is a small group of PATs that display six TMDs with an extended N-terminal region encoding ankyrin repeats . The yeast PAT called Akr1 is a member of this group [16] , [20] . All these findings were crucial in defining palmitoylation as an enzymatic process and led to subsequent identification of protein acyltransferases in many other organisms , such as mammals [21] , [22] , plants [23] , and protozoan parasites like Toxoplasma gondii [24] , [25] , Plasmodium [26] , [25] , and Trypanosoma brucei [27] . There is scarce knowledge about palmitoylation in Giardia , but some findings indicate that this PTM may play an important role in pathogenesis . It was shown that α19-giardin , one of the major protein components of the Giardia cytoskeleton , can be both myristoylated and palmitoylated [28] and that the variant-specific surface proteins ( VSPs ) may be palmitoylated within their C-terminal domains [29] , [30] . Later , Touz et al . determined the exact site of palmitoylation of the VSPs , characterized the enzyme responsible for this modification , and determined the participation of palmitoylation during antigenic variation [31] , a process in which the trophozoite continuously changes its surface antigen coat [32] . Antigenic variation and encystation are two distinctive mechanisms of defense that the parasite has developed to survive in hostile environmental conditions during its life cycle , and it has been suggested that both are mechanistically related processes [33] . Accumulation of material in membrane vesicles followed by transport and vesicle fusion and secretion are some of the main events involved in Giardia encystation . Because palmitoylation has been reported to play a key role in these events in other cell types [12] , [10] , [34] , [35] , [36] , it is likely that this PTM may also play a role in Giardia encystation . In this work , we address the question of whether PATs and palmitoylation itself are involved in Giardia encystation . We provide evidence about the role of palmitoylation in Giardia encystation biology by inhibiting this PTM with 2-bromopalmitate ( 2-BP ) or 2-fluoropalmitate ( 2-FP ) . Using bioinformatics , we identified the potential PATs ( hereafter called DHHC proteins ) in the Giardia lamblia proteome and performed a phylogenetic analysis of these proteins . We evaluated the expression of the total collection of DHHC proteins in trophozoites and encysting parasites . Using dhhc transgenic Giardia parasites , we revealed the intracellular localization of DHHC proteins and their influence in CWP expression and cyst yield when parasites were induced to encyst . Our data suggest a role of palmitoylation and DHHC proteins in encystation , providing an insight into the impact of this PTM in Giardia survival .
Trophozoites of the isolate WB , clone 1267 [37] , were cultured in TYI-S-33 medium supplemented with 10% adult bovine serum and 0 . 5 mg ml−1 bovine bile ( Sigma , St . Louis , MO ) as described [38] . GL50806_40376 ( High Cysteine Non-variant Cyst protein; HCNCp ) , GL50803_1908 , GL50803_2116 , GL50803_16928 , and GL50803_8711 open reading frames ( ORF ) were amplified from genomic DNA . GL50806_40376 was cloned into the vector pTubV5-pac [39] to generate pHCNCp-V5 plasmid . GL50803_1908 , GL50803_2116 , GL50803_16928 , and GL50803_8711 were each one cloned into the vector pTubHA-pac [39] to generate the pDHHC-HA plasmids . Trophozoites were transfected with the constructs by electroporation and selected by puromycin ( Invivogen , San Diego , CA ) as previously described [40] , [41] , [42] . Trophozoites transfected with empty pTubHA-pac or pTubV5-pac plasmids were used as control . Primer sequences used for DHHC proteins cloning are depicted in table S1 . Encystation was induced by growing trophozoites for one culture cycle in TYI-S-33 medium without bile ( pre-encystation ) . Bile-deficient medium was poured off along with unattached trophozoites and replaced with warmed encysting medium containing 0 . 45 mg ml−1 porcine bile ( Sigma , St . Louis , MO ) and 0 . 25 mg ml−1 lactic acid ( Sigma , St . Louis , MO ) , pH 7 . 8 , and incubated at 37°C for 48 h [43] . Total encysting cultures were harvested at 48 h by chilling and centrifugation , and subsequently used for palmitoylation assay , RNA extraction , western blot , immunofluorescence , or flow cytometry . The assay followed the procedure described by Papanastasiou et al . and Corvi et al . [29] , [44] . Briefly , 8×106 growing and encysting wild-type or dhhc transgenic parasites were washed , suspended in 1 ml of RPMI ( Gibco , Invitrogen , Carlsbad , CA ) containing 200 µCi of [9 , 10-3H ( N ) ]-palmitic acid ( Perkin-Elmer , MA ) , previously conjugated to BSA fatty acid free ( 1∶1 , mol∶mol ratio ) , and incubated for 4 h at 37°C . The samples were then suspended on SDS–PAGE loading buffer without any reducing agent and loaded onto SDS-PAGE gel . The gel was then incubated for 30 min in ddH2O and for 30 min more in 1M sodium salicylate pH 6 . 5 . The gel was then incubated with 3% glycerol , 10% acetic acid , and 40% methanol for 30 min , dried for 2 h at 80°C using a gel dryer machine , and exposed to autoradiographic film for a month . For hydroxylamine treatment , the gel was soaked in either 1 M NH2OH- NaOH pH 7 . 0 or 1 M Tris-HCl pH 7 . 0 ( Control ) for 48 h . Finally , the gel was incubated for 30 min in ddH2O and for 30 min more in 1M sodium salicylate pH 6 . 5 , dried as described above , and exposed to autoradiographic film for a month . Total cellular palmitoylated proteins from growing and encysting wild-type or transgenic ( overexpressing HCNCp ) parasites , were purified following the procedure described by Wan et al . [45] . Briefly , 5×107 trophozoites or 48 h encysting parasites were harvested and lysed with Lysis buffer ( LB; 50 mM Tris-HCl pH 7 . 4 , 5 mM EDTA , 150 mM NaCl ) with 10 mM N-Ethylmaleimide ( NEM; Thermo Scientific Pierce Rockford , IL ) plus protease inhibitors . After sonication , 1 . 7% of Triton X-100 was added to each sample and incubated for 1 h at 4°C under shacking . The samples were then centrifuged at 500× g for 5 min at 4°C . The supernatant was collected in a new tube and solubilized proteins were precipitated with chloroform-methanol . Proteins were resolubilized in 4% SDS buffer ( SB; 4% SDS , 50 mM Tris-HCl pH 7 . 4 , 5 mM EDTA ) with 10 mM NEM by incubating at 37°C under shacking . Each sample was then diluted with 3 vol of LB with 1 mM NEM , protease inhibitors , and 0 . 2% Triton X-100 and incubated overnight at 4°C under shacking . Proteins were then precipitated by three sequential chloroform-methanol extractions after which each sample was dissolved in SB and split into two equal fractions: one for neutral pH hydroxylamine treatment ( hyd+ ) and the other for neutral pH Tris buffer treatment ( hyd− ) . The hyd+ portion was diluted with 4 vol of hyd+ buffer ( 1M hydroxylamine pH 7 . 4 , 150 mM NaCl , 1 mM HPDP-Biotin , 0 . 2% Triton X-100 , protease inhibitors ) , and the hyd- portion with 4 vol of the hyd- buffer ( 50 mM Tris-HCl pH 7 . 4 , 5 mM EDTA , 150 mM NaCl , 1 mM HPDP-Biotin ( Thermo Scientific Pierce , Rockford , IL ) , 0 . 2% Triton-X-100 , protease inhibitors ) and incubated for 1 h at room temperature under shacking , followed by chloroform-methanol precipitation . The samples were then resuspended in SB at 37°C under shacking . Protein pellets were solubilized in LB containing 0 . 2% Triton X-100 . Streptavidin-agarose ( Thermo Scientific Pierce , Rockford , IL ) was added at concentration of 25 µl beads ml−1 and the lysate and samples were incubated for 1 h at room temperature . Unbound proteins were removed by four sequential washes with LB containing 0 . 2% Triton X-100 . Samples were finally eluted with 100 mM DTT containing 0 . 2% Triton X-100 . Each eluate was then analyzed by Western blotting . Giardia trophozoites were cultured as described above . 2-bromopalmitate ( 2-BP ) ( Sigma-Aldrich , St . Louis , MO ) or 2-fluoropalmitate ( 2-FP ) ( Cayman Chemical , Ann Arbor , MI ) were added to the media for 48 h to reach a final concentration of 10 , 20 , 40 , 50 , 75 or 100 µM for 2-BP , and 100 , 150 or 200 µM for 2-FP . The inhibitors were diluted in DMSO ( Sigma-Aldrich , St . Louis , MO ) following manufacturer indications . The parasites were then analyzed by staining them with Trypan blue to distinguish live from dead cells and by counting them in a Neubauer chamber . To perform a growth curve , parasites from three independent experiments were counted . Parasites were induced to encyst as described above . 2-BP or 2-FP were added with encysting media for 48 h to reach a final concentration of 10 , 20 or 40 µM for 2-BP , and 100 µM for 2-FP . The inhibitors were diluted in DMSO as mentioned above . For immunofluorescence the parasites were subcultured onto 12 mm round glass coverslips ( Glaswarenfabrik Karl Hecht , Sondhein , Germany ) in 24-well culture plates for 1 h , fixed with 4% paraformaldehyde in PBS for 20 min at 4°C , washed twice in PBS and blocked with 10% normal goat serum ( Invitrogen , Carlsbad , CA ) in 0 . 1% Triton X-100 in PBS for 30 min at 37°C . The samples were then incubated with FITC labeled anti-CWP1 mAb ( Waterborne Inc . , New Orleans , LA ) diluted 1∶250 in PBS containing 3% normal goat serum and 0 . 1% Triton X-100 for 1 h at 37°C or anti-CWP1 mAb and DAPI diluted in PBS ( dilution 1∶500 ) ( Sigma , St . Louis , MO ) . The coverslips were then mounted onto glass slides using FluorSave reagent ( Calbiochem , La Jolla , CA ) . Fluorescence was visualized in a Zeiss Axiovert 200 microscope ( Carl Zeiss , Jena , Germany ) . To quantify the percentage of encysting parasites , 55 cells from three separate experiments were counted and classified as encysting I , encysting II , or cyst according to the cell shape , membrane staining , and number and size of the encystation-specific vesicles . The average was taken in each of the three groups . A proteome database was constructed gathering complete proteomes for 25 Metazoa ( Amphimedon queenslandica ( aqu ) , Anolis carolinensis ( aca ) , Apis mellifera ( apm ) , Bombyx mori ( bmo ) , Caenorhabditis elegans ( cae ) , Canis familiaris ( cfa ) , Ciona intestinalis ( cin ) , Danio rerio ( dre ) , Daphnia pulex ( dpu ) , Drosophila melanogaster ( dme ) , Equus caballus ( eqc ) , Felis catus ( fca ) , Gallus gallus ( gga ) , Gorilla gorilla ( ggo ) , Homo sapiens ( hsa ) , Ixodes scapularis ( ixs ) , Mus musculus ( mmu ) , Nematostella vectensis ( nve ) , Ornithorhynchus anatinus ( oan ) , Petromyzon marinus ( pma ) , Pteropus vampyrus ( pva ) , Rattus norvegicus ( rno ) , Schistosoma mansoni ( sma ) , Sus scrofa ( ssc ) and Xenopus tropicalis ( xtr ) ) , 18 Fungi ( Aspergillus nidulans ( and ) , Batrachochytrium dendrobatidis ( bde ) , Botryotinia fuckeliana ( bfu ) , Candida albicans ( clb ) , Encephalitozoon cuniculi ( ecu ) , Gibberella zeae ( gze ) , Leptosphaeria maculans ( lem ) , Nematocida sp ( nsp ) , Neurospora crassa ( ncr ) , Pichia pastoris ( ppa ) , Puccinia graminis ( pug ) , Saccharomyces cerevisiae ( sce ) , Schizosaccharomyces pombe ( szp ) , Sclerotinia sclerotiorum ( scl ) , Tuber melanosporum ( tme ) , Ustilago maydis ( uma ) , Vittaforma corneae ( vco ) and Yarrowia lipolytica ( yli ) ) , 12 Plants ( Arabidopsis thaliana ( ath ) , Brachypodium distachyon ( bdi ) , Glycine max ( gmx ) , Medicago truncatula ( met ) , Oryza sativa ( osa ) , Physcomitrella patens ( php ) , Populus trichocarpa ( pot ) , Selaginella moellendorffii ( smo ) , Solanum lycopersicum ( sly ) , Solanum tuberosum ( stu ) , Sorghum bicolor ( sbi ) and Vitis vinifera ( vvi ) ) , 1 Brown alga ( Aureococcus anophagefferens ( aan ) ) , 1 Red alga ( Cyanidioschyzon merolae ( cym ) ) , 3 Green algae ( Ostreococcus taurii ( ota ) , Chlamydomonas reinhardtii ( chr ) and Chlorella variabilis ( chv ) ) , and 24 Protists ( Babesia bovis ( bbo ) , Bigelowiella natans ( bna ) , Chlamydomonas reinhardtii ( chr ) , Chlorella sp ( chl ) , Cryptosporidium parvum ( cpv ) , Dictyostelium discoideum ( ddi ) , Entamoeba histolytica ( ehi ) , Giardia lamblia ( gla ) , Guillardia theta ( gth ) , Leishmania major ( lma ) , Paramecium tetraurelia ( pat ) , Perkinsus marinus ( pem ) , Phaeodactylum tricornutum ( pht ) , Phytophthora capsici ( pcs ) , Phytophthora ramorum ( pra ) , Plasmodium falciparum ( pfa ) , Polysphondylium pallidum ( pop ) , Tetrahymena thermophila ( tet ) , Thalassiosira pseudonana ( thp ) , Theileria parva ( thp ) , Toxoplasma gondii ( tgo ) , Trichomonas vaginalis ( tva ) , Trypanosoma brucei ( trb ) and Trypanosoma cruzi ( tcz ) ) from Ensembl , the Joint Genome Institute ( JGI ) and the NCBI databanks . zf-DHHC HMMer profile was obtained from Pfam [46] , and used to search the proteomes database [47] . Incomplete sequences or those that did not start with the M residue were deleted from the dataset . Also , 90% similar amino acid sequences were clustered using CD-HIT web server with default settings , to reduce the redundancy of the set [48] . The final dataset contained 1034 amino acid sequences . Multiple sequence alignment of DHHC-CRD amino acid sequences was carried out using PROMALS3D online server with default settings [49] . Following manual curation using GeneDoc software [50] , sequences lacking conservation in the regions of interest ( i . e . , DPG , DHHC-CRD and TTxE ) were removed . Block Mapping and Gathering with Entropy ( BMGE ) [51] was used to select columns suitable for phylogenetic inference with the following settings: m = BLOSUM30 , g = 0 . 2 , b = 4 . Phylogenetic analysis was performed by Maximum Likelihood ( ML ) using PhyML [52] with approximate likelihood-ratio test ( aLRT ) , in combination with the LG+G amino acid replacement matrix , which was determined by ProtTest to be the model of protein evolution which best fit the data [53] . Phylogenetic trees were generated and edited with Itol [54] . RNA from WB1267 trophozoites or 48 h encysting WB1267 was extracted and purified using TRIzol reagent ( Invitrogen , Carlsbad , CA ) and SV total RNA Isolation System ( Promega , Madison , WI ) . Total RNA were reverse transcribed using Revertaid reverse transcriptase according to the manufacturer's specifications ( Fermentas , Thermo Scientific , PA ) . DNA contamination was tested by performing PCR in a “-RT” control ( a mock reverse transcription containing all the RT-PCR reagents , except the reverse transcriptase . For PCR , 30 cycles ( 30 s at 94°C , 30 s at 55°C and 1 min at 72°C ) were used ending with a final extension of 10 min at 72°C . The expression of the Giardia glutamate dehydrogenase ( gdh ) gene was assayed for positive control . Aliquots ( 50 µl ) of the RT-PCR reaction were size-separated on 1% agarose gel prestained with SYBR Safe ( Invitrogen , Carlsbad , CA ) . Primers sequences used in RT-PCR are displayed in table S2 . These assays were performed four times in duplicates . RNA from WB1267 trophozoites , 48 h encysting WB1267 or dhhc transgenic 48 h encysting cells ( GL50803_1908 , GL50803_2116 , GL50803_16928 , GL50803_8711 ) was extracted and purified as described above . 2 µg of total RNA were reverse transcribed using Revertaid reverse transcriptase according to the manufacturer's specifications ( Fermentas , Thermo Scientific , PA ) . DNA contamination was tested as described above . cDNA samples were stored at −80°C until use . Control samples were prepared as above using nuclease-free ddH2O in place of RNA . Primers for PCR were designed using Primer express 3 . 0 software ( Applied Biosystems , Forster City , CA ) and were synthesized by Invitrogen , Inc . ( Carlsbad , CA ) . Amplification was performed in a final volume of 20 µl , containing 2 µl of each cDNA sample which were previously diluted 1∶1000 ( for dhhc genes ) or 1∶10000 ( for cwp genes ) , and 10 µl of SYBR Green Master Mix ( Applied Biosystems , Foster City , CA ) . qRT-PCR was performed in a StepOne thermal cycler ( Applied Biosystems , Foster City , CA ) . The mRNA levels of the genes studied were normalized to the expression of the Giardia glutamate dehydrogenase ( gdh ) gene . The relative-quantitative RT-PCR conditions were: holding stage: 95°C for 10 min , cycling stage: 40 cycles at 95°C for 15 s , 60°C for 1 min and melt curve stage: 95°C for 15 s , 60°C for 1 min , and 95°C for 15 s . Expression data were determined by using the comparative ΔΔCt method [55] . Primer sequences used in qRT-PCR are displayed in table S3 . For Western Blot assays , parasite lysates or purified palmitoylated proteins were incubated with 2× Laemmli buffer , boiled for 10 min , and separated in 10% Bis-Tris gels using a Mini Protean II electrophoresis unit ( Bio-Rad ) . Samples were transferred to nitrocellulose membranes ( GE Healthcare Biosciences , Pittsburgh , PA ) , blocked with 5% skimmed milk and 0 . 1% Tween 20 in PBS , and later incubated with anti-HA mAb or anti-V5 mAb ( Sigma , St . Louis , MO; dilution 1∶1000 or 1∶50 respectively ) diluted in the same buffer for 1 h . The membrane was then washed , incubated with IDRye 800CW conjugated goat anti-mouse Ab ( LI-COR , Lincoln , NE; dilution 1∶10000 ) for 1 h , and analyzed on the Odyssey scanner ( LI-COR , Lincoln , NE ) . For the analysis of VSPs expression , blockage was performed with 5% skimmed milk and 0 . 1% Tween 20 in TBS , and then incubated with 5C1 anti-VSP1267 mAb diluted in the same buffer for 1 h . After washing and incubation with an enzyme-conjugated secondary antibody , proteins were visualized with the SuperSignal West Pico Chemiluminescent Substrate ( Pierce , Thermo Fisher Scientific Inc . , Rockford , IL , USA ) and autoradiography . Controls included the omission of the primary antibody , the use of an unrelated antibody , or assays using non-transfected cells . For immunofluorescence assays ( IFA ) , trophozoites or encysting cells cultured in growth medium or encysting medium , respectively , were harvested and washed two times with PBSm ( 1% growth medium in PBS , pH 7 . 4 ) and allowed to attach to multi-well slides in a humidified chamber at 37°C for 30 min . After fixation with 4% formaldehyde ( Sigma , St . Louis , MO ) in PBS for 40 min at room temperature , the cells were washed with PBS and blocked with 10% normal goat serum ( Invitrogen , Carlsbad , CA ) in 0 . 1% Triton X-100 in PBS for 30 min at 37°C . Cells were then incubated with the anti-HA mAb ( Sigma , St . Louis , MO; dilution 1∶500 ) in PBS containing 3% normal goat serum and 0 . 1% Triton X-100 for 1 h at 37°C , followed by incubation with Alexa 546-conjugated goat anti-mouse ( dilution 1∶500 ) secondary antibody at 37°C for 1 h . Encysting cells were also incubated with FITC-conjugated anti-CWP1 mAb ( Waterborne Inc . , New Orleans , LA; dilution 1∶250 ) . Alternatively , cells were incubated with 9C3 anti-BiP mAb ( marker for ER ) [56] or 5D2 anti-AP2 mAb ( marker for peripheral vacuoles ) [57] in PBS containing 3% normal goat serum and 0 . 1% Triton X-100 for 1 h at 37°C , followed by incubation with Alexa 546-conjugated goat anti-mouse ( dilution 1∶500 ) secondary antibody at 37°C for 1 h . Samples were then incubated with FITC-conjugated anti-HA mAb ( Sigma , St . Louis , MO; dilution 1∶100 ) . Preparations were stained with DAPI diluted in PBS ( dilution 1∶500 ) ( Sigma , St . Louis , MO ) . Finally , preparations were washed with PBS and mounted in Vectashield mounting medium ( Vector Laboratories , Burlingame , CA ) . Fluorescence staining was visualized with a motorized FV1000 Olympus confocal microscope ( Olympus UK Ltd , UK ) , using 63× or 100× oil immersion objectives ( NA 1 . 32 ) . The fluorochromes were excited using an argon laser at 488 nm and a helio-neon laser at 543 nm . Detector slits were configured to minimize any cross-talk between the channels . Differential interference contrast images were collected simultaneously with the fluorescence images , by the use of a transmitted light detector . Images were processed using Fiji software [58] and Adobe Photoshop 8 . 0 ( Adobe Systems ) software . The colocalization and deconvolution were also performed using Fiji . For the analysis of the amount of cyst yield in dhhc transgenic trophozoites by flow cytometry , the parasites were induced to encyst for 48 h . Trophozoites , encysting cells , and cysts were collected from confluent cultures . Parasites were pelleted by centrifugation at 1455 g for 15 min at 4°C , resuspended in cool sterile ddH2O and placed at 4°C overnight . Mature water-resistant cysts were then processed following the protocol for immunofluorescence ( see above ) without permeabilization . Briefly , parasites were washed two times with PBSm ( 1% growth medium in PBS , pH 7 . 4 ) . After blockade with 10% normal goat serum , the parasites were labeled with anti-CWP1 mAb ( Waterborne Inc , New Orleans , LA; dilution 1∶250 ) diluted in PBSm for 1 hour at 4°C . Cells were then washed twice in PBS and fixed with 4% formaldehyde ( Sigma , St . Louis , MO ) in PBS for 40 min at room temperature . Unlabeled samples were used to determine background fluorescence , and subsequently , fluorescently labeled cysts were analyzed in triplicate on a FACSCanto II flow cytometer ( Becton & Dickinson , New Jersey , NY ) . All samples were analyzed in parallel by IFA to assess encystation efficiency . Results were analyzed for statistical significance ( defined as p<0 . 05 and indicated by asterisks in figures ) by performing unpaired , two-sided Student's t-test with GraphPad Prism 5 Data Analysis Software ( GraphPad Software , Inc . , La Jolla , CA ) . Mean and standard error of mean ( SEM ) values were calculated from at least three biologically and technically independent experiments .
It has been shown that protein palmitoylation actively participates in cell differentiation in a variety of cells [59] , [60] , [61] . The analysis of the expression of palmitoylated proteins , using metabolic labeling with [3H] palmitic acid , showed that encysting Giardia parasites displayed a different pattern of total protein palmitoylation than growing parasites ( Figure 1A , T-ET/hyd− ) . The results showed a band of ∼60 kDa in trophozoites that may correspond to the expressed VSPs [31] ( Figure 1A , T/hyd− ) . However , when Giardia encysting cells were analyzed , the assay displayed a larger amount of palmitoylated proteins , as can be judged by the larger number of bands displayed compared to trophozoites ( Figure 1A , ET/hyd− ) . When we performed neutral treatment with hydroxylamine , almost complete removal of the attached palmitates was observed in both growing and encysting parasites ( Figure 1A , T-ET/hyd+ ) . This confirms that palmitate is attached through a labile thioester linkage ( S-palmitoylation ) in Giardia , as has been observed in other cell types including parasites [62] , being most common among palmitoylated proteins [63] . Protein S-palmitoylation reversibility makes it a flexible , rapid and precise way of protein activity regulation [64] which may be crucial in the encystation process . The fact that the amount of total S-palmitoylated proteins was higher in encysting cells compared to trophozoites suggested that this PTM may play an important role during Giardia differentiation . This observation is in accordance with previous reports showing an important role of protein S-palmitoylation in controlling several crucial processes in parasites such as invasion or motility [44] . During Giardia encystation , the cyst wall proteins ( CWPs ) are sorted , concentrated within encystation-specific vesicles ( ESVs ) , and exported to the nascent cyst wall [65] , [66] , [67] . Thus , the larger amount of palmitoylated proteins observed in encysting parasites ( Figure 1A , ET/hyd− ) may be explained by this additional requirement of protein sorting and export during this stage . In addition to the CWP1 , 2 and 3 , another type of cyst wall protein has been identified , a High Cysteine Non-variant Cyst protein ( HCNCp ) [68] . HCNCp belongs to a large group of cysteine-rich , non-VSPs , Type I integral membrane proteins ( HCMp ) [68] . The palmitoylation prediction algorithm CSS-Palm 3 . 0 [69] strongly predicts that HCNCp is palmitoylated at cysteines 1602 ( CSS-Palm score 6 . 57 , high stringency cut-off 0 . 31 ) and 1603 ( CSS-Palm score 4 . 99 , high stringency cut-off 0 . 31 ) , which are located in the transmembrane region and in the cytosolic tail respectively ( HMMTOP , ( http://www . enzim . hu/hmmtop/ ) [70] , [71] ) . In order to find out whether HCNCp is palmitoylated or not , we performed the following approach: first , we expressed full length HCNCp as a fusion protein containing a C-terminal V5-tag and a tubulin promoter [39] . The expression of the ∼169 kDa HCNCp protein was equally observed in hcncp-V5 transgenic growing and encysting parasites , together with fragments of 21 , 42 and 66 kDa already reported by Davids et al . [68] ( Figure S1 ) . Second , hcncp-V5 transgenic trophozoites ( HCNCp T ) and encysting ( HCNCp ET ) parasites were subjected to acyl biotin exchange ( ABE ) as described in Methods . Parallel plus- and minus-hydroxilamine ( hyd ) samples were analyzed by Western blotting using an anti-V5 mAb ( Figure 1B ) . Only the samples that were treated with hydroxylamine had free cysteine residues able to be detected by biotin/streptavidin ( see Methods ) . When we assayed HCNCp T purified samples , we observed three bands ( 169 , 66 and 21 kDa ) and a weak band of 42 KDa ( Figure 1B , HCNCp T/hyd+ ) . Also , the four bands ( 169 , 66 , 42 , and 21 kDa ) were observed for HCNCp ET purified sample compared to the control ( hyd− ) , showing that not only the full length but also the smaller epitope-tagged fragments of the HCNCp protein were palmitoylated in encysting parasites ( Figure 1B , HCNCp ET/hyd+ ) . The presence of these four bands may account , at least in part , for the bands shown in figure 1A ( Figure 1A , ET/hyd− ) . Although we showed that the constitutively expressed HCNCp can be palmitoylated during growth and encystation , it was clearly reported that HCNCp is almost exclusively expressed during encystation when its expression was analyzed at the mRNA and protein ( expression under its own promoter ) levels [68] . Altogether , these results suggest that HCNCp is likely important during encystation , while the machinery necessary for its palmitoylation remains unaltered during growth and differentiation . Despite the need of additional assays to accurately identify additional palmitoylation substrates , it seems that this PTM is more frequently founded in encysting cells compared to trophozoites . In parallel to HCNCp T and HCNCp ET samples , we also performed ABE in wild-type trophozoites and encysting parasites and analyzed the purified samples by Western blotting using anti-VSP1267 mAb ( Figure 1C ) . The results showed the specific protein band of VSP1267 ( MW ∼60 KDa ) , in both growing and encysting parasites , suggesting that this PTM may be important for VSP function during the entire Giardia life cycle . Further analysis using ABE or click chemistry [72] assays , together with different methods for Mass spectrometry-based proteomics , including Multidimensional protein identification technology [45] , will expand our knowledge about other palmitoylated proteins in Giardia , defining the palmitoyl proteome of this parasite and shedding light on the role of this PTM in its life cycle . The fact that Giardia encysting cells displayed a large amount of palmitoylated proteins prompted us to find out whether inhibition of protein palmitoylation would influence Giardia encystation . Several compounds have been reported to block protein palmitoylation [73] . The 2-bromopalmitate ( 2-BP ) [74] and the 2-fluoropalmitate ( 2-FP ) [73] inhibitors are non-metabolizable palmitate analogs that block palmitate incorporation into proteins using a still unclear mechanism . These two compounds have been widely used , act as broad inhibitors of palmitate incorporation and do not appear to selectively inhibit the palmitoylation of specific protein substrates . To test the effect of these inhibitors during encystation , Giardia wild-type trophozoites were induced to encyst together with the addition of either 2-BP or 2-FP . It has been reported that 2-BP is not well tolerated by in vitro cultured cells and causes cell death even after a brief exposure to 100 µM of 2-BP [75] . Thus , a growth curve was performed to determine the optimal concentrations that do not affect Giardia growth ( 10 , 20 or 40 µM for 2-BP and 100 µM for 2-FP ) , observing that trophozoites died under concentrations higher than 50 µM of 2-BP or 150 µM of 2-FP ( Figure 2A ) . After 48 h of encystation , treated or control parasites were harvested , permeabilized , stained with anti-CWP1 mAb and analyzed by fluorescence microscopy ( Figure 2B ) . Wild-type encysting trophozoites were classified as encysting I ( EI ) ( corresponding to 6 h of encystation [76] ) , encysting II ( EII ) ( corresponding to 12 h of encystation [76] ) , and cysts ( corresponding to 24–48 h of encystation [76] ) ( Figure 2B , upper panel ) , based on the following features: cell shape , membrane staining , and number and size of the ESVs . As shown in figure 2B ( lower panel ) , there was a significant reduction in the amount of cysts when parasites were treated with 2-BP ( 20 µM or 40 µM ) or 2-FP ( 100 µM ) . The effect of 2-BP as a generic palmitoylation inhibitor has been reported in a wide variety of cells [77] , [74] , [78] including parasites like Toxoplasma gondii [62] , although the concentrations used were much higher than the ones we used in this work . Interestingly , with 20 and 40 µM of 2-BP , there was an increase of the encysting II parasites compared to the control , reaching its highest levels when the concentration of 2-BP was 40 µM and resulting also in a diminution of encysting I cells ( Figure 2B , lower panel ) . Thus , the decrease in the amount of cysts may be at the expense of the arrest of the cells at the encysting II stage of differentiation . In order to find out whether the treatment with palmitoylation inhibitors affect DNA replication , we analyzed the number of nuclei in the population of EII cells that were increased , observing no differences compared to the control ( Figure 2C ) . Although a pleiotropic effect of 2-BP cannot be excluded , it is very likely that the observed decrease in cyst formation is associated with the inhibition of palmitoylation and the subsequent defect in ESVs docking and fusion , as was shown to be the case for other cells [79] , [80] . Some results have suggested that palmitoylation in cells may occur nonenzymatically , i . e . spontaneous formation of thioester linkage in the presence of palmitoyl-CoA [81] . However , studies in yeast showed that DHHC protein family-mediated palmitoylation accounted for most of the palmitoylated proteins found in this organism [79] . Therefore , we decided to explore the Giardia proteome to study the presence of DHHC proteins in this parasite . PATs , the discovery of which has been crucial for the enzymology of palmitoylation , are a widespread evolutionary family of proteins [16] , [82] ranging from eight in Saccharomyces cerevisiae [82] , twelve in Trypanosoma brucei [27] , eighteen in Toxoplasma gondii [25] , twelve in Plasmodium [26] , [25] to twenty-three members in humans [82] . To identify the complete set of Giardia putative PATs , we performed a HMMER search against the Giardia complete proteome using a DHHC PAT HMMer profile from Pfam ( zf-DHHC ) . As shown in figure 3A , we found nine DHHC proteins in the Giardia proteome that displayed conserved sequences when compared to other organisms: i ) the DHHC-CRD domain , ii ) the two short motifs DPG ( aspartate-proline-glycine ) and iii ) TTxE ( threonine-threonine-any-glutamate ) motif [20] , [82] . One protein ( gla_8711 ) contained a DHYC amino acid motif , instead of the canonical DHHC motif . However , this DHYC motif has been reported to be functional in the yeast PAT Akr1 [16] . We next analyzed the molecular identity of Giardia DHHC proteins with bioinformatics tools . In agreement with previous reports for other PATs [20] , [18] , [25] , Giardia DHHC proteins were predicted to be polytopic membrane proteins , mainly harboring between three and six TMDs with the DHHC domain facing the cytosol ( Figure 3B ) . There is a small group of DHHC proteins , including yeast DHHC protein Akr1 , displaying the conserved 33 amino acid ankyrin repeats , which are frequently involved in protein-protein interactions [83] . By contrast , none of the Giardia DHHC proteins showed ankyrin repeats in their structure . Moreover , gla_8619 displayed a coiled coil structure and gla_96562 a signal peptide . As already described for other organisms [18] , [25] , Giardia DHHC proteins displayed a conserved structure , sharing domains and motifs that are present across all members of this enzyme family . The names used in this paper , GiardiaDB , NCBI , and UniProt accession numbers for Giardia DHHC proteins are indicated in table 1 . In order to elucidate the phylogenetic relationship among the PATs and to infer the evolutionary history of Giardia DHHC proteins , we retrieved 1034 DHHC-CRD protein sequences from 84 completely sequenced eukaryotic genomes , including the Giardia lamblia genome ( Assemblage A , isolate WB ) , by means of the DHHC PAT HMMer profile from Pfam ( zf-DHHC ) . A Multiple Sequence Alignment was constructed with PROMALS3D [49] , and Block Mapping and Gathering with Entropy ( BMGE ) [51] was used to select columns suitable for Maximum Likelihood ( ML ) phylogenetic inference . Maximum likelihood phylogenetic trees were calculated using PhyML [52] , and Branch support was evaluated by approximate likelihood-ratio test ( aLRT ) [84] . The resultant phylogenetic tree can be divided in six monophyletic clades ( MC ) , three of which together contain almost 90% of all sequences ( MC D , E and F ) . Four MC have Giardia DHHC proteins: MC A and D contain one DHHC sequence each , while MC E and F contain five and two Giardia sequences respectively ( Figure 4A and figures S2 , S3 , S4 , S5 ) . Without any further consideration than the topology of the tree and the early divergent phylogenetic status of Giardia , it can be argued that the Most Recent Common Ancestor of Giardia and the rest of the eukaryotic lineage ( MRCA ) had a minimum of four and a maximum of six groups of PATs . However , of the two Giardia-lacking MC one is almost entirely composed of Plant paralogues ( MC C ) . Moreover , many MC contain subclades composed mostly or even only by Plant paralogues , suggesting that gene duplication have largely taken place in this group . All these can be seen as an indication of functional diversification among Plants , which also constitutes a plausible evolutionary mechanism for the origin of the MC C . If we hypothesize that all DHHC sequences evolve from 4 PATs groups in the MRCA , we should be able to explain , in a parsimonious way , the MC lacking Giardia sequences as examples of evolutionary innovation . As we mentioned before , this is suitable in the case of the MC C , but not for the MC B ( the other Giardia sequences-lacking MC ) . This is because MC B is composed of sequences from a greater variety of organisms compared to MC C , making the possibility of a common functional diversification very unlikely . Nevertheless , it is possible for the MC B to be the result of reductive evolution , meaning that Giardia lost sequences during its adaptation to a parasitic lifestyle , since the more stable environment provided by the host can cause relaxation or loss of selective constraints . We tested gene loss across DHHC-CRD protein family by examining the heavily duplicated genomes of Trichomonas vaginalis , given that duplicated genes are most likely to be released from functional constraints ( Figure 4B ) . For this , we retrieved all DHHC sequences from Trichomonas ( http://trichdb . org/trichdb/ ) using the same pipeline described above , except that this time no sequences were excluded from the posterior analysis . Variations in the HC , C and DHHC portions of the DHHC-CRD domain were extracted from the MSA , and mapped onto a phylogenetic tree . Contrary to what is found in Plants , there is a substantial presence of poorly conserved sequences among Trichomonas genome that cluster together in the tree . Moreover , we found a strong correlation between the degree of conservation in the HC , C and DHHC portions of the DHHC-CRD domain within each sequence . Altogether , our findings suggest that the MRCA had five groups of DHHC sequences from which the other sequences eventually evolved by functional diversification , and that Giardia lost at least one representative sequence presumably during its adaptation to a parasitic lifestyle . We also determined the orthology relationships between sequences from different assemblages . For this , we retrieved DHHC sequences from Giardia isolates WB , GS and P15 ( Assemblages A , B and E , respectively; http://giardiadb . org/giardiadb/ ) , following the pipeline described above . As expected , every DHHC sequence in the isolate WB has a highly similar ortholog in the other isolates , which cluster together in the tree ( Figure 5 ) . Only one WB sequence , EAA36893 , escapes this pattern , but this probably constitutes a case of defective annotation in isolates GS and P15 . Semi-quantitative RT-PCR indicated that all the dhhc genes were expressed in trophozoites and in encysting parasites ( Figure S6 ) . This prompted us to explore further the expression levels of these genes in growing and encysting parasites by performing qRT-PCR analysis of mRNA expression from these cells . As shown in figure 6 , many of the dhhc transcripts were present at relatively constant levels , but gla_8619 , gla_1908 , and EAA36893 were downregulated in encysting parasites while gla_2116 was upregulated in 48 h encysting cells . Considering that Giardia contains minimal systems , either as a result of reductive processes associated with a parasitic lifestyle , as a reflection of basic evolutionary characteristics , or both [85] , [86] , the fact that the nine dhhc genes found by bioinformatics were expressed in vegetative and encysting parasites suggests that protein palmitoylation and the PATs themselves may be playing a key role during the entire life cycle of this parasite . We next sought to characterize four of the nine DHHC proteins that are expressed in Giardia based on their expression profile . We chose two that are expressed at similar levels in growing and encysting parasites ( gla_8711 and gla_16928 ) , one that is downregulated during encystation ( gla_1908 ) , and one that is upregulated in encysting parasites ( gla_2116 ) . To further analyze these DHHC proteins , we expressed full-length gla_1908 , gla_2116 , gla_16928 and gla_8711 as fusion DHHC proteins containing C-terminal HA-tag [39] and evaluated their protein expression profiles by Western blotting using an anti-HA mAb ( Figure 7 ) . Analysis by semi-quantitative RT-PCR indicated that the overexpression of these fusion proteins was 2 to 3-times higher in transgenic cells , as reported for protein expression using a similar vector [9] . Immunofluorescence assays showed that HA-tagged gla_1908 , gla_2116 , and gla_16928 partially co-localized with BiP in the endoplasmic reticulum ( ER ) or around the nuclei of transgenic trophozoites ( Figure 8 , trophozoite ) . Our results confirmed the localization of gla_16928 already shown by Touz et al . [31] . Analysis of intracellular localization of yeast and mammalian DHHC proteins revealed that the majority of these localize to the ER and Golgi [20] , [87] . However , there are a few exceptions , including human DHHC5 protein [87] and Giardia DHHC protein ( EAA36893 ) [31] , which localize to the plasma membrane . Also , we found that gla_8711 partially co-localized with the adaptor protein AP-2 [57] at the lysosomal-like peripheral vacuoles ( PVs ) as well as in plasma membrane and flagella ( Figure 8 , trophozoite ) . Ongoing experiments intended to knock-down this protein may reveal its importance during the Giardia life cycle . The hallmark of encystation in Giardia is the synthesis of CWP1 , CWP2 , and CWP3 [88] . These proteins are expressed and concentrated within the ESVs before they are targeted to the cyst wall [89] , [6] , [90] . To address the influence of the overexpression of these HA-tagged DHHC proteins during encystation , dhhc-ha transgenic trophozoites were induced to encyst in vitro . The localization of DHHC-HA proteins as well as CWP1 expression , intracellular localization , and vesicle formation were addressed by IFA . To examine in detail the results obtained , we decided to analyze each dhhc-ha transgenic cell following the protocol described above , in which the cells were classified as encysting I , encysting II , and early cyst . We observed that gla_1908 ( Figure 8A ) , gla_2116 ( Figure 8B ) , and gla_8711 ( Figure 8D ) transgenic parasites displayed normal encystation . It was noteworthy that gla_16928 ( Figure 8C ) had enlarged ESVs , with co-localization between gla_16928-HA and CWP1 observed in those vesicles ( Figure 8C , inset ) . Additionally , it was noted that gla_16928 early cysts had a larger size and an abnormal shape compared with wild-type cells ( not shown ) and other transgenic early cysts . When CWP expression was analyzed in dhhc transgenic parasites by qRT-PCR , we observed that , except for gla_2116 transgenic cells , which displayed similar levels or even moderate decrease in the mRNA expression of CWPs compared to the control , the other dhhc-ha transgenic parasites showed increased expression of CWP1 , CWP2 , and CWP3 ( Figure 9A ) . Several transcription factors have been described as involved in the regulation of cwp gene transcription [91] , [92] , [93] , [94] , [95] , [96] , [97] . However , the mechanisms underlying transcription control in this parasite have not been completely elucidated . It has always been assumed that the mobilization mechanism for transcription factors in many organisms is based on proteolytic processing [98] , [99] , [100] , [101] . Nevertheless , there is a group of lipid-modified transcription factors whose mobilization mechanism to the nucleus is not based on proteolytic processing but on reversible palmitoylation [102] . If that were the case for the transcription factors involved in Giardia encystation , DHHC proteins would be palmitoylating different transcriptions factors that , in turn , may regulate CWP expression . It would be interesting to explore the molecular architecture of Giardia transcription factors to find out whether palmitoylation is involved in regulating their shuttling between the cytoplasm and the nuclei . Analyzing the amount of water-resistant cysts , we observed that gla_1908 and gla_8711 transgenic cells yielded a significantly higher amount of cysts than the control ( Figure 9B ) . In contrast , gla_2116 transgenic cells , while displaying an apparently normal encystation process ( Figure 8B ) and CWP expression ( Figure 9A ) , produced a reduced number of mature cysts ( Figure 9B ) . A likely explanation is that gla_2116 may be involved in the palmitoylation of a protein in charge of turning encystation-specific genes off and ending the encystation process . In the case of gla_16928 transgenic parasites , these cells produced a low percentage of cysts ( Figure 9B ) although the CWP expression was increased ( Figure 9A ) . These findings , in addition to the large ESVs seen in figure 8C ( encysting II ) and the large size of early cysts ( Figure 8C , early cyst ) , may be explained by a high rate of synthesis of CWPs in gla_16928 transgenic parasites , which may exceed the mechanisms of vesicle discharge regulation , leading to the formation of immature non-water-resistant cysts . Further experiments using knock-down strategies are needed to completely address the role of each DHHC protein in the encystation process . Table 2 summarizes the main features of the Giardia DHHC proteins analyzed in this work . The different localization of DHHC-HA proteins in trophozoites and the differential effect of DHHC overexpression in encystation prompted us to evaluate the palmitoylation pattern in the dhhc transgenic parasites ( Figure 10 ) . gla_1908 , gla_2116 , gla_16928 , and gla_8711 transgenic trophozoites or encysting parasites displayed a similar global protein palmitoylation pattern compared to wild type ( Figure 1A ) . Mass spectrometry-based proteomics analyses will be necessary to accurately identify any differences in the palmitoylation substrates among the dhhc transgenic parasites . This work presents a detailed analysis of Giardia lamblia DHHC protein structure and phylogeny and reveals a possible role of palmitoylation in Giardia encystation . Our data , suggesting the presence of DHHC proteins in growing and encysting parasites , reinforced the idea that this PTM has conserved and important functions in cell-signaling , protein-sorting and protein-export throughout evolution . Without being able to assign a specific substrate candidate to each Giardia DHHC proteins , we showed that overexpression of these enzymes had consequences on CWP expression and on the amount of cysts produced . Proteomic analysis of Giardia palmitoyl proteome would be a great contribution to elucidating the mechanisms by which palmitoylation participates in encystation biology . Finally , the suggested role of palmitoylation in Giardia encystation , a key event that enables the parasite to survive in the environment , infect a new host and evade the immune response [1] , [103] , could open new ways to intervene in the process of Giardia infection . | Giardiasis is a major cause of non-viral/non-bacterial diarrheal disease worldwide and has been included within the WHO Neglected Disease Initiative since 2004 . Infection begins with the ingestion of Giardia lamblia in cyst form , which , after exposure to gastric acid in the host stomach and proteases in the duodenum , gives rise to trophozoites . The inverse process is called encystation and begins when the trophozoites migrate to the lower part of the small intestine where they receive signals that trigger synthesis of the components of the cyst wall . The cyst form enables the parasite to survive in the environment , infect a new host and evade the immune response . In this work , we explored the role of protein S-palmitoylation , a unique reversible post-translational modification , during Giardia encystation , because de novo generation of endomembrane compartments , protein sorting and vesicle fusion occur in this process . Our findings may contribute to the design of therapeutic agents against this important human pathogen . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"Discussion"
] | [
"biology",
"and",
"life",
"sciences",
"parasitology",
"microbiology"
] | 2014 | Identification of Giardia lamblia DHHC Proteins and the Role of Protein S-palmitoylation in the Encystation Process |
Since its emergence in 2007 in Micronesia and Polynesia , the arthropod-borne flavivirus Zika virus ( ZIKV ) has spread in the Americas and the Caribbean , following first detection in Brazil in May 2015 . The risk of ZIKV emergence in Europe increases as imported cases are repeatedly reported . Together with chikungunya virus ( CHIKV ) and dengue virus ( DENV ) , ZIKV is transmitted by Aedes mosquitoes . Any countries where these mosquitoes are present could be potential sites for future ZIKV outbreak . We assessed the vector competence of European Aedes mosquitoes ( Aedes aegypti and Aedes albopictus ) for the currently circulating Asian genotype of ZIKV . Two populations of Ae . aegypti from the island of Madeira ( Funchal and Paul do Mar ) and two populations of Ae . albopictus from France ( Nice and Bar-sur-Loup ) were challenged with an Asian genotype of ZIKV isolated from a patient in April 2014 in New Caledonia . Fully engorged mosquitoes were then maintained in insectary conditions ( 28°±1°C , 16h:8h light:dark cycle and 80% humidity ) . 16–24 mosquitoes from each population were examined at 3 , 6 , 9 and 14 days post-infection to estimate the infection rate , disseminated infection rate and transmission efficiency . Based on these experimental infections , we demonstrated that Ae . albopictus from France were not very susceptible to ZIKV . In combination with the restricted distribution of European Ae . albopictus , our results on vector competence corroborate the low risk for ZIKV to expand into most parts of Europe with the possible exception of the warmest regions bordering the Mediterranean coastline .
Zika virus ( ZIKV ) ( genus Flavivirus , family Flaviviridae ) is an emerging arthropod-borne virus transmitted to humans by Aedes mosquitoes . ZIKV infection in humans was first observed in Africa in 1952 [1] , and can cause a broad range of clinical symptoms presenting as a “dengue-like” syndrome: headache , rash , fever , and arthralgia . In 2007 , an outbreak of ZIKV on Yap Island resulted in 73% of the total population becoming infected [2] . Following this , ZIKV continued to spread rapidly with outbreaks in French Polynesia in October 2013 [3] , New Caledonia in 2015 [4] , and subsequently , Brazil in May 2015 [5 , 6] . During this expansion period , the primary transmission vector is considered to have been Aedes aegypti , although Aedes albopictus could potentially serve as a secondary transmission vector [7] as ZIKV detection has been reported in field-collected Ae . albopictus in Central Africa [8] . As Musso et al . [9] observed , the pattern of ZIKV emergence from Africa , throughout Asia , to its subsequent arrival in South America and the Caribbean closely resembles the emergence of Chikungunya virus ( CHIKV ) . In Europe , returning ZIKV-viremic travelers may become a source of local transmission in the presence of Aedes mosquitoes , Ae . albopictus in Continental Europe and Ae . aegypti in the Portuguese island of Madeira . Ae . albopictus originated from Asia was recorded for the first time in Europe in Albania in 1979 [10] , then in Italy in 1990 [11] . It is now present in all European countries around the Mediterranean Sea [12] . This mosquito was implicated as a vector of CHIKV and DENV in Europe [13] . On the other hand , Ae . aegypti disappeared after the 1950s with the improvement of hygiene and anti-malaria vector control . This mosquito reinvaded European territory , Madeira island , in 2005 [14] , and around the Black Sea in southern Russia , Abkhazia , and Georgia in 2004 [12] . The species was responsible for outbreaks of yellow fever in Italy in 1804 [15] and dengue in Greece in 1927–1928 [16] . To assess the possible risk of ZIKV transmission in Europe , we compared the relative vector competence of European Ae . aegypti and Ae . albopictus populations to the Asian genotype of ZIKV .
The Institut Pasteur animal facility has received accreditation from the French Ministry of Agriculture to perform experiments on live animals in compliance with the French and European regulations on care and protection of laboratory animals . This study was approved by the Institutional Animal Care and Use Committee ( IACUC ) at the Institut Pasteur . No specific permits were required for the described field studies in locations that are not protected in any way and did not involve endangered or protected species . Four populations of mosquitoes ( two populations of Ae . aegypti: Funchal ( 32°40’N , 16°55’W ) and Paul do Mar ( 32°45’N , 17°13’W ) , collected on island of Madeira and two populations of Ae . albopictus: Nice ( 43°42’N , 7°15’E ) and Bar-sur-Loup ( 43°42’N , 6°59’E ) in France ) were collected using ovitraps . Eggs were immersed in dechlorinated tap water for hatching . Larvae were distributed in pans of 150–200 individuals and supplied with 1 yeast tablet dissolved in 1L of water every 48 hours . All immature stages were maintained at 28°C ± 1°C . After emergence , adults were given free access to a 10% sucrose solution and maintained at 28°C ± 1°C with 70% relative humidity and a 16:8 light/dark cycle . The F1 generation of Ae . aegypti from Madeira and F7-8 generation of Ae . albopictus from France were used for experimental infections . The ZIKV strain ( NC-2014-5132 ) originally isolated from a patient in April 2014 in New Caledonia was used to infect mosquitoes . The viral stock used was subcultured five times on Vero cells prior to the infectious blood-meal . The NC-2014-5132 strain is phylogenetically closely related to the ZIKV strains circulating in the South Pacific region , Brazil [5] and French Guiana [17] . Infectious blood-meals were provided using a titer of 107 TCID50/mL . Seven-day old mosquitoes were fed on blood-meals containing two parts washed rabbit erythrocytes to one part viral suspension supplemented with ATP at a final concentration of 5 mM . Rabbit arterial blood was collected and erythrocytes were washed five times with Phosphate buffered saline ( PBS ) 24 h before the infectious blood-meal . Engorged females were transferred to cardboard containers with free access to 10% sucrose solution and maintained at 28°C and 70% relative humidity with a 16:8 light/dark cycle . 16–24 female mosquitoes from each population were analyzed at 3 , 6 , 9 , and 14 days post-infection ( dpi ) to estimate the infection rate , disseminated infection rate and transmission efficiency . Briefly , legs and wings were removed from each mosquito followed by insertion of the proboscis into a 20 μL tip containing 5 μL FBS for 20 minutes . The saliva-containing FBS was expelled into 45 μμL serum free L-15 media ( Gibco ) , and stored at -80°C . Following salivation , mosquitoes were decapitated and head and body ( thorax and abdomen ) were homogenized separately in 300 μL L-15 media supplemented with 3% FBS using a Precellys homogenizer ( Bertin Technologies ) then stored at -80°C . Infection rate was measured as the percentage of mosquitoes with infected bodies among the total number of analyzed mosquitoes . Disseminated infection rate was estimated as the percentage of mosquitoes with infected heads ( i . e . , the virus had successfully crossed the midgut barrier to reach the mosquito hemocoel ) among the total number of mosquitoes with infected bodies . Transmission efficiency was calculated as the overall proportion of females with infectious saliva among the total number of tested mosquitoes . Samples were titrated by plaque assay in Vero cells . For head/body homogenates and saliva samples , Vero E6 cell monolayers were inoculated with serial 10-fold dilutions of virus-containing samples and incubated for 1 hour at 37°C followed by an overlay consisting of DMEM 2X , 2% FBS , antibiotics and 1% agarose . At 7 dpi , overlay was removed and cells were fixed with crystal violet ( 0 . 2% Crystal Violet , 10% Formaldehyde , 20% ethanol ) and positive/negative screening was performed for cytopathic effect ( body and head homogenates ) or plaques were enumerated ( head and saliva samples ) . Vero E6 cells ( ATCC CRL-1586 ) were maintained in DMEM ( Gibco ) supplemented with 10% fetal bovine serum ( Eurobio ) , Penicillin and Streptomycin , and 0 . 29 mg/mL l-glutamine . All statistical tests were conducted with the STATA software ( StataCorp LP , Texas , USA ) using 1-sided Fisher’s exact test and P-values>0·05 were considered non-significant .
To test whether Ae . aegypti from a European territory were able to transmit ZIKV , we analyzed the vector competence of two Ae . aegypti populations collected on the island of Madeira based on three parameters: viral infection of the mosquito midgut , viral dissemination to secondary organs , and transmission potential , analyzed at 3 , 6 , 9 , and 14 dpi . Only mosquitoes presenting an infection ( i . e . infected midgut ) were analyzed for viral dissemination . The two populations presented similar infection ( P = 0 . 50 ( 3 dpi ) , 0 . 17 ( 6 ) , 0 . 36 ( 9 ) , 0 . 50 ( 14 ) ; Fig 1 ) and disseminated infection ( P = 0 . 59 ( 3 dpi ) , 0 . 63 ( 6 ) , 0 . 43 ( 9 ) , 0 . 06 ( 14 ) ; Fig 1 ) with the highest rates measured at 9 dpi and 9–14 dpi , respectively . When examining transmission efficiency , only Ae . aegypti Funchal were able to transmit ZIKV at 9 ( 1 individual among 20 tested ) and 14 dpi ( 1 among 20 ) ( Fig 1 ) . When considering the number of viral particles in heads , no significant difference was detected between Ae . aegypti Funchal and Ae . aegypti Paul do Mar ( P = 1 ( 3 dpi ) , 0 . 22 ( 6 ) , 0 . 60 ( 9 ) , 0 . 38 ( 14 ) ; Fig 2 ) . When examining viral loads in saliva , only Ae . aegypti Funchal exhibited 1550 particles at 9 dpi and 50 at 14 dpi ( Fig 2 ) . To determine if Ae . albopictus present in continental Europe were able to sustain local transmission of ZIKV as previously observed with CHIKV and DENV , we evaluated the vector competence of two Ae . albopictus populations collected in Nice and Bar-sur-Loup in the South of France . When compared with Ae . aegypti , the two Ae . albopictus populations showed similar infection rates at 3 dpi ( P = 0 . 08 ) and 6 dpi ( P = 0 . 11 ) and disseminated infection rates at 9 dpi ( P = 0 . 62 ) and 14 dpi ( P = 0 . 10 ) ( Fig 1 ) . Only one individual among 24 Ae . albopictus Bar-sur-Loup tested at 14 dpi was able to transmit ZIKV ( Fig 1 ) . When analyzing the number of viral particles in heads , only few mosquitoes were infected ( Fig 2 ) . When examining saliva , one Ae . albopictus Bar-sur-Loup exhibited 2 viral particles at 14 dpi ( Fig 2 ) . In summary , ZIKV dissemination through Ae . aegypti was noticeably superior and the virus in saliva was detected earlier in Ae . aegypti than in Ae . albopictus . However both mosquito species showed similar transmission efficiencies at 9–14 dpi .
ZIKV could be transmitted , spread and maintained in Europe either via ( i ) Madeira where the main vector Ae . aegypti has been established since 2005 or ( ii ) Continental Europe where Ae . albopictus is known to have been present since 1979 [12] . We demonstrated that ZIKV was amplified and expectorated efficiently in saliva by European Ae . aegypti from Madeira . This contrasts with the lower vector competence for ZIKV of French Ae . albopictus . Taking these observations and the overall average lower temperatures of most regions of Europe into account , the risk of major outbreaks of Zika fever in most areas of Europe , at least for the immediate future , appears to be relatively low . Our results highlight the potential risk for ZIKV transmission on Madeira where two main factors are present: the presence of the main vector , Ae . aegypti introduced in 2005 [18] and imported cases from Brazil with which Madeira , an autonomous region of Portugal , maintains active exchanges of goods and people sharing the same language . Thus Madeira Island could be considered as a stepping stone for an introduction of ZIKV into Europe . Autochthonous cases of CHIKV and DENV have been reported in Europe since 2007: CHIKV in Italy in 2007 , South France in 2010 , 2014 , and DENV in South France in 2010 , 2013 , 2015 , and Croatia in 2010 [19] . The invasive species Ae . albopictus first detected in Europe in 1979 [10] has played a central role in this transmission [19] . Thus , there might be a risk of a similar establishment of ZIKV in Europe upon the return of viremic travelers [20 , 21] . We showed that Ae . albopictus from South France were less competent for ZIKV infection requiring 14 days to be expectorated in the mosquito saliva after infection . Therefore , we can suggest that the Asian tiger mosquito from Southern France and more widely , Europe , are less suitable to sustain local transmission of ZIKV compared to CHIKV and perhaps , DENV . Ae . albopictus Nice were not able to expectorate ZIKV in saliva at day 14 post-infection like Ae . albopictus Bar-sur-Loup suggesting two populations genetically differentiated . Considering the extensive airline travel between Latin America and Europe , the risk for local transmission of ZIKV in the European area where the mosquito Ae . albopictus is widely distributed , is assumed to be minimal based on our studies of vector competence . Nevertheless , reinforcement of surveillance and control of mosquitoes should remain a strong priority in Europe since Aedes mosquitoes also transmit DENV and CHIKV and virus adaptation to new vectors cannot be excluded , as previously observed with CHIKV in La Reunion [22 , 23] . | In May 2015 , local transmission of Zika virus ( ZIKV ) was reported in Brazil and since then , more than 1 . 5 million human cases have been reported in Latin America and the Caribbean . This arbovirus , primarily found in Africa and Asia , is mainly transmitted by Aedes mosquitoes , Aedes aegypti and Aedes albopictus . Viremic travelers returning from America to European countries where Ae . albopictus is established could become the source for local transmission of ZIKV . In order to estimate the risk of seeding ZIKV into local mosquito populations , the susceptibility of European Ae . aegypti and Ae . albopictus to ZIKV was measured using experimental infections . We demonstrated that Ae . albopictus and Ae . aegypti from Europe were not very susceptible to ZIKV . The threat for a Zika outbreak in Europe should be limited . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"chikungunya",
"infection",
"pathogens",
"geographical",
"locations",
"microbiology",
"tropical",
"diseases",
"saliva",
"animals",
"alphaviruses",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"neglected",
"tropical",
"diseases",
"insect",
"vectors",
"infectious",
"diseases",
"south",
"america",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"brazil",
"arthropoda",
"people",
"and",
"places",
"france",
"mosquitoes",
"anatomy",
"flaviviruses",
"viral",
"pathogens",
"physiology",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"europe",
"organisms",
"zika",
"virus"
] | 2016 | Zika Virus, a New Threat for Europe? |
Iran , despite its size , geographic location and past cultural influence , has largely been a blind spot for human population genetic studies . With only sparse genetic information on the Iranian population available , we pursued its genome-wide and geographic characterization based on 1021 samples from eleven ethnic groups . We show that Iranians , while close to neighboring populations , present distinct genetic variation consistent with long-standing genetic continuity , harbor high heterogeneity and different levels of consanguinity , fall apart into a cluster of similar groups and several admixed ones and have experienced numerous language adoption events in the past . Our findings render Iran an important source for human genetic variation in Western and Central Asia , will guide adequate study sampling and assist the interpretation of putative disease-implicated genetic variation . Given Iran’s internal genetic heterogeneity , future studies will have to consider ethnic affiliations and possible admixture .
The highlands of Iran have been at the crossroads of human migrations [1–6] since the dispersal of modern humans out of Africa due to their geostrategic position . While exercising a strong cultural influence on neighboring regions , Iran has also repeatedly received migratory influx in the past millennia . Among others , this includes the successive southward migration of groups of Indo-European ( IE ) language speakers ( e . g . Scythians , Medes and Persians ) [7] , the Arab arrival in the 7th century CE and the later influx of Turkic-speaking people from Central Asia . As a result of migrations , internal splits , admixture and other movements , today’s Iranian population comprises numerous ethnic , religious and linguistic groups ( S1 Appendix , S1 Fig ) , prominently including Persians ( 65% in 2008 [8] ) , Iranian Azeris ( 16% ) , Iranian Kurds ( 7% ) , Iranian Lurs ( 6% ) , Iranian Arabs ( 2% ) , Iranian Baluchis ( 2% ) , Iranian Turkmen ( 1% ) , Qashqai and other Turkish-language tribal groups ( 1% ) as well as Armenians , Assyrians , Georgians , Jews , Zoroastrians ( all <1% ) and others , although definitions [9] and reported proportions vary between sources ( e . g . [10–12] ) . Speakers of an Iranian , i . e . Indo-European , language or language dialect ( e . g . Persian , Kurdish , Luri , Baluchi ) by far outnumber speakers of either a Turkic or Semitic language . With Iran being located within a belt of countries where consanguineous marriages are widely practiced , Iranian samples have featured prominently in disease-related studies , facilitating the identification of genes involved in rare autosomal recessive diseases by linkage analysis and autozygosity mapping and contributing to a deeper etiological understanding also of complex disorders [13 , 14] . Examples demonstrating the value of these populations for human genetic research are ample ( e . g . [15–21] for Iran alone ) , likely moving from the study of few families to population-based studies in the future [22–25] . Still , consanguinity levels are not homogenous across the Iranian population . Early studies , based not on actual genetic data but on familial relation assessment , found these levels to vary between geographic regions and between ethnic groups [26 , 27] . A recent study , also based on familial relation assessment , refined these results and reported differences in consanguinity by province , area of residence , birth and marriage cohort as well as with educational level [28] . Patterns of runs of homozygosity ( ROHs ) or haplotype sharing by descent ( HBD ) can be indicative of autozygosity , but vary between populations and across genomic locations [29–34] , as do the frequency of consanguinity and the moderately correlated degree of genomic inbreeding [35] . Furthermore , autozygosity mapping is predominantly able to detect comparatively recent , local founder mutations [13] . Moreover , carrier frequencies of disease-predisposing variants have been reported to strongly differ between geographic regions in Iran , e . g . for mutations in the GJB2 gene [36] and for β-thalassemia [37] , with different ethnic affiliations being the likely cause and possibly helping to determine the pathogenicity of those variants [38] . Finally , studies on copy-number variation ( CNV ) in the Iranian population ( e . g . [39] ) were so far disease-specific but not with respect to the general , healthy population . Perhaps somewhat surprisingly , Central Asia and parts of Western Asia have largely been a blind spot for non-medical genetic studies in the past decades . Until recently , dedicated genetic projects of extant human populations with a global or continental focus ( e . g . [33 , 40–51] only sporadically included samples , if any , from Iran and did not comprehensively cover the area . Of note , studies that did include Iranian samples frequently treated them as coming from or being representative of a single homogeneous population . Studies on sporadic ancient DNA ( aDNA ) samples from the Early Neolithic up to the Chalcolithic in Iran showed the existence of highly genetically differentiated populations that were not ancestral to Europeans but , in the case of specimen from the Zagros Mountains , exhibited some affinity to Zoroastrians [1 , 2 , 6 , 52] . An early study on ABO blood groups found extreme differences between some of 21 considered ethnic groups in Iran [53] , whereas another study , published a year later and additionally based on serum proteins and cell enzymes , presented evidence for population substructure between the six included groups ( Iranian Turks , Kurds , Lurs , Zabolis , Baluchis and Zoroastrians ) with an average FST value of 0 . 02 , based on blood groups , serum proteins and cell enzymes , and some degree of inbreeding [54] . More regionally focused studies on Iran , based on uniparental markers such as Y-chromosomal haplogroups and short tandem repeat ( STR ) marker haplotypes as well as mitochondrial ( mtDNA ) haplogroups , confirmed high degrees of genetic diversity in the Iranian population [3–5 , 55–61] . These studies reported the respective variation to be predominantly of Western Eurasian origin , with only limited contributions from eastern Eurasia , South Asia and Africa most pronounced in the southern Iranian provinces . These studies also reported ancient and recent gene flow between Iran and the Arabian Peninsula , a surprisingly close relationship between Persians and Iranian Turkic-speaking Qashqai and generally high levels of variation comparable to those in the South Caucasus , Anatolia and Europe . These observations all support the notion of Iran forming a crossroads of human migrations . Notably , a study on Armenians , located to the North of Iran , also suggested multiple admixture events and a general role as bridge between different geographic regions [49] . Using genome- or exome-wide genotype data , a number of studies have analyzed samples of populations that can be considered proxies for ethnic groups in Iran from surrounding countries . In a study of 156 individuals , the population of Qatar was reported to comprise three distinct groups , with one ( “Q2” ) showing strong affinity to Persians and patterns of admixture [3 , 62 , 63] . A study on 22 Kuwaitis with Persian ancestry found comparatively high levels of genetic diversity for a non-African population , explicable by past admixture events [34] . A study of 43 individuals belonging to the Parsis , a Zoroastrian religious community in India and Pakistan , demonstrated a closer genetic affinity to today’s Iranian and Caucasus populations than to South Asian populations , but , quite remarkably , an even stronger similarity to Neolithic aDNA samples from Iran compared to modern Iranians , consistent both with the historic record of a southward migration induced by the 7th century’s Arab entry to Iran and more recent admixture events with the modern Iranian population [64] . Findings of increased homogeneity and the dating of past admixture events in further samples of Iranian and Indian Zoroastrians [65] complemented these results . Analysis of 24 individuals from the Indo-European speaking Kalash , a population isolate at the Hindu Kush , Afghanistan , indicated a genetically drifted ancient northern Eurasian population that split during the very early Neolithic and subsequently migrated southwards [66] . Finally , a recent study restricted to exome data merged 87 Iranian with 136 Pakistani samples and demonstrated a somewhat extreme or isolated position when compared to other populations from the Maghreb and from the Arabian Peninsula through Turkey [33] . Still , none of these studies has directly and comparatively studied ethnic groups in Iran . Correlation between genetic and linguistic proximity of populations has frequently been assumed to be the rule , while language adoption is usually considered as an exception to the rule of co-evolution ( e . g . [67–69] ) , although such claims have repeatedly been disputed ( e . g . [70] ) . Evidence for such correlation is ample in Europe , including autosomal and mitochondrial data [71–77] , Y-chromosomal data [77–80] and even , with respect to the spread of Indo-European languages into Europe , ancient DNA data [81] . In-depth studies on other parts of the world found some correlation of language dispersal with Y-chromosomal lineages [82–87] , although not in all parts [88] . Furthermore , some instances of male-mediated gene flow over major linguistic barriers have been inferred as well [89 , 90] . An early study already observed close genetic relationship between Semitic-speaking and Indo-European-speaking groups in Iran [58] . Studies on neighboring Armenia found evidence for a language replacement [91] event , possibly facilitated by the mixing of multiple source populations during the Bronze Age [49] . However , the relationship between genetic and linguistic proximity has been rarely investigated for Iran and neighboring countries . While Iran appears to be destined to make further important contributions to human genetic research , an adequate design and interpretation of future medical and population genetic studies is mandatory to arrive at interpretable findings . Here , we comprehensively analyzed the genome-wide diversity of eleven ethnic groups in Iran , their relation to each other as well as with global and local reference populations . Furthermore , we investigated , stratified by ethnicity , levels of consanguinity , the distribution of homozygous and copy-number regions and CNVs as well as the extent of population stratification within Iran and the possible effects in association studies if not accounted for properly and the relationship between spoken language family and genetic proximity .
The 11 included Iranian ethnic groups featured distinct and substantial genetic heterogeneity ( Fig 1A ) . Seven groups ( Iranian Arabs , Azeris , Gilaks , Kurds , Mazanderanis , Lurs and Persians ) strongly overlapped in their overall autosomal diversity in an MDS analysis ( Fig 1B ) , suggesting the existence of a Central Iranian Cluster ( CIC ) , notably also including Iranian Arabs and Azeris . The other four groups ( Iranian Baluchis , Persian Gulf ( PG ) Islanders , Sistanis and Turkmen ) presented as strongly admixed populations with contributions by different ancestral populations but always with an orientation towards the CIC , being strikingly different from the CIC and from each other , except for Baluchis and Sistanis who partially overlapped ( Fig 1A ) . On a global scale ( Fig 2 including “Old World” populations only; see S2 Fig for all 1000G populations ) , CIC Iranians closely clustered with Europeans , while Iranian Turkmen showed similar yet distinct degrees of admixture compared to other South Asians . The degree was less pronounced for Baluchis , Sistanis and PG Islanders , with the latter showing a pointed orientation towards Sub-Saharan Africans and a co-localization with numerous Latin American samples . Notably , Iranian Arabs now showed some detachment from the CIC towards Sub-Saharan populations . A local comparison corroborated the distinct genetic diversity of CIC Iranians relative to other geographically close populations [2 , 6 , 44] ( Fig 3 and S3 Fig ) . Strikingly , the relative genetic location of the Iranian ethnic groups mirrored their geographic location at the nexus between South and Central Asia and West Asia , Northern Africa and the Caucasus . Iranian Baluchis and Sistanis clustered with or nearby Pakistani and other South Asian populations , whereas Iranian Turkmen located next or atop Central Asian populations , respectively . Iranian Arabs appeared distinct from other Arab populations in West Asia and Northern Africa . Furthermore , Zoroastrian samples [6] located as essential CIC members . These results were closely mirrored by the pairwise fixation index ( FST ) values ( Table 2 and S5 Table ) . CIC groups showed little differentiation ( FST~0 . 0008–0 . 0033 ) , whereas non-CIC groups consistently yielded much larger values , most extreme for PG Islanders vs Iranian Turkmen ( FST = 0 . 0110 ) . Still , genetic substructure was much smaller among Iranian groups than in relation to any of the 1000G populations , supporting the view that the CIC groups form a distinct genetic entity , despite internal heterogeneity . European ( FST~0 . 0105–0 . 0294 ) , South Asians ( FST~0 . 0141–0 . 0338 ) , but also some Latin American populations ( Puerto Ricans: FST~0 . 0153–0 . 0228; Colombians: FST~0 . 0170–0 . 0261 ) were closest to Iranians , whereas Sub-Saharan Africans and admixed Afro-Americans ( FST~0 . 0764–0 . 1424 ) as well as East Asians ( FST ~ 0 . 0645–0 . 1055 ) showed large degrees of differentiation with Iranians . If not corrected for , the observed degree of population substructure could severely confound population-based genetic association studies in Iran . In the extreme scenario of cases being sampled exclusively from one ethnic group and controls from another , CIC groups would yield moderate , although still problematic , genomic inflation factor ( GIF ) values ( 1 . 17–1 . 61 ) , whereas non-CIC groups may yield values up to 3 . 0 ( Table 2 ) . We further explored the genetic composition and origin of the Iranian ethnic groups . ADMIXTURE [111] analyses corroborated the existence of the postulated CIC and pointed to the existence of a distinct Iranian ancestral component . In the analysis of the 11 Iranian groups alone ( best-fit model for k = 4 ) , all seven CIC groups featured a single predominant ancestry and slightly varying proportions for the other three ancestral groups , whereas the other four varied in their degree of admixture with different ancestral populations ( Fig 4A ) . Even more strikingly , the global data set analysis ( best-fit k = 13 ) yielded three ancestral populations that substantially and almost exclusively contributed to the 11 Iranian groups but were barely seen in the 1000G populations , with one ancestral population shared across all 11 groups ( colored blue in Fig 4B ) and another one shared by all groups except for PG Islanders which featured a different dominant ancestral population ( colored light-green and light-blue in Fig 4B , respectively ) . A notable exception was the Tuscans ( TSI ) , sharing a substantial proportion of ancestry with Iranians , in particular those from the CIC . A regional comparison corroborated the unique composition of the Iranian ethnic groups ( Fig 4C ) , with Zoroastrian and other Iranian samples showing a concordant picture . Random down-sampling of our Iranian data set to sizes similar those of the reference groups confirmed that this result was not due to our comparatively large sample sizes ( S4 Fig ) . Explicit modeling of 0–15 migration events using TreeMix [112] evidenced the robustness of the close clustering of all Iranian groups , with Europeans always closest to Iranians ( S5–S10 Figs ) . An influx of ancestors from Asian populations to both Turkmen and Finns was consistently inferred , while Iranian Arabs apparently received some African influx . Modelling Iranians as resulting from admixture between pairs of 1000G populations resulted in positive f3 statistics [113] throughout , thus supporting the primarily autochthonous origin of the CIC groups , except for non-CIC Turkmen that consistently showed negative f3 values ( median -0 . 0083; range -0 . 0023 –-0 . 0096 ) for any pair of an European and an East Asian population ( S6 Table ) , yielding the strongest evidence for Tuscans admixing Han Chinese or Japanese ( f3 = -0 . 0093 –-0 . 0096; Z = -29 , 2370 –-30 , 1030 ) . Modelling non-CIC groups as resulting from admixture between a CIC group and a 1000G population yielded a more nuanced picture ( S7 Table ) . While Sistanis consistently appeared to be admixed between CIC and South Asian groups and , less pronouncedly , with Southern Han Chinese , Turkmen revealed components from CIC , African , European , East Asian and , less pronounced , South Asian groups . PG Islanders and also Baluchis comprised a limited African component but no apparent influx from other groups besides the CIC . When relating our extant Iranian samples with published ancient DNA ( aDNA ) samples of different time strata from Iran and beyond to trace temporal-spatial movements of human populations , we did not find indications for substantial migrations into the CIC groups except for Caucasus populations during Neolithic through Bronze Age times ( Figs 5–7 ) , with the latter presenting either as a source or as a refuge , i . e . a migration target . In particular , contributions by Steppe people were apparently very limited and restricted to the Bronze Age or briefly before ( Fig 6 ) . Overall , the CIC groups appeared to have experienced a largely autochthonous development over at least the past 5 , 000 years . Remarkably , Early Neolithic Iranian samples [6 , 107] from Western Iran and Tappeh Hesar co-localized with the more remotely located extant PG Islanders ( Fig 5 ) , whereas later Bronze Age samples from Tappeh Hesar showed a trend towards the CIC ( Fig 6 ) , possibly indicating ongoing admixture between these groups . Of note , Central Asian aDNA samples from the Neolithic and the Bronze Age also co-localized with PG Islanders and showed a similar trend ( Figs 5 and 6 ) . Sistani samples most distant from the CIC clustered close to Iron Age Pakistani samples ( Fig 7 ) and may have undergone a similar admixture with CIC groups , however , a lack of samples from the past millennia renders this an open question . Languages spoken by the 11 Iranian ethnic groups fell into three different families , namely Afro-Asiatic ( Semitic; Arabs ) , Altaic ( Turkic; Turkmen , Azeris ) and Indo-European ( IE; all others ) . This linguistic diversity was only partially mirrored by genetic proximity , with Turkic-speaking Iranian Azeris and Semitic-speaking Iranian Arabs closely genetically resembling IE speakers from the CIC , whereas IE-speaking Baluchis , PG Islanders and Sistanis appeared genetically detached from the other IE-speaking groups . After re-classifying our local data set with respect to language family ( S2 Table ) , a general trend of closer genetic proximity , as assessed by a principal-components analysis , for speakers of a language from the same family became obvious ( S11A Fig ) . However , IE speakers fell apart into broadly two distinct groups ( corresponding to the European and Indo-Iranian subbranches ) , while Altaic language speakers comprised widely spread genetic diversity . An approximate autocorrelation analysis based on genetic distance in the first two principal components confirmed a strong localized positive correlation between genetic proximity and spoken language family ( S11B Fig ) . Iran’s ethnic groups strongly differed in their levels of consanguinity . Iranian Arabs , Baluchis and Sistanis showed very high inbreeding coefficient values ( FI ~ 0 . 0122–0 . 0132 ) , exceeding those of the most consanguineous 1000G population ( STU ) . Iranian Gilaks ( FI = 0 . 0001 ) and Kurds ( FI = 0 . 0010 ) showed almost no consanguinity , whereas the other groups showed considerably elevated consanguinity ( FI ~ 0 . 0024–0 . 069 ) in comparison to the 1000G populations ( S12A Fig and Table 3 ) . Of note , consanguinity varied widely within each group , with 50% of individuals showing FI values below 0 . 0051 ( Iranian Arabs ) , 0 . 0042 ( Iranian Sistanis ) and 0 . 0036 ( Iranian Baluchis ) , respectively , and virtually equal to zero in the remaining groups . Cumulative lengths of IBDseq-inferred autozygous regions and of PLINK-defined runs of homozygosity ( ROHs ) closely mirrored the distribution of inbreeding values ( S12B and S12C Fig ) . Likelihood-based ROH definition and subsequent length classification by GARLIC ( S12D–S12F Fig ) revealed substantial amounts of ancestral class-A cumulative ROH length in virtually all Iranian ethnic groups and 1000G populations but also generally much shorter recent class-C cumulative ROH length . Iranian Arabs , Baluchis and Sistanis most prominently deviated from this trend , while most other Iranian groups showed still elevated values , indicating ongoing consanguinity through the past millennia . Akin to previously studied populations , the genomic distribution of PLINK-defined ROHs followed a highly non-uniform pattern that was highly concordant across all groups ( S13A Fig ) and similar to that obtained for the non-African 1000G populations ( S14 Fig; analysis performed on the markers present in the merged data set ) , with a number of ROHs reaching substantial frequencies in the Iranian population ( S8 Table ) . CNVs , as defined by the Axiom Analysis Suite v4 . 0 software , were predominantly detected in Iranian Gilaks , Mazanderanis and Sistanis ( S15 Fig ) and also comprised a highly non-uniform genomic distribution that showed virtually no systematic overlap with ROHs ( S13B and S13C Fig ) , resulting in a number of high-frequency CNV regions ( “CNV islands”; S9 Table ) in healthy individuals . The observed genetic diversity and partially different ancestry was also evident in the frequency differences for numerous trait-related or predisposing alleles in the Iranian ethnic groups ( S10 Table ) . In general , CIC groups tended to have very similar allele frequencies that were nevertheless often markedly different from those of Europeans , while Iranian Baluchis and Sistanis showed a tendency towards South Asians , although these trends were not present at all markers . A notable exception was lactase persistence-causing marker allele rs4988235-T whose frequency in Iranian Baluchis ( 22% ) was much higher than in any of the other Iranian groups , raising the prospect of convergent evolution [114] . However , we did not find evidence for a selective sweep based on Tajima’s D ( S16 Fig ) nor when using the integrated haplotype score ( iHS ) approach [115] ( S17 Fig ) . Although rs4988235 showed a substantial absolute score in Baluchis ( |iHS| = 2 . 42 ) , this value was not significant ( two-sided p>0 . 05 ) and we also did not observe a clear clustering of SNPs with extreme values as a possible indication for positive selection [116] .
Our study , based on genome-wide data from a stratified ethnic-group sampling and also including groups previously not well covered , such as Iranian Gilaks , Kurds , Mazanderanis and Sistanis , revealed the distinct and rich genetic diversity of the Iranian population , corroborating previous reports based on uniparental markers . The majority of Iran’s ethnic groups comprise largely overlapping genetic autosomal diversity , implicating a shared and largely autochthonous ancestry , designated as the Central Iranian Cluster ( CIC ) . Notably , the CIC also includes Iranian Arabs and Azeris ( Fig 1 ) as well as the religious group of Zoroastrians ( Fig 3 ) , being consistent with the suggestion that Zoroastrians have lived in the area of present-day Iran for millennia and had formed an early group of Indo-European speakers . Still , the CIC comprised substantial internal structure , with pairwise FST values up to an order of magnitude higher than those for more homogeneous populations of similar population size , such as Germany [117] , but below the level of substructure reported for Europe , Central Asia , the Near East or Southeast Asia as a whole [45] and much lower than for neighboring Armenia in the Caucasus [118] . Iranian Baluchis , Sistanis , Turkmen and Persian Gulf Islanders showed strong admixture , with the CIC ( or its ancestral population ) consistently contributing to all of them and contributions from different respective ‘opposite’ ancestral populations , evidencing CIC’s strong impact on human demography in this world region . Since substantial proportions of the Iranian population belong to non-Persian ethnic groups or are admixed , more precise reference to the particular ethnic groups appears mandatory when conducting future genetic studies . In comparison with global and local reference data , the CIC represents a distinct entity comprising an autochthonous genetic component , clustering closely with geographically adjacent populations and assuming a location in the ‘genetic map’ that corresponds to its geographic location at the nexus between South , Central and West Asia , Northern Africa and the Caucasus . This observation is consistent with limited gene flow reported in previous uniparental marker-based studies and adding a further example on the correspondence between genetic diversity and geographic location , such as Europe [73 , 119] , explicable by genetic drift as well as admixture . The largely autochthonous development of CIC groups , consistent with an early branching from the Eurasian population before the Neolithic [6] , is further corroborated by the distinctiveness of these groups in comparison to different time strata represented by aDNA samples , indicating a genetic continuity for at least several past millennia and eventually mirrored by Zoroastrian genomic diversity . Both , Early Neolithic farmers from West Iran and people from the Steppe appear to have made very limited contributions to CIC groups . In turn , the ‘African’ component shared between PG Islanders and some Sub-Saharan populations likely predates the beginning of the Neolithic and , thus , renders PG Islanders as an early autochthonous group that subsequently became strongly admixed with CIC groups . Notably , Iranian Arabs appear to be slightly genetically detached from other Arab populations in West Asia and Northern Africa . The small ancestry component shared between the CIC and Tuscans may mirror early migrations from the Near East although this requires further investigation . Correlating genetic affinity with spoken language yielded evidence for a number of language adoption cases in Iran . CIC’s distinct and autochthonous genetic variation indicates that Indo-European ( IE ) language ( s ) were likely adopted by some ancient population in Iran several millennia ago , although it remains unclear if this was driven by previously suggested aggressive warrior-bands migration [120] given the lack of Y-chromosomal data in our study . The observed close genetic proximity , based on genome-wide data , of Turkic-speaking Iranian Azeris as well as of Semitic-speaking Iranian Arabs to IE-speaking groups within the CIC , confirms previous reports on Semitic-speaking groups in Iran [58] and Turkic-speaking Azerbaijanis [91 , 121–123] . Given their genetic vicinity to other Arab and Caucasian populations , respectively , this is well explained by admixture between some overwhelmingly contributing ancestral IE population ( s ) and a minor genetic contributor whose language was adopted in the course of past entries . Finally , the spread of IE-speaking Iranian Baluchis , Sistanis and PG Islanders from the other IE-speaking CIC groups is explicable by repeated admixture of some IE-speaking ancestral population ( s ) with ancient South or West Asian populations , such as Early Neolithic West Iranians , respectively , while retaining their language , causing its adoption by the admixed offspring . The heterogeneous levels of substantial population substructure as well as of elevated consanguinity in the Iranian population have profound implications for future human genetic studies . They corroborate previous reports on different predisposing variant frequencies across Iran ( e . g . [36 , 37] ) and emphasize the need for an ethnicity-aware approach when performing human genetic studies or genetic counseling in Iran . Population-based association studies should focus on CIC groups to minimize biasing effects due to population stratification , applying to common single-marker analysis but in particular to rare-variant collapsing tests where regional and ethnic group-specificity is to be expected due to the average young age of these variants . Given the genetic diversity even within the CIC , ancestry correction appears mandatory while sample inclusion from the highly admixed groups may increase the risk of biased results . The observed elevated consanguinity in some ethnic groups is in line with previous reports on Iran and other West Asian populations , indicating past and ongoing consanguineous marriage practice and also possibly explaining reported differences between Iranian provinces and residential areas . Family-based linkage or homozygosity-mapping studies should preferentially target groups featuring increased consanguinity levels , namely Iranian Arabs , Baluchis and Sistanis , to increase power especially for studying autosomal-recessive diseases . When studying runs of homozygosity and copy-number variants in diseased individuals , for example in whole-exome and whole-genome sequencing studies , the frequent occurrence of such features in healthy individuals , as shown in this work , requires caution in the interpretation of these features . In summary , Iranians feature distinct genetic variability , resulting from long-standing genetic continuity , as well as substantial genetic heterogeneity and can , thus , not be treated as a single homogeneous entity . Future human genetic studies have to consider ethnic affiliations for sampling and analyses and should expect widespread admixture in both extant and ancient samples . The observed concordance between genetic diversity and geographic location and examples of lineage break up between language and genetic proximity are consistent with the archeological and historical evidence on Iran as occupying a stretch of land that has seen multiple migration and admixture events in the past millennia . By providing genome-wide population data for Western Asia , thereby filling a lack that has characterized this region for over a decade despite its known diversity and prominent place in human history , we hope to encourage future population genetic , evolutionary and medical studies in Iran and beyond .
This study has been approved by the Research Ethics Committee of the University of Social Welfare and Rehabilitation Sciences ( USWR ) , Tehran , Iran ( approval number IR . USWR . REC . 1395 . 376 ) . Prior to gathering information on sex , ethnicity , demographic and health status , we obtained written informed consent from each individual , according to the guidelines of the Research Ethics Committee , University of Social Welfare and Rehabilitation Sciences ( USWR ) , Tehran , Iran . We included 1069 healthy unrelated individuals from 11 major Iranian ethnic groups , including 800 from the Iranome project [124] as well as 269 additionally sampled individuals in the study , comprising Iranian Arabs , Azeris , Baluchis , Kurds , Lurs , Gilaks , Mazanderanis , Sistanis , Persians , Turkmen and Persian Gulf Islanders living in Iran ( Table 1 ) . Prior to gathering information on sex , ethnicity , demographic and health status , we obtained written informed consent from each individual , according to the guidelines of the Research Ethics Committee , University of Social Welfare and Rehabilitation Sciences ( USWR ) , Tehran , Iran . Individuals were required to have the same ethnic background for at least two generations . The majority of individuals were more than 40 years old at the time of recruitment , lowering the risk of manifesting genetic disorders in later life . All subjects were re-examined by a clinician . This study has been approved by the Research Ethics Committee of USWR , Tehran , Iran . Language family assignment was obtained from Glottolog 3 . 2 ( http://glottolog . org/ ) . Venous blood was taken from individuals . DNA extraction from blood samples was done using the salting out method [125] . Samples were genotyped using the Axiom Precision Medicine Research Array ( PMRA ) by Life Technologies , comprising about 903 , 000 markers . Samples were randomly assigned to genotyping array probes without regard to ethnic affiliation in order to avoid batch effects . Life Technologies’ AxiomAnalysisSuite v2 . 0 . 0 . 35 was used for evaluating and genotyping CEL files . After removing low quality samples ( quality < 97 ) , genotypes of 1058 samples were assigned using the Axiom_PMRA . na35 . annot . db annotation file . Further quality control was performed on those 1058 samples using PLINK [126] v1 . 9 and R v3 . 5 . 1 [127] . Variants were required to have call-rates ≥95% and deviations from Hardy-Weinberg equilibrium with p>10−5 . Samples were required to have a call-rate of ≥97% , to not show excessive hetero- as well as homozygosity ( <5 sd ) . Cryptic relatives ( mean identity-by-descent [IBD] sharing π>0 . 4 ) were detected using PLINK’s—genome option and 20 samples were excluded from the study for representing parent-child pairs , sib pairs or identical individuals . After quality control , the cleaned data set comprised 1021 samples ( Table 1 ) comprising genotypes for 829 , 779 autosomal markers . The overwhelming majority of sample pairs within an ethnic group ( typically ~99% or more ) were unrelated or only distantly related ( π<0 . 04125 ) , with only few pairs showing elevated IBD sharing ( S11 Table ) . For some analyses , we additionally considered only markers with common alleles ( minor allele frequency ≥5%; 311 , 262 markers ) , only markers in no strong linkage disequilibrium ( LD; r2≤0 . 5 , 500kb window size , 25 SNPs step size ) by using PLINK’s—indep-pairwise option ( 475 , 665 markers ) , or both ( 203 , 495 markers ) . In order to put the Iranian samples in a global as well as local context , we merged our cleaned data set with those of publicly available reference data sets , using only markers that were present in each of the datasets being merged . For a global comparison , we used 2492 unrelated samples assigned to 26 populations from the 1000 Genomes Project [41–43] ( “1000G”; accessed May 2017 ) . For a more localized comparison , we used samples from three different curated data sets , namely 120 samples from the Simons Genetic Diversity Panel ( SGDP ) [44] , 1345 samples from Lazaridis et al . [2] , partially including previously published samples , and 45 samples from Broushaki et al [6] . Notably , these reference data also included samples from a wide variety of ethnicities , such as Semitic groups ( e . g . Arabs , Assyrians , Jews ) , Caucasian groups ( e . g . Armenians , Georgians , Circassians ) , Zoroastrians and many others . We further grouped these samples for their corresponding geographic region ( S1 Table ) and language family ( S2 Table ) . Only markers with genotypes in both the Iranian and the respective reference data set ( s ) were included in the analysis and underwent additional quality control using the same thresholds as before . Again , for some analyses , markers in strong LD or with infrequent alleles were removed . After QC , the ‘global data set’ ( merger with 1000G ) included 782 , 127 markers , while 232 , 138 common markers remained after additional LD pruning and frequency filtering . The ‘local data set’ ( alternative merger with the other three reference data sets ) included 59 , 837 markers in total and 43 , 198 common , LD-pruned markers , respectively . A growing number of human aDNA samples from Iran and beyond have been published . We compiled 798 aDNA samples from 21 different publications and one pre-print [2 , 6 , 81 , 92–110] ( S3 and S4 Tables ) for spatial-temporal analysis . We applied multidimensional scaling ( MDS ) analysis based on identity-by-state ( IBS ) allele sharing to the LD-pruned data sets using PLINK’s—mds-plot implementation . PCA analysis was independently performed for each of the considered , possibly merged , data sets , except for the aDNA samples which were projected onto the components obtained from the merged data set of our 1021 extant Iranians and 118 SGDP samples geographically co-localizing with the aDNA samples ( S4 Table; S18 Fig ) . This data set underwent quality control , LD pruning , and frequency filtering using the same thresholds as before . We generated PCs of reference samples running TRACE from LASER [128] v2 . 04 in PCA mode ( -pca 1 ) using default parameters . Then we projected each aDNA sample independently onto the reference PCA using TRACE [128] v1 . 03 with default parameters . The number of markers each aDNA sample shared with reference PCA ranged from 30 , 000 to 80 , 000 . PLINK’s—fst option was used to estimate Weir & Cockerham’s FST fixation index [129] . For an approximate assessment of the upper limit of the impact of population substructure on genetic association studies in the Iranian population , we deliberately assigned , for each pair of ethnic groups , case status to all samples from one group and control status to all from the other and subsequently calculated the genomic inflation factor [130] ( GIF ) , where values of 1 . 0 correspond to no inflation , by using PLINK’s—adjust option . For exploratory admixture and migration analysis , we ran ADMIXTURE [111] v1 . 3 . 0 in parallel for K = 2 , … , 20 using random seeds and TreeMix [112] v1 . 13 through the Treemix_bootstrap . sh script of the BITE R package [131] v1 . 1 . 0004 allowing for 0 , 1 , 2 , 5 , 10 and 15 migration events to be replicated 100 times and made consensus trees based on replications using PHYLIP [132] v3 . 697 . The final tree was then plotted using treemix . bootstrap from BITE . The qp3Pop program of ADMIXTOOLS v5 . 0 package with default parameters was used to calculate f3 statistics [113] . Autocorrelation analysis for language family with respect to genetic distance based on the Euclidian distance in the first two principal components and Moran’s I [133] , obtained from running the TRACE software from LASER v2 . 0 [128] on the local data set , was performed using the Moran . I function from R package ape v5 . 1 [134] . To this end , language families were assigned numeric class values ( 1 , … , 8 ) . To avoid spurious effects due to this numeric ( instead of categorical ) coding , which would imply order and distance between classes , we performed 100 random permutations of the assigned values , thereby destroying potential biases introduced by arbitrary numbering , and report the respective distributional statistics . Tajima’s D was estimated with VCFtools v0 . 1 . 13 ( https://vcftools . github . io/man_latest . html ) using the—TajimaD option with a window size of 100 kb and the—from-bp/—to-bp commands for a sliding window analysis . We also performed an integrated Haplotype Score ( iHS ) analysis [115] of the LCT region on chromosome 2 . To this end , haplotypes were estimated using ShapeIt v2 . r790 [135] based on the 1 , 000 Genomes Phase 1 haplotype reference panel and genetic map of chromosome 2 ( downloaded from: https://mathgen . stats . ox . ac . uk/impute/data_download_1000G_phase1_integrated . html ) . A total of 34 , 746 SNPs on chromosome 2 coincided between our data set and the reference panel . The iHS scan was performed using R package rehh v2 . 0 . 2 [136 , 137] ( downloaded from: https://cran . r-project . org/web/packages/rehh/index . html ) . Inbreeding coefficients ( FI ) were estimated using PLINK’s—ibc option ( ‘Fhat3’; [138] ) based on LD-pruned autosomal markers and separately for each ethnic group . Furthermore , we defined runs of homozygosity ( ROHs ) using PLINK v1 . 9 ( LD-pruned autosomal markers; ethnic groups combined ) and GARLIC [31 , 139] v1 . 1 . 4 ( autosomal markers; separately for each ethnic group ) and autozygous genomic regions using IBDseq [32] ( LD-pruned autosomal markers; separately for each ethnic group ) , using default options and applying them to separate data subsets containing only a single population or ethnic group , respectively . We used the Axiom Analysis Suite ( AxAS v4 . 0 ) with default options in order to detect copy-number variants ( CNVs ) . We divided the set of 1021 Iranian samples into 5 groups where each group comprised similar proportions of males and females from the 11 ethnic groups . Samples in each group were used to construct a reference for CNV detection , subsequently running the CNV detection for the same groups . CNVs were required to comprise at least 25 and 50 markers for homozygous and heterozygous variants , respectively . All statistical analyses were performed and graphs were created using R with in-house scripts , unless noted otherwise . Two-dimensional kernel density estimates were obtained using the Hpi and kde functions from the ks package v1 . 11 . 3 [140] for R . Map data were obtained from GADM v2 . 8 ( November 2015; www . gadm . org ) and maps were plotted using functions in the sp package v1 . 3–1 [141 , 142] for R . Bar plots were created using functions in ggplot2 v3 . 0 . 0 [143] for R . | Based on genome-wide genotype data on over 1000 samples from eleven ethnic groups present in Iran and by comparison to reference data sets of both extant populations and ancient DNA samples , we show that the Iranian population comprises distinct genetic variation with respect to populations in close geographic proximity , a cluster of genetically largely overlapping ethnic groups as well as a number of strongly admixed groups . These observations , also corroborated by f3 migration statistics and other approaches , indicate genetic continuity of and limited influx into the cluster groups over several millennia , despite Iran’s geographic position at a crossroads in West Asia . They also suggest , correspondingly , several instances of language adoption instead of demic replacement in the past . Future human genetic studies , both with a focus on population and medical genetics , will have to consider differences in heterogeneity , consanguinity and degree of admixture between the ethnic groups for an adequate design and interpretation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] | [
"population",
"genetics",
"geographical",
"locations",
"social",
"sciences",
"neuroscience",
"ancient",
"dna",
"ethnicities",
"cognitive",
"psychology",
"paleontology",
"dna",
"population",
"biology",
"paleogenetics",
"persian",
"people",
"arabic",
"people",
"language",
"iran",
"people",
"and",
"places",
"biochemistry",
"psychology",
"asia",
"nucleic",
"acids",
"earth",
"sciences",
"genetics",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"evolutionary",
"biology",
"cognitive",
"science"
] | 2019 | Distinct genetic variation and heterogeneity of the Iranian population |
Recommendations for soil-transmitted helminth ( STH ) control give a key role to deworming of school and pre-school age children with albendazole or mebendazole; which might be insufficient to achieve adequate control , particularly against Strongyloides stercoralis . The impact of preventive chemotherapy ( PC ) against STH morbidity is still incompletely understood . The aim of this study was to assess the effectiveness of a community-based program with albendazole and ivermectin in a high transmission setting for S . stercoralis and hookworm . Community-based pragmatic trial conducted in Tartagal , Argentina; from 2012 to 2015 . Six communities ( 5070 people ) were enrolled for community-based PC with albendazole and ivermectin . Two communities ( 2721 people ) were re-treated for second and third rounds . STH prevalence , anemia and malnutrition were explored through consecutive surveys . Anthropometric assessment of children , stool analysis , complete blood count and NIE-ELISA serology for S . stercoralis were performed . STH infection was associated with anemia and stunting in the baseline survey that included all communities and showed a STH prevalence of 47 . 6% ( almost exclusively hookworm and S . stercoralis ) . Among communities with multiple interventions , STH prevalence decreased from 62% to 23% ( p<0 . 001 ) after the first PC; anemia also diminished from 52% to 12% ( p<0 . 001 ) . After two interventions S . stercoralis seroprevalence declined , from 51% to 14% ( p<0 . 001 ) and stunting prevalence decreased , from 19% to 12% ( p = 0 . 009 ) . Hookworm’ infections are associated with anemia in the general population and nutritional impairment in children . S . stercoralis is also associated with anemia . Community-based deworming with albendazole and ivermectin is effective for the reduction of STH prevalence and morbidity in communities with high prevalence of hookworm and S . stercoralis .
Soil-transmitted helminth ( STH ) infections are the most prevalent Neglected Tropical Diseases ( NTD ) worldwide affecting over 2 billion people . Four nematode species ( Ascaris lumbricoides , Trichuris trichiura , Necator americanus and Ancylostoma duodenale ) are the most common STH infections of humans [1 , 2] . Due to their common biological characteristics and risk factors , geographic overlap and anthelmintic treatment of choice , the recommendations from the World Health Organization ( WHO ) for STH control target the four species together [3] . Strongyloides stercoralis is an STH with similar distribution but with some distinctive characteristics regarding its diagnosis and therapy that have prevented its inclusion in the current guidelines for STH control; however , it could be targeted with updated comprehensive control strategies [4 , 5] . STH morbidity is of public health concern because of the population affected , the high prevalence in low income countries and the long lasting consequences of the infection , all contribute to the economic impact of the disease and perpetuation of poverty [6] . Chronic STH infection has been associated with cognitive impairment in school age children ( SAC ) and negative impact on motor and language development of preschool age children ( PSAC ) [7–9] . Negative effects of STH infection on nutritional status of children have also been described such as stunting , reduced weight gain and specific micronutrient deficiencies ( i . e . iron and vitamin A ) [10] . A major consequence of STH infection is iron-deficiency anemia , of significant relevance in infants and pregnancy outcomes . STH infections and other tropical infectious diseases , such as schistosomiasis and malaria , are implicated in the etiology of iron-deficiency anemia in lesser-developed countries [11 , 12] . Among them , hookworm infection has been strongly associated with the development of iron-deficiency anemia , due to chronic intestinal blood loss [13] . Comprehensive STH control strategies include health education , improvements in water , sanitation and hygiene , and Preventive chemotherapy ( PC ) through mass drug administration ( MDA ) of albendazole or mebendazole [14 , 15] . Implementation of PC for school age children ( SAC ) , the group targeted by the current recommendations , has shown to be effective in reducing worm burden and STH prevalence as much as in improving nutritional status and anemia [16 , 17] . However , critical groups of the population like PSAC and women of reproductive age are not directly reached by this approach and would benefit if another strategy for deworming such as community-based PC were applied [9 , 18] . The use of a drug active against S . stercoralis ( ivermectin ) along with albendazole or mebendazole would enhance the PC effectiveness in the control of S . stercoralis and T . trichiura and would carry additional benefits such as the reduction in the prevalence of scabies and impetigo [5 , 19–21] . Argentina has a heterogeneous prevalence of STH infection , with areas of high prevalence in the north . Previous studies in Salta province , in the Northwest , showed a cumulative prevalence of STH near 50% , with preponderance of S . stercoralis ( 20–48% ) and hookworm ( 20–45% ) [22 , 23] . Anemia is a public health problem that affects 18% of the Argentinian population , particularly in the northwest , where 38% of the PSAC and 19% of women between 10 and 49 years old are anemic [24 , 25] . The aim of the present study was to assess the effectiveness of a community based PC program against STH , using a combination of albendazole and ivermectin . For that purpose , we compared the results of fecal , blood and anthropometric surveys carried out before each PC intervention through a longitudinal community based operational research . The effectiveness of the intervention was monitored through the evaluation of STH cumulative prevalence of the participant communities and variation in morbidity indicators such as anemia and nutritional status of children . The study was part of a larger project with the objective to incorporate a program of PC for STH control into the regular activities of the public primary health care system in high prevalence regions of Salta province , Argentina .
Community-based pragmatic non-randomized trial conducted in Tartagal , Salta province , Argentina between August 2012 and May 2015 . All participants selected for surveillance provided written informed consent prior to the study and parents/guardians provided informed consent on behalf of minor participants . The research protocol and the informed consent forms were approved by the Bioethics Committee of the Colegio de Médicos de la Provincia de Salta and by the Bioethics Committee of the Faculty of Health Sciences at the Universidad Nacional de Salta ( FWA registered committee ) . The anthelmintic drugs were used according to currently approved and recommended regimens . All members of six communities from Tartagal were invited to participate in the study . A community is defined as the group of people with a common ethnic origin that lives in neighboring households and shares a unique community leader; generally including a group of around 100 households . Four of the communities enrolled in the study were peri-urban: Lapacho Alto , Kilometro 6 , Las Moras and Lapacho I; and two communities were urban: Pablo Secretario and Tapiete , These communities are served by the Provincial primary health care system; therefore , trained public health personnel ( “sanitary agents” ) visited each household every three months . The communities were selected by the local public health authorities based on the sanitary risk indicators collected in the census the year before the study started . Communities with report of malnutrition in PSAC , child or maternal mortality of preventable causes or known STH prevalence above 50% were enrolled . All these communities also share similar water and sanitation conditions [23] and are homogeneous in their economic status , most of the families have a low monthly income that comes from informal and transitory jobs of a single breadwinner and from small economic benefits from the government . A sample was recruited for stool and blood surveillance , using stratified random selection with community as the stratification factor and household as the unit for choice . Inhabitants of the selected households , of any age and gender , were asked to provide a stool and blood samples prior to deworming . Since this is a population-based trial , the subjects recruited for surveillance were not necessarily the same individuals at baseline and follow-up . Nevertheless , the random selection strategy and sample size of the survey group were similar throughout the study . The number of stool samples collected within the survey group was 397 at baseline; 130 at first follow up and 181 at second follow up . Blood samples were 409 at baseline; 156 at first follow up and 165 at second follow up . All the participants were offered anthelmintic therapy , independent of their age ( age ranging from 1 to 91 years old ) or recruitment to the survey group . Two anthelmintic drugs in single doses were used simultaneously , albendazole 400mg tablets , dosed at 1 tablet ( half the dose in children 12 to 24 months of age ) and ivermectin 6mg tablets dosed at 200 micrograms/kg; produced by GlaxoSmithKline and ELEA Argentina respectively . The drugs were administered according to the following inclusion and exclusion criteria . Inclusion criteria: 1 ) Permanent resident of the enrolled community; 2 ) Willing to take anthelminthic drugs . Exclusion criteria: 1 ) Age < 12 months ( albendazole ) ; 2 ) Body weight < 15 kg ( ivermectin ) ; 3 ) Confirmed or suspected pregnancy , contraindication for albendazole in the first trimester and for ivermectin in any trimester; 4 ) Women breast-feeding new born babies , contraindication for ivermectin in the first week of puerperium; 5 ) Rejection to take anthelminthic drugs; 6 ) Known allergy to albendazole or ivermectin; 7 ) Taking an alternative anthelminthic therapy at the moment of the intervention . PC was carried out by health personnel and members of the research team who distributed the drugs through intensive deworming campaigns in the communities ( house by house ) and local schools . Each campaign lasted 4 days per community and was carried out sequentially , due to operative limitations . During the deworming campaigns , all the participants who met all the inclusion and none of the exclusion criteria received the anthelmintic therapy . Passive pharmaco-vigilance activities were carried at the local tertiary hospital ( Hospital Juan D . Peron ) and the respective sanitary posts of ambulatory and emergency services were aware of the interventions . The first intervention took place in August 2012 and the last one in May 2015; during that period , some communities were treated once and others received three rounds of PC with an interval of 9 to 16 months between rounds . Stool and blood surveillance were performed at baseline and follow-up ( before each round of PC ) . The results of the interventions were entered in the data base for each participant as: treated ( if the subject was administered one or two drugs ) ; excluded ( if both drugs were contraindicated ) ; absent ( if the subject was not found at home during the deworming round or the household could not be found ) ; death; migration ( if the subject moved from the community before the intervention ) and rejection ( if the subject did not want to take the anthelminthic drugs when offered ) . Coverage was calculated through the formula: n people treated/n people eligible for treatment and eligibility was calculated as: total population– ( excluded + death + migrants ) . The following data were gathered for the study: a ) baseline socio-demographic information; b ) baseline and follow-up anthropometric data; c ) baseline and follow-up stool results; d ) baseline and follow-up hematologic results , and e ) baseline and follow-up serologic results for S . stercoralis . The methods used for data collection are detailed below: a ) Socio-demographic assessment: public health census forms include data about each household of the community , collected through direct observation by the sanitary agent , along with individual demographic information for each inhabitant of the household . Demographic information was entered in the study database using the sanitation and drinking-water ladders designed by WHO/UNICEF for sanitary census [26] . b ) Nutritional survey: during the deworming visits , members of the research team registered weight and height of children from one to fifteen years-old . Weight was measured using a standard electronic scale in kilograms and grams . Height was measured with a steel tape while the child was standing near to a wall and registered in centimeters . In order to make the results comparable , population based sampling was used; the number of baseline height and weight observations was above the minimum sample of 400 suggested by WHO for nutritional surveys . Results were compared to the NCHS/WHO reference population and the expected ranges of standard deviations of the anthropometric indicators were considered to control the accuracy of the measurements . Anthropometric data was analyzed through WHO Anthro and WHO Anthro Plus softwares ( Department of Nutrition , WHO ) to calculate relevant Z-scores . Weight , height , Weight-for-Age z-score ( WAZ ) , Height-for-Age z-score ( HAZ ) and Weight-for-Height z-score ( WHZ ) of each participant child at baseline ( before the first PC round ) and follow-up ( before second and third PC rounds ) were registered . Children were classified as stunting ( HAZ <-2 SD from the international reference median value ) ; underweight ( WAZ <-2 SD from the international reference median value ) or wasting ( WHZ <-2 SD from the international reference median value ) according to the WHO recommendations on z-scores interpretation [27] . c ) Parasitological surveillance: a single fresh stool sample without preservatives was collected from each participant of the survey group . Sterile stool containers and instructions were distributed house by house , collected the following morning and analyzed within 24 hours of collection in a reference laboratory . Five parasitological techniques were used: sedimentation/concentration; agar plate culture; Harada-Mori filter-paper culture; Baermann concentration of charcoal-cultured fresh stool , and McMaster egg counting method as described elsewhere ( 28 , 29 ) . If the sample volume was insufficient to perform all methods , concentration technique was prioritized due to its overall higher sensitivity in preliminary studies [28] . A high level of certainty in the distinction between hookworm and S . stercoralis larvae in the culture techniques was assured by the different incubation time of the methods ( 24 hours for Baermann and 7 days for Harada-Mori ) , adapted to each specie life cycle , and by the long experience of the technicians who observed and supervised the microscopic exams . The findings of the different methods were grouped and entered in the database as positive if at least one method was positive or negative if all the methods were negative , for each STH species . McMaster´s results were recorded as egg per gram ( EPG ) . The stool survey was performed at baseline and before each PC campaign . d ) Hematological surveillance: participants of survey group had 5 mL blood drawn through venipuncture . A complete blood count was performed using a SYSMEX automated hematology analyzer KX 21N . The results of Hemoglobin value ( Hgb ) , white blood cell count ( WBC ) and eosinophil relative count were registered . Subjects were classified as anemic or not anemic using the Hgb thresholds to define anemia according to sex and age set by WHO/UNICEF [12] . Absolute eosinophil count was calculated and eosinophilia was defined as absolute eosinophil values >500 cells/mm3 [29] . e ) Blood samples were centrifuged and an aliquot of serum was preserved frozen at –20°C and analyzed with the in-house enzyme-linked immunosorbent assay ( NIE-ELISA ) method for the diagnosis of S . stercoralis . NIE-ELISA detects IgG antibodies against recombinant NIE antigen of S . stercoralis L3 larvae , as has been described previously [30 , 31] . Patient’s sera were tested in duplicate and compared to a standard positive IgG curve run on each plate . The averages of duplicate results were calculated and corrected for background reactivity ( no serum added ) . The primary outcome was STH prevalence . The cumulative and species specific STH prevalence before and after deworming was compared to monitor the effectiveness of community-based PC . Secondary outcomes were indicators of morbidity potentially due to STH infection such as anemia , eosinophilia and nutritional impairment . The study was analyzed in two phases: i ) baseline assessment that included a description of the study population targeted with PC and an exploration of the survey group at the individual level , searching for associations of STH infection with nutritional and hematological findings; and ii ) longitudinal assessment to evaluate the impact of PC on population-related parameters in the communities where repeated PC was carried out . Sample size was estimated considering a predicted prevalence of 50% , a confidence level of 95% , and a design effect of 2 . Sample size calculation was based on observing a specific reduction of 10% or more in STH prevalence; the estimated sample size was n = 190 . Continuous quantitative measures evaluated at different time points were described using proportions with 95% confidence intervals ( 95% CI ) ; means with standard deviations ( SD ) and medians with interquartile ranges ( IQR ) . Comparisons between infected and uninfected people at baseline and between pre and post intervention parameters were carried out using T test and Mann-Withney U test . Significant associations between STH infection and morbidity indicators were explored through stratified bivariate analysis and , afterwards , adjusted through multivariate logistic regression models; statistical significance was assessed by Chi-square test with 95% significance . Correlation between continuous quantitative measures was explored through linear regression tests of Pearson´s or Spearman´s ( according to the underlying distribution ) . All data was entered in Microsoft Access 11 . 5 ( Microsoft , Redmond , WA ) with an Epi Info 3 . 5 . 4 ( CDC , Atlanta , GA ) view . Duplicate data entry was performed by trained collaborators . The analysis was performed with EPIDAT 3 . 1 ( PAHO , Washington , DC ) and R 3 . 1 . 1 ( The R Foundation for Statistical Computing , GNU General Public License ) .
We found a heterogeneous distribution of STH prevalence between communities varying from 11% to 72% , urban communities ( Pablo Secretario and Tapiete ) had a significantly lower prevalence than peri-urban communities ( Lapacho Alto , Kilometro 6 , Lapacho I and Las Moras ) , p<0 . 001 . Table 1 details the baseline prevalence of STH infections and the coverage reached with the first intervention in each one of the studied communities . Among the 6 communities , hookworm ( 34% ) and Strongyloides stercoralis ( 26% ) were the most frequent parasitological findings . Regarding hookworm , species identification was done through Harada-Mori in 73 of the 135 positive cases: 63 were A . duodenale; 9 were N . americanus and 1 was a co-infection of both species . The other 62 cases were negative in Harada-Mori . Therefore , A . duodenale accounted for 86% of the hookworm cases identified by species . A total of 104 S . stercoralis cases were diagnosed , 53 of them were detected by parasitological exam , 53 were detected by NIE-ELISA serology and 18 were positive by both methods . Table 2 summarizes a baseline description of the study population and the findings in the survey group . We found significant differences in mean hemoglobin level , median eosinophil count and mean HAZ according to STH infection status at the individual level ( uninfected versus infected subjects ) . No significant differences were found in mean WAZ for children aged 1 to 15 years old and WHZ of children aged 1 to 4 years old . Table 3 displays the comparison of hematological and nutritional results between infected and uninfected groups . Hookworm infection was significantly associated with anemia ( odds ratio [OR] = 5 . 5; 95% CI: 2 . 7–11 . 4 ) ; eosinophilia ( OR = 7 . 48; 95% CI: 3 . 0–20 . 8 ) and stunting ( OR = 1 . 7; 95% CI: 0 . 8–3 . 8 ) . S . stercoralis infection was also significantly associated with anemia ( OR = 3 . 5; 95% CI: 1 . 7–7 . 1 ) and eosinophilia ( OR = 2 . 3; 95% CI: 1 . 2–4 . 9 ) but not with stunting ( OR = 1 . 01; 95% CI: 0 . 4–2 . 3 ) . Fig 2 displays the adjusted odds ratios after controlling for potential confounders through multivariate logistic regression models . We found a linear negative correlation between hookworm intensity , calculated as EPG and hemoglobin level . Non-parametric Spearman´s coefficient of correlation ( rs ) = -0 . 46; p< 0 . 001 . Fig 3 shows the scatter plot of this correlation . It should be noted that the frequency of heavy and moderate intensity infections was low with most cases having light infections . However , even light egg burden was significantly associated with anemia since 65% ( 95% CI: 48–81 ) of the subjects with light hookworm infection were anemic compared to 12% ( 95% CI: 5–19 ) of uninfected subjects . Lapacho Alto and Kilometro 6 inhabitants ( n = 2685 ) received three community deworming interventions . We found significant improvements in the prevalence of STH infection , anemia , eosinophilia and children´s chronic malnutrition ( stunting ) after community-based PC with albendazole and ivermectin . Table 4 summarizes the indicators measured at baseline ( before first PC ) and at follow-up ( before second and third interventions ) . We found a significant decrease in the cumulative prevalence of STH infection and in the species-specific prevalence of hookworm and S . stercoralis after the first intervention . Hookworm infection intensity decreased with deworming , showing fewer moderate and no heavy intensity infections at follow-up . We were not able to calculate hookworm’s Eggs Reduction Rate ( ERR ) after deworming for several reasons that attempted against the validity of this indicator . The number of stool samples collected at baseline and follow up was below the recommended sample size of 200 for this type of calculation[32] . Follow up infection intensity was assessed much later than the recommended two weeks after PC; therefore , reinfection could have occurred between baseline and follow-up assessments . Not all the samples that tested positive for hookworm through other techniques showed eggs to be counted through McMaster method , suggesting that those cases correspond to low-intensity infections . [32] . S . stercoralis´s seroprevalence showed non-significant changes after the first intervention but diminished significantly after the second intervention . Fig 4 shows the evolution of relative IgG titers across the study . Anemia and eosinophilia prevalence and severity showed a significant response to the first intervention . When comparing hemoglobin levels between baseline and follow up surveys we found that mean hemoglobin level increased with deworming but also the curve shifted to the right showing higher minimum and maximum levels and eliminating the most extreme cases of anemia . Age and gender distribution of anemia also showed a modification following deworming . At baseline the most affected groups were PSAC and adult women but SAC and adult men were affected as well . In follow-up surveys the prevalence in PSAC , SAC and adult men declined to low levels , while adult women remained anemic; although with lower prevalence ( Fig 5 ) . We found an impact of deworming on stunting prevalence after two deworming campaigns but no changes in underweight and wasting , which were infrequent already at baseline . The severity of the deficit in height showed improvement after one PC intervention: baseline mean HAZ = -0 . 93 ( SD = 1 . 2 ) was statistically different than the measurement at first follow up which had a mean HAZ = -0 . 72 ( SD = 1 . 3 ) ( p = 0 . 009 ) ; although it should be noted that mean HAZ remained negative after PC . When discriminating the evolution of HAZ in PSAC from SAC , we found a significant difference in PSAC between baseline and first follow up and between first and second follow up . While SAC , showed no difference in HAZ after deworming . However , the number of HAZ observations within each group was below the minimum of 400 recommended to perform comparisons . Fig 6 shows the evolution of HAZ values in PSAC and SAC .
The current WHO strategy for the control of STH as a public health problem relies on the use of PC with albendazole or mebendazole for those groups and countries that carry the heaviest burden of disease . The long term and sustainable solution for the STH problem probably remains more closely linked to the provision of water and sanitation . Such an approach will impact several NTDs while addressing broader goals of shared prosperity and equity [33] . While aiming for coverage indicators to achieve public health and morbidity goals , evidence supporting the relationship between coverage and morbidity in STH is still incomplete . With the understanding that WHO recommendations are a minimum set of goals for resource limited settings , this report describes a strategy for STH control in resource-limited communities with significant deficits in sanitation , where a primary care public health system is in place and Strongyloides stercoralis has been described at high prevalence [23 , 28] , through a comprehensive baseline survey and a community based intervention with albendazole-ivermectin , achieving significant impact on morbidity indicators . The significance of STH as a public health problem in the study region was confirmed in this study both in terms of prevalence and morbidity indicators [22 , 28] , with a baseline prevalence of 48% ( IC95% 42–53% ) . Most of the prevalence was due to hookworm and S . stercoralis . We believe that the scarcity of A . lumbricoides and T . trichiura infections in our study population might be due to the wide distribution of piped water in the households , which might have a protecting effect against orally acquired infections . Similarly , the high prevalence of hookworm and S . stercoralis found might be related to the lack of sanitation facilities in most households , since this could increase the risk of infection by skin-penetrating species . This selectiveness of the risk factors ( unimproved sanitation and unimproved water access ) to the route of transmission of each STH species has been explored in a previous study of our group [23] . An alternative explanation for the low frequency of A . lumbricoides and T . trichiura infection could be the issue that we used only two methods for the detection of these species , compared to the three or five methods used for the diagnosis of hookworm and S . stercoralis respectively . However , the sensitivity of the two methods that detected eggs ( sedimentation/concentration and McMaster ) is appropriate ( around 90% ) as reported in other studies [34 , 35] . As a matter of fact , the use of Harada-Mori did not add new diagnosis of hookworm to the findings of concentration , this method only contributed with the specie identification; while additional cases of S . stercoralis , that went unnoticed with concentration , were detected through Baermann or agar plate cultures , suggesting an adequate sensitivity of sedimentation/concentration for the detection of eggs but a limited capacity to find larvae . The prevalence of anemia of 31% puts these communities in the category of “Severe public health significance” according to WHO definitions [12] . Anemia was significantly associated with hookworm and S . stercoralis infection in the baseline cross-sectional analysis . Even though the design of the study is not proper to demonstrate causality , the existence of previous evidence linking STH infections ( especially hookworm ) with anemia , explained by already well known pathogenic mechanisms [13 , 36–38]; added to the fact that hematologic parameters improved in the population after the intervention with anthelminthic ( and nothing else ) make us infer a probably causal relationship between infection and anemia . Guidelines for anemia and STH both recommend deworming as a significant although not sufficient measure to improve the public health situation of communities with high prevalence of anemia[11 , 12 , 39] . Impact indicators following PC demonstrated significant reductions in terms of prevalence and morbidity . Statistically significant changes in STH and anemia prevalence were both achieved after a single intervention . Interestingly , the residual anemia in the post intervention assessment was concentrated in adult females , the group that has a significant complementary cause of anemia and depleted iron stores due to menstrual blood loses; still , this group also benefited from the intervention lowering the prevalence of anemia . These results highlight the vulnerability of this group and the need for STH control measures that target women of childbearing age [40 , 41] . In accordance with other studies showing intervals of up to a year , S . stercoralis prevalence measured through NIE-ELISA only showed significant decreases in prevalence after 2 rounds of MDA [42] . The use of a comprehensive diagnostic panel allowed the identification of S . stercoralis as a significant pathogen and its morbidity , justifying the choice of the selected drug regimen . Should the diagnostic approach have been limited to a quantitative egg detection method , like Kato-Katz , McMaster`s or Mini-FLOTAC , this species would have gone completely unnoticed . Another significant observation from this diagnostic approach is the recognition that low burden hookworm infections carry a significant burden of anemia ( Fig 3 ) , which is different than in previous reports [43] . If confirmed , these results should stimulate a reassessment of the current approach that links morbidity solely to moderate and high burden infections . This finding is possibly related to co-existing conditions present in these communities , such as inadequate iron dietary intake and depleted iron stores of the subjects , which increases the likelihood that even relatively small chronic blood losses cause anemia; therefore , our results might not be generalizable . The impact on growth indicators in children , which are probably multifactorial , showed that height-for-age z-scores ( HAZ ) were significantly lower in STH infected children at baseline; moreover , as indicated by WHO nutritional guidelines [27] , the fact that the mean HAZ is below zero suggests that the whole population is affected or at risk . Stunting , as the nutritional parameter indicating more severe compromise of HAZ , was significantly associated with hookworm infection and underwent significant improvement after the second ( but not the first ) drug intervention , possibly indicating the interval of time free of STH needed to observe changes in this growth parameter . Even though the mean HAZ significantly increased after two PC interventions it remained negative suggesting that a chronic nutritional impairment persisted despite deworming; these might be related to the relative short time of follow up or to concomitant causes of malnutrition such as deficient dietary intake . We were not able to demonstrate any association of hookworm or S . stercoralis infection with weight for age z-score ( WAZ ) and weight for height z-score ( WHZ ) , or impact of deworming on these parameters . The association of STH infection with anemia and nutritional impairment needs to be considered in the context of possible confounding factors such as family income . Poverty is a well-known risk factor for STH infection but may also be a cause or contributing cause of anemia and malnutrition , since poor families tend to have diets insufficient in quantity and quality with low iron content . We did not include poverty in the multivariate analysis to confirm or discard a confounding effect in the association found between hookworm or S . stercoralis infection and anemia or malnutrition because data of family income was unavailable . However , the homogeneity of the socio-economic status of the study population makes us infer that the low family income was similarly distributed in the infected and uninfected groups . Our study has limitations to be considered: first , the surveillance samples at baseline and follow-up are not matched samples; therefore , we compared the results before and after PC by community and not individually . In a recently published article , we demonstrated in a smaller group in a community project close to the study area , with paired samples , that after a year of follow-up , decreases in NIE-ELISA antibody titers were similar to those observed in this study [44] . Second , some communities that were studied at baseline and treated for the first time were not surveyed at follow up and did not received successive deworming; we included these communities in a cross sectional baseline assessment looking for associations of STH infection but for the prospective analysis only the communities where follow up surveillance was performed were compared . Third , false negative results could occur due to suboptimal sensitivity of the diagnostic methods , which can overestimate the impact of deworming since a reduction in burden might be read and interpreted as a negative result . To deal with this limitation , various diagnostic methods were used . More sensitive methods such as PCR for the detection of STH might overcome this limitation [45] . Fourth , some self-selection bias might have happened related to the fact that some individuals selected for surveillance refused to participate and those among the selected that provided samples for the study might be at higher risk of infection . Fifth , the coverage reached with the interventions was not optimal , which could underestimate the real effectiveness of the drug regimen . Finally , we do not report the impact of deworming on infection intensity; due to the limitations of our study for the estimation of ERR , even at group level . Most of our limitations are typical of population-based trials in a resource-restricted setting . Despite these limitations , the analysis performed , including multivariate analysis , and the strength of the associations and effects , in the context of no other changes in conditions than the PC , provide the foundations for our interpretations regarding the benefits of PC with albendazole and ivermectin . As previously mentioned , we found difficulty in conducting PC in some communities where the collaboration of the sanitary agents responsible for the areas was meager and were geographically isolated which resulted in low coverage rates and the discontinuation of the study in those communities . For those situations , a school based intervention , as recommended by WHO [46] , is a more feasible option , even though with more limited results in terms of morbidity reduction since women and preschool age children are not targeted . The drug regimen used in these interventions has been used in a few clinical trials for the treatment of STH , mostly aiming at improving efficacy for T . trichiura [20 , 47] , however most of the data regarding its use , safety and even impact on STH , comes from lymphatic filariasis ( LF ) programmes in areas with onchocerciasis , where millions of individuals have been treated , highlighting the safety of this regimen [48 , 49] . Since currently most PC programs are dependent on drug donations , the addition of IVM needs to be considered in the context of the critical issue of drug availability . This is , to our knowledge , the first report of an intervention of PC with albendazole-ivermectin for pure STH control; its positive short-term effects on morbidity are echoed by recent reports proposing this drug combination as a feasible strategy for delaying the emergence of resistance and offering improved efficacy against T . trichiura and S . stercoralis in clinical trials and mathematical models [20 , 50–52] . In summary , albendazole/ivermectin applied as community based PC in communities with high prevalence of hookworm and S . stercoralis resulted in significant reduction of STH prevalence and improvements in anemia in the general population and nutritional status in children . These results should be further explored in randomized-controlled trials with cost-effectiveness evaluation to overcome the limitations of our study . | Soil-transmitted helminth ( STH ) infections are a relevant public health problem in resource restricted settings due to their potential to perpetuate poverty , since chronic infections are associated with learning and grow impairment in children and reduced productivity in adults . The current strategy for STH control gives a key role to preventive chemotherapy of risk groups ( preschool and school age children and women of childbearing age ) with anthelmintic drugs . The drugs recommended for regular deworming are albendazole or mebendazole . This strategy does not target Strongyloides stercoralis , an STH resistant to the recommended drugs in single doses . The efficacy of ivermectin , alone or in combination , for the treatment of Strongyloides stercoralis infection has been reported in controlled trials . We conducted a pragmatic study aiming to assess the effectiveness of community based preventive chemotherapy with albendazole plus ivermectin for the control of STH prevalence and morbidity , in endemic communities of Northwestern Argentina . We found high baseline prevalence of hookworm and Strongyloides stercoralis and significant nutritional and hematological morbidity associated with these infections . After three rounds of preventive chemotherapy with albendazole and ivermectin we observed a significant decline in the prevalence of infection and in the prevalence and severity of morbidity . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"children",
"medicine",
"and",
"health",
"sciences",
"helminths",
"tropical",
"diseases",
"hookworms",
"parasitic",
"diseases",
"anemia",
"animals",
"health",
"care",
"age",
"groups",
"neglected",
"tropical",
"diseases",
"strongyloides",
"stercoralis",
"families",
"morbidity",
"public",
"and",
"occupational",
"health",
"strongyloides",
"hematology",
"people",
"and",
"places",
"helminth",
"infections",
"eukaryota",
"health",
"statistics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"soil-transmitted",
"helminthiases",
"organisms"
] | 2017 | Albendazole and ivermectin for the control of soil-transmitted helminths in an area with high prevalence of Strongyloides stercoralis and hookworm in northwestern Argentina: A community-based pragmatic study |
B cells can contribute to acquired immunity against intracellular bacteria , but do not usually participate in primary clearance . Here , we examined the endogenous CD4 T cell response to genital infection with Chlamydia muridarum using MHC class-II tetramers . Chlamydia-specific CD4 T cells expanded rapidly and persisted as a stable memory pool for several months after infection . While most lymph node Chlamydia-specific CD4 T cells expressed T-bet , a small percentage co-expressed Foxp3 , and RORγt-expressing T cells were enriched within the reproductive tract . Local Chlamydia-specific CD4 T cell priming was markedly reduced in mice lacking B cells , and bacteria were able to disseminate to the peritoneal cavity , initiating a cellular infiltrate and ascites . However , bacterial dissemination also coincided with elevated systemic Chlamydia-specific CD4 T cell responses and resolution of primary infection . Together , these data reveal heterogeneity in pathogen-specific CD4 T cell responses within the genital tract and an unexpected requirement for B cells in regulating local T cell activation and bacterial dissemination during genital infection .
Chlamydia trachomatis is an obligate intracellular pathogen that causes the most prevalent bacterial sexual transmitted infection worldwide [1] . In the US , Chlamydia is now the most common notifiable disease reported to the US Centers for Disease Control ( CDC ) . The 1 . 4 million cases of Chlamydia infection reported in 2011 represent an 8% increase over the previous year and is the largest number of annual infections ever reported to the CDC for any condition [2] . The introduction of a Chlamydia screening and control program in the mid-1990s has not prevented annual increases in infection , although a portion of this increase is due to improved disease surveillance [3] . Overall , the CDC reports a median 8 . 3% Chlamydia positivity test among women aged 15–24 , making this one of the most prevalent bacterial infections in the US . Most Chlamydia infections are initially asymptomatic and therefore unlikely to be treated . However , 5–15% of females with untreated infection will eventually develop serious pelvic inflammatory disease ( PID ) as a consequence . Furthermore , 1 in 6 women who develop PID will become infertile , and many others will develop chronic pelvic inflammation and pain , or suffer from ectopic pregnancy [4]–[6] . The combination of an extraordinarily high number of infections , the asymptomatic nature of initial disease , and the potential for serious reproductive pathology in young women , means that Chlamydia is now recognized as a growing health care problem in the US . The current consensus among scientists and clinicians is that an effective Chlamydia vaccine is urgently needed [7] . The development of an effective Chlamydia vaccine would likely alleviate the burden of Chlamydia on the public health care system . However , the rational design of a Chlamydia vaccine would be aided by improved understanding of the cellular immune response to infection of the female reproductive tract . As Chlamydia is an obligate intracellular pathogen , IFN-γ production by CD4 Th1 cells is essential for protective immunity to primary and secondary infection [8]–[13] . Unfortunately , we have at present only a rudimentary understanding of the development of protective Th1 responses in the context of the female upper reproductive tract and the extent of T helper heterogeneity is unclear . One of the major roadblocks to improving this situation is the lack of antigen-specific reagents that would allow detailed investigation of Chlamydia-specific CD4 T cell responses using a relevant genital challenge model . Previous studies have demonstrated that antibody production by B cells can assist protective immunity during secondary Chlamydia infection [14]–[16] . In contrast , B cells are thought to be dispensable for resolving primary Chlamydia infection , and B cell-deficient and wild type mice shed similar numbers of Chlamydia , as measured by vaginal swabs [17] , [18] . However , another study using the respiratory route of infection demonstrated that intranasal infection with Chlamydia requires B cells for efficient CD4 T cell activation [19] . Therefore , the issue of whether B cells contribute to initial CD4 T cell priming during vaginal infection requires additional analysis . In this study , we generated MHC class-II tetramers to visualize the endogenous CD4 T cell response to systemic and genital tract Chlamydia infection . We show that , unlike intravenous infection , reproductive tract infection is associated with a short delay in the clonal expansion of Chlamydia-specific CD4 T cells in the local draining lymph node . While almost all expanded CD4 T cells expressed the Th1 marker , T-bet , we detected an expanded pool of Chlamydia-specific Tregs that co-expressed T-bet and Foxp3 , and a population of Chlamydia-specific Th17 cells that were specifically enriched in the reproductive tract . In addition , we noted a surprising requirement for B cells in Chlamydia-specific CD4 cell priming within local draining lymph nodes . Loss of local priming in the absence of B cells coincided with bacterial dissemination to the peritoneal cavity inducing inflammatory infiltrate and ascites . Together , these data demonstrate heterogeneity in Chlamydia-specific T helper responses and an unexpected role for B cells in local CD4 T cell priming and bacterial containment within the upper reproductive tract .
In order to develop an overview of the adaptive response to Chlamydia infection , we initially examined the kinetics of bacterial growth and Chlamydia-specific CD4 T cell expansion during systemic infection with Chlamydia . When C57BL/6 female mice were infected intravenously ( i . v . ) with 1×105 inclusion-forming units ( IFUs ) of C . muridarum , the bacterial burden in the spleen peaked around day 4 post-infection and decreased quickly thereafter ( Fig . 1A ) . At day 20 , no infectious Chlamydia was detected in the spleen ( Fig . 1A ) . Consistent with previous findings [20] , a small number of Chlamydia were found in the lung during the first week of systemic infection , but no bacteria were detected in kidney or heart at any time point ( data not shown ) . Numerous C . muridarum MHC class-II epitopes have been uncovered by Immunoproteomic analysis of infected APCs [21] . We used an ELISPOT assay to monitor the frequency of CD4 T cells responding to multiple C . muridarum epitopes after systemic infection . A population of IFN-γ-secreting CD4 T cells responding to RplF51–59 , Aasf24–32 , and PmpG-1303–311 was detected as early as 4 days after infection ( Fig . 1B and C ) . Expansion of IFN-γ-secreting CD4 T cells peaked around day 4–7 , and was followed by a slow contraction of the population over the next 90 days , before a plateau was reached that lasted for at least 352 days ( Fig . 1B and 1D ) . Thus , peak expansion of IFN-γ-secreting CD4 cells closely mirrored peak bacterial burdens in vivo , and stable Chlamydia-specific CD4 T cell memory frequencies were maintained in the absence of active Chlamydia infection . Previous studies have demonstrated that pMHC class-II tetramers can be used in conjunction with tetramer enrichment , to visualize low frequency endogenous antigen-specific CD4 T cells in infected and immunized mice [22] , [23] . We constructed three distinct pMHC class-II tetramers , containing I-Ab with a Chlamydia-specific epitope ( RplF51–59 , Aasf24–32 , or PmpG-1303–311 ) bound to the MHC class-II β chain . Uninfected C57BL/6 mice contained low frequency antigen-specific CD4 T cell population specific for each Chlamydia epitope ( Fig . 2A ) . However , in mice immunized subcutaneously with peptide/CFA , or infected intravenously with C . muridarum , an expanded population of CD44hi CD4 T cells was detectable 7 days post immunization or infection ( Fig . 2A ) . Tetramer staining was specific , as infection with Salmonella Typhimurium did not induce expansion of tetramer-specific CD4 T cells ( Fig . 2A ) . Furthermore , no CD8 T cells were detected that bound to Chlamydia tetramers ( Fig . 2A ) Together , these results demonstrate that all three tetramers , RplF51–59:I-Ab , Aasf24–32:I-Ab , and PmpG-1303–311:I-Ab can be used to detect endogenous C . muridarum-CD4 T cells in vivo . We next used the PmpG-1303–311:I-Ab tetramer to visualize clonal expansion of antigen-specific CD4 T cells during intravaginal ( i . vag . ) infection . We focused on PmpG-1 because CD4 T cell clonal expansion against RplF , Aasf and PmpG-1 is similar ( Fig . 1B , 1C and 1D ) yet PmpG-1 is a promising Chlamydia vaccine candidate [24] . Consistent with previous findings , C . muridarum bacterial loads measured by vaginal swab peaked around day 4 post-infection and steadily decreased until clearance around day 35 ( Fig . 2B ) . To visualize the primary site of endogenous T cell priming to Chlamydia infection , we examined PmpG-1-specific T cell activation in multiple secondary lymphoid tissues . One week after infection , PmpG-1-specific CD4 T cells expanded in iliac lymph nodes and spleen , but were barely detectable in other non-draining lymph nodes ( Fig . 2C and 2D ) . At day 14 , endogenous PmpG-1-specific CD4 T cell expansion peaked in all secondary lymphoid tissues , and was followed by a notable contraction phase ( Fig . 2D ) . The kinetics of the CD4 T cell response to local Chlamydia genital tract infection was therefore markedly delayed in comparison to systemic infection with the same pathogen ( Fig . 1 ) . To examine CD4 T helper differentiation , the expression of lineage-specific transcription factors was examined in expanded CD44hi PmpG-1303–311:I-Ab+ CD4 T cells . Following either systemic or intravaginal infection , almost all PmpG-1-specific CD4 T cells expressed T-bet while no GATA3 expression was detected ( Fig . 3A ) . This is consistent with previous reports that Th1 CD4 T cells are the dominant helper subset following Chlamydia infection [10] , [11] . However , a distinct population of PmpG-1-specific CD4 T cells that co-expressed Foxp3 and T-bet was also detected after i . v . and i . vag . infection ( Fig . 3B ) , suggesting that induced Chlamydia-specific Treg cells are also contained within the expanded CD4 pool . We also utilized RORγt-GFP reporter mice and combined staining with all three Chlamydia tetramers to examine the potential development of Chlamydia-specific Th17 cells after vaginal infection . While GFP-positive cells were undetectable among expanded tetramer-positive cells in the spleen or draining lymph nodes of infected mice , approximately 7% of CD4+CD44hitetramer+ T cells in infected non-lymphoid tissues expressed GFP ( Fig . 3C ) . Furthermore , stimulation of lymphocytes purified from the genital tract confirmed the presence of T cells producing both IL-17A and IFN-γ ( Fig . 3D ) . Thus , Chlamydia infection of the reproductive tract induces a heterogenous T helper response that comprises expanded T-bet+ Th1 cells , T-bet+Foxp3+ Tregs , and Th17 cells that are enriched in infected non-lymphoid tissues . Next , we examined bacterial shedding after vaginal infection of WT and μMT mice with Chlamydia . Consistent with previous reports [17] , [18] , bacterial shedding was unaffected by the absence of B cells ( Fig . 4A ) . However , more detailed analysis of the local draining lymph nodes of μMT mice suggested significantly reduced activation of CD4 T cells ( Fig . 4B ) . Indeed , using the PmpG-1303–311:I-Ab tetramer , we detected much lower clonal expansion of Chlamydia-specific CD4 T cells in the local draining lymph node of infected μMT mice compared to WT mice ( 60±12 in μMT mice vs 166±36 in WT mice , p<0 . 01; Fig . 4C ) . This reduced local response was also accompanied by dissemination of Chlamydia to the spleen and peritoneal cavity , where ascites was noted 14 days post infection ( Fig . 5A and 5B ) . Analysis of ascites fluid from μMT mice revealed a large proportion of macrophages ( F4/80+ ) , monocytes ( Gr-1+ ) and T lymphocytes ( Fig . 5C ) . In addition , PmpG-1-specific CD4 T cells were abundant in ascites , demonstrating that much of the lymphocyte infiltrate into the peritoneal cavity is likely to be Chlamydia-specific ( Fig . 5C ) . Thus , the absence of B cells reduces local CD4 T cell priming and allows bacterial dissemination . In contrast to the local response , polyclonal CD4 T cells in the spleen of μMT mice displayed evidence of increased activation ( Fig . 4B ) . Consistent with increased systemic activation , expansion of PmpG-1-specific CD4 T cells was markedly increased in the spleen of μMT mice ( 33600±6044 in μMT mice vs 5494±1164 in WT mice , p<0 . 001; Fig . 4C ) . Furthermore , Chlamydia-specific CD4 T cells in μMT mice expressed higher levels of T-bet and produced more IFN-γ than CD4 T cells from WT mice ( Fig . 4D and E ) . A greater percentage of multifunctional CD4 T cells producing IFN-γ and TNF-α was also detected ( Fig . 4F ) . Consistent with an enhanced effector response , the percentage of Chlamydia-specific CD4 T cells expressing CCR7 was considerably lower in μMT mice ( Fig . 4D ) . These data suggest a compensatory systemic T cell response is induced to clear the disseminated bacteria that accompany B cell deficiency .
In vivo visualization studies of pathogen-specific CD4 T cell responses have typically involved adoptive transfer of monoclonal TCR transgenic T cells and have rarely focused on sexually transmitted infections [25] , [26] . Here , we describe the generation of peptide-MHC class II tetramers that allow direct visualization of endogenous , polyclonal antigen-specific CD4 T cell responses to Chlamydia infection . Using Chlamydia tetramers and ELISPOT assays , we detected differences in the tempo of the initial CD4 T cell responses to systemic or vaginal challenge with Chlamydia . Unlike other mucosal tissues , the female genital mucosa lacks defined lymphoid structures [7] , which may explain the delayed clonal expansion of CD4 T cells after Chlamydia genital tract infection . Alternatively , this delay may represent an unappreciated virulence mechanism of Chlamydia to delay early T cell priming , as has been noted during respiratory infection with MTB [27] . Thus , delayed CD4 T cell priming in the draining lymph node may be due to limited access to bacterial antigen by tissue dendritic cells or impeded migration to the local lymph nodes . Further studies will be required to examine this issue directly . Our studies show that Chlamydia-specific memory CD4 T cells persist for at least one year after infection , which may suggest the presence of persistent antigen and/or persistent Chlamydia without culturable Chlamydia organisms [28] , after the clearance of infectious bacteria . Interestingly , a recent study by Johnson et al suggested that the PmpG-1303–311 epitope can be detected on splenic APC for at least 6 months after the clearance of primary genital tract infection [29] . Our data , together with others , also support the potential role of PmpG-1 specific CD4 T cells in Chlamydia protective immunity [24] , [29] . Our data confirm directly the previous notion that T-bet expressing Th1 cells are the predominant CD4 effector lineage among Chlamydia-specific CD4 T cells . However , we also identify small populations with the phenotypic marker of Tregs in the lymph node and spleen following both systemic and mucosal infection , as well as the enrichment of Chlamydia-specific Th17 cells at the genital mucosa itself . These data suggest that Chlamydia-specific Tregs are expanded to regulate the large Th1 response that develops during infection [30] . The enrichment of Th17 cells at mucosal surfaces has also been detected in other infection models [23] , [31]–[33] . For example , flagellin-specific Th17 cells are predominantly enriched in the gut mucosal sites after Salmonella oral infection [23] . Further experimentation is required to determine whether these pathogen-specific Th17 cells contribute to Chlamydia clearance and/or the induction of upper genital tract pathology . The presence of large numbers of B cells in the lower and upper genital tract during Chlamydia infection , as shown by both immunohistochemical staining and flow cytometry [34] ( and data not shown ) , led us to speculate that B cells play an important role during primary Chlamydia infection . Indeed , our data show that in the absence of B cells , there is a marked reduction in antigen-specific CD4 T cell priming within the draining iliac lymph nodes . Given this effect of early CD4 T cell expansion it is possible that B cells participate directly in antigen presentation during the early stages of primary infection . Although this has not previously been observed during genital tract infection , a similar finding was reported after C . muridarum lung infection [19] . A recent study has suggested that macrophage deficiency could also account for protective defects of μMT mice against viral infections [35] . However , our observations that ascites did not occur at early time point ( 7 days post infection , data not shown ) suggested that the phenotype we observed is mediated by altered adaptive immune mechanisms . Although mild Chlamydia dissemination to other mucosal sites has been reported previously in C57BL/6 mice [36] , a large number of disseminated Chlamydia has only previously been found in IFN-γ-deficient and SCID mice and largely attributed to deficient Th1 development [8] . B cell deficient mice therefore provide an unexpected additional model where Chlamydia also disseminate to non-mucosa tissues . Notably , our data provides a clear example where the marked pathology of Chlamydia infection does not always correlate with the inability of host to clear the infection . A likely explanation for highly efficient bacteria clearance in μMT mice is the robust systemic CD4 T cell response that may compensate for the loss of initial CD4 T cell priming within the local draining lymph nodes of the genital tract . The fact that Chlamydia genital tract infection can lead to ascites in the absence of B cells is also clinically relevant: Chlamydia infection induces ascites in patients with salpingitis and peritonitis [37]–[40] , although it is unclear whether Chlamydia dissemination in mouse models reflects the same mechanism that leads to symptoms in human such as PID and Fitz-Hugh-Curtis syndrome . Further pathological studies are needed to understand differences of mouse and human infections . Overall , our data demonstrate unappreciated heterogeneity of the CD4 T cell response to genital tract infection in a model where Th1 cells are essential for protective immunity . In addition , our data uncover a surprising involvement of B cells in local expansion of effector Chlamydia-specific T cells in the genital tract and prevention of bacterial dissemination . Greater understanding of the mechanism of Chlamydia dissemination from the genital tract and the T and B cell responses that restrain this spread of bacteria may reduce the risk of sequelae after Chlamydia genital infection in infected women .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The University of California Davis is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) . All animal experiments were approved by University of California Davis Institutional Animal Care and Use Committee ( IACUC ) ( Protocol number 16612 ) . C57BL/6 ( B6 ) mice were purchased from the National Cancer Institute ( Frederick , MD ) and The Jackson Laboratory ( Bar Harbor , ME ) . μMT mice were purchased from The Jackson Laboratory ( Bar Harbor , ME ) . RORγt-GFP reporter mice were obtained from Dr . Marc Jenkins ( University of Minnesota ) . Mice used for experiments were 6–12 weeks old , unless otherwise noted . All mice were maintained in accordance with University of California Davis Research Animal Resource guidelines . Chlamydia muridarum strain Nigg II was purchased from ATCC ( Manassas , VA ) . The organism was cultured in HeLa 229 cells in Dulbecco's Modified Eagle Medium ( DMEM ) ( Life Technologies , Grand Island , NY ) supplemented with 10% fetal bovine serum ( FBS ) . Elementary bodies ( EBs ) were purified by discontinuous density gradient centrifugation as previously described and stored at −80 degree [41] . The number of IFUs of purified EBs was determined by infection of HeLa 229 cells and enumeration of inclusions that were stained with anti-Chlamydia MOMP antibody . Heat-killed EBs ( HKEBs ) were prepared by heating at 56°C for 30 min . A fresh aliquot was thawed and used for every infection experiment . Salmonella enterica serovar Typhimurium strains BRD509 ( AroA−D− ) were kindly provided by Dr . D . Xu ( University of Glasgow , Glasgow , U . K ) . For systemic infection , mice were injected intravenously in the lateral tail vein with 1×105 C . muridarum . To enumerate the bacteria burden in tissues , the spleen , liver and kidney , were crushed in 5 mL SPG buffer and tissue homogenate was placed in a tube with glass beads to disrupt cells . After shaking for 5 min , and centrifugation at 500 g for 10 minutes , supernatants were collected and serial dilutions were plated on HeLa 229 cells . For intravaginal infections , estrus was synchronized by subcutaneous injection of 2 . 5 mg medroxyprogesterone acetate ( Greenstone , NJ ) 7 days before infection . 1×105 C . muridarum in 5 µL SPG buffer were then deposited into vaginal vaults . To enumerate bacteria , vaginal swabs were collected , shaken with glass beads , and serial dilutions were plated on HeLa 229 cells . Spleen and lymph nodes ( axillary , brachial , inguinal and mesenteric ) were harvested and a single-cell suspension prepared . After RBC lysis , CD4+ T cells were enriched using LS MACS columns and anti-CD4 magnetic beads ( Miltenyi Biotec , Auburn , CA ) . Enriched CD4+ T cells were incubated with irradiated APCs from naive mice in the presence of 10 µM Chlamydia peptide ( RplF51–59 , FabG157–165 , Aasf24–32 or PmpG-1303–311 ) in 96-well ELISPOT plates ( Millipore , Billerica , MA ) that had been pre-coated with purified anti-IFN-γ ( BD Biosciences , San Diego , CA ) . The RplF51–59 , FabG157–165 , Aasf24–32 and PmpG-1303–311 epitopes used for stimulation have been described previously [21] , [42] . After 20 h of incubation at 37°C , cells were washed and cytokine spots developed using biotinylated anti-IFN-γ ( BD Biosciences ) , AKP Streptavidin ( BD Biosciences ) , and 1-Step NBT/BCIP substrate ( Thermo Scientific , Waltham , MA ) . Cytokine spots were counted using an ImmunoSpot S5 Core Analyzer ( C . T . L . , Shaker Heights , OH ) , and the total number of IFN-γ-producing CD4+ T cells per spleen was calculated . The methodology for construction of pMHCII tetramers has been described in detail [22] . Briefly , biotinylated I-Ab monomers containing a covalently linked C . muridarum peptide ( RplF51–59 , Aasf24–32 or PmpG-1303–311 ) were expressed by S2 Drosophila insect cell lines and cultured in a Wave Bioreactor ( GE Healthcare Biosciences , Pittsburgh , PA ) . After purification , I-Ab monomers were tetramerized by co-incubated with fluorochrome-conjugated streptavidin at a pre-determined optimal ratio at room temperature for 30 min [22] . To test for tetramer specificity , C57BL/6 mice were immunized with each of the three peptides and CFA ( Sigma-Aldrich , St . Louis , MO ) . Seven days post-immunization , draining lymph nodes were isolated and tetramer positive cells enriched using the methodology described below . Spleen and LNs were harvested from naïve or infected mice . Single cell suspensions were prepared in FACS buffer ( PBS with 2% FCS ) and stained with tetramers in Fc block ( culture supernatant from the 24G2 hybridoma , 2% mouse serum , 2% rat serum , and 0 . 01% sodium azide ) for 1 h at room temperature in the dark . Cells were then washed and tetramer positive cells enriched via LS MACS columns and anti-fluorochrome magnetic beads ( Miltenyi Biotec , Auburn , CA ) . Bound and unbound fractions were stained with a panel of monoclonal antibodies ( listed below ) and analyzed on a FACSCanto or an LSRFortessa flow cytometer ( BD Biosciences , San Jose , CA ) . To stain for intracellular transcription factors and cytokines , cells were left untreated or stimulated with phorbol 12-myristate 13-acetate ( PMA , 50 ng/ml ) , ionomycin ( 200 ng/ml ) in the presence of brefeldin A ( 10 µg/ml ) for 4 h at 37°C . After surface staining , cells were fixed , permeablized and stained using the Foxp3 staining Kit ( eBioscience , San Diego , CA ) . Antibodies for staining included FITC-CD11b , CD11c , F4/80 , B220 , TNFα; PerCP-eFlour710-CD4; APC-CCR7; eFlour660-T-bet; Alexa700-CD44; eFlour450-CD3 , Foxp3 , IFN-γ ( eBioscience , San Diego , CA ) ; FITC-IL-17A and APC-Cy7-CD8 ( BD Biosciences , San Diego , CA ) . Flow data were analyzed using FlowJo software ( Tree Star , Ashland , OR ) and endogenous , tetramer-specific CD4 T cells were identified using a previously published gating strategy [22] . All data sets were analyzed by unpaired Student's t-test using Prism ( GraphPad Software , La Jolla , CA ) . A p value<0 . 05 were considered statistically significant . | Sexually transmitted infections caused by Chlamydia are increasing every year in the US and an effective vaccine is urgently required . Unfortunately , we currently only have a rudimentary understanding of the natural host immune response to Chlamydia infection , especially in the context of the female genital tract . Here , we have developed new reagents that allow direct visualization of the host T cell responses to vaginal Chlamydia infection using a mouse model of infection . These new tools reveal an unexpectedly complex CD4 T cell response to infection and a surprising role for B cells in preventing the spread of bacteria to multiple host tissues . This greater understanding of the host response to infection may eventually allow the construction of an effective vaccine . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | B Cells Enhance Antigen-Specific CD4 T Cell Priming and Prevent Bacteria Dissemination following Chlamydia muridarum Genital Tract Infection |
Eye movements affect object localization and object recognition . Around saccade onset , briefly flashed stimuli appear compressed towards the saccade target , receptive fields dynamically change position , and the recognition of objects near the saccade target is improved . These effects have been attributed to different mechanisms . We provide a unifying account of peri-saccadic perception explaining all three phenomena by a quantitative computational approach simulating cortical cell responses on the population level . Contrary to the common view of spatial attention as a spotlight , our model suggests that oculomotor feedback alters the receptive field structure in multiple visual areas at an intermediate level of the cortical hierarchy to dynamically recruit cells for processing a relevant part of the visual field . The compression of visual space occurs at the expense of this locally enhanced processing capacity .
Our visual experience is derived from a multitude of rapid scanning gaze shifts called saccades . However , perception is even more tightly coupled to saccades than by the mere selection of gaze position . This can be observed in at least three dynamic phenomena that occur time-locked to an upcoming saccade . First , in dual-task experiments that require discrimination of an object and the execution of a target-directed saccade , it has been observed that visual discrimination is best when the discrimination stimulus is located at the saccade target [1 , 2] . Second , receptive fields dynamically change their position and shape . In area V4 , receptive fields tend to shrink and shift towards the saccade target [3] . In many other areas such as V3a , LIP and FEF , a receptive field translation along the saccade vector has been described [4–7] . The latter observation is commonly referred to as remapping , since neurons begin to fire towards a stimulus located in the future , post-saccadic receptive field . Remapping has been suggested to play an essential role in visual stability , i . e . our subjective experience of a stable world despite the change of the retinal image with every saccade [4 , 7] . Third , around saccade onset briefly flashed objects are seen close to the saccade target . This transient distortion of perceptual geometric relationships has been termed peri-saccadic compression [8–11] . There is a general agreement that these phenomena depend on extraretinal signals . Their precise link to particular extraretinal signals , however , is unknown . Among those extraretinal signals is corollary discharge , a copy of a motor command that is sent to the perceptual pathways of the brain . For example , the corollary discharge from the superior colliculus to the frontal eye field encodes the saccade target information , i . e . , saccadic eye displacement [12] . Corollary discharge has been primarily associated with the remapping of receptive fields to construct a continuously accurate , retinocentric representation of visual space [4–7] . This remapping of receptive fields , however , would require an extraretinal signal that is distributed across the whole visual space changing the effective connectivity of neurons in retinotopic maps [13] . Another extraretinal signal of the oculomotor system codes for eye position [14] . Information about eye position is crucial for coordinate transformation from a retinocentric to a head-centered reference frame by tuning the response selectivity [15 , 16] . Localization errors of stimuli flashed in total darkness , known as uniform peri-saccadic shift [17 , 18] , suggest that the eye position signal is erroneous around a saccade [19] . The mislocalization of brief flashes in direction to the saccade target , the peri-saccadic compression [9 , 11] , is virtually not understood , but it has been attributed to a translation in cortical coordinates [20] or a stretching of receptive fields [21] . The facilitated visual discrimination at the saccade target position is usually interpreted as the result of spatially focused visual attention . Presumably , attention-related extraretinal signals during eye movements lead to an enhanced response of neurons that encode a target object selected for saccade [22] . The processing in parietal and frontal cortex has often been associated with attentional spatial selection—the source of spatial attention [23–25] . Here , for the first time , we develop a computational theory of peri-saccadic vision that explains three of the mentioned peri-saccadic phenomena: the enhancement of visual discrimination at the saccade target , the shift of receptive fields , and peri-saccadic compression . Basically , we will demonstrate that these three phenomena can be linked to a single neural mechanism . Our proposed theory assumes that corollary discharge , or more general , a plan to move the eye , is used to transiently boost visual performance at the target location of the saccade immediately before the saccade . While this performance boost is beneficial for visual discrimination peri-saccadic compression is a direct consequence of it , and thus a cost to pay .
Early to mid-level visual processing is organized in retinotopic maps in many brain areas . Likewise , saccadic targeting information is organized in visuo-motor maps in cortical ( frontal eye field ) and subcortical ( superior colliculus ) structures . In our model , saccade target information is sent back as an oculomotor feedback signal from visuo-motor maps to visual-spatial maps in topographic correspondence and is only available around saccade onset . Each single visual map consists of an input stage with “simple” cells for feature detection , a spatial pooling stage with “complex” cells to obtain increasing spatial invariance [26 , 27] , and an intermediate gain stage at which the oculomotor feedback signal acts . This feedback signal increases the gain of the visual responses of the neurons as observed in electrophysiological studies [25 , 28] . A hierarchy of visual processing is then obtained by simulating multiple layers where the mechanism of gain modulation acts in each additional layer ( Figure 1A ) . A visual stimulus initially exerts a corresponding activity hill on the cortical surface of layer 1 ( L1 , Figure 1C , left ) . The position and shape of this activity hill is determined by the magnification factor ( Figure 1B ) and the receptive field sizes of the neurons in the simulated area . Prior to an eye movement , activity increases at the location of the saccade target in the oculomotor map ( Figure 1C , top ) . The feedback of this activity distorts the population response of the flashed stimulus towards the saccade target ( Figure 1C , center ) . By assuming that the visual system relies on this population response for stimulus localization , we can decode the perceived position ( Figure 1C , right ) . Figure 1D illustrates the underlying mechanism of mislocalization in detail . Each panel shows the population activities in the input ( gray ) , feedback ( blue ) , and gain ( red ) layers along a horizontal stretch of visual space . The leftmost panel depicts the case when the flash is presented 150 ms before the saccade . Input and gain layer activities are identical since there is no feedback signal at this point in time . The flash exerts a distribution of activity over the entire population that peaks at the position where the flash was presented ( 10° ) . The perceived position is decoded from the distribution of the population activity by a template matching procedure ( see “Decoding” in Methods ) . The decoded position ( red vertical line ) is identical to the true flash position . The three panels to the right depict the interaction between feedback and gain layers for flashes presented at three time points before saccade onset ( −40 ms , −30 ms , and −20 ms ) . Over this time course , the feedback signal ( blue ) rises in strength but is always centered at the saccade target position at 20° . In the gain layer ( red curve ) the responsivity of the neurons near the saccade target increases and the shape of the population activity is distorted . The decoding of the perceived flash position shifts the perceived position ( red vertical line ) gradually away from the true position ( gray vertical line ) and towards the saccade target . As the strength of the feedback signal increases as time gets closer to saccade onset , the strength of mislocalization of a flash presented at that particular time increases as well . For flashes presented spatially beyond the saccade target , the mislocalization would be in the opposite direction , and again towards the saccade target . Mathematical details of the model are described in the Methods section . In the simulations we found that a model with a hierarchy of only two gain modulated layers ( L1 , L2 ) with increasing receptive field sizes is consistent with three particularly relevant experimental data sets of peri-saccadic localization: the spatial range of compression [9] , the time course of compression [9] and the spatial pattern of compression [11] ( Figure 2 ) . We estimated the goodness of fit by the proportional reduction in error measure ( pre ) , which is the reduction in the sum of squared error ( SSE ) of the data by the model ( section “Proportional reduction in error measure” in Methods ) . The model shows strong compression in the range of ±20° around the saccade target ( Figure 2A ) . The mislocalization originates in L1 for stimuli flashed close to the saccade target and in L2 for stimuli flashed further away . The effect occurs prior to saccade onset and ceases during the saccade ( Figure 2B ) . Mislocalization of small stimuli occurs also orthogonal to saccade direction ( Figure 2C ) as the feedback signal acts on the two-dimensional cortical surface . However , only a model of anisotropic cortical magnification in L1 results in an adequate fit to the data for all four saccade amplitudes ( Figure 2C ) . The presented model is the first neural explanation that accounts for the essential data of peri-saccadic compression . Because we put much emphasis on neuroanatomical and physiological details , the model , as defined by the parameter fit to the available data , can provide quantitatively testable predictions . Since most of the relevant anatomical data is not well known in humans we primarily relate to investigations with monkeys . One prediction is concerned with the origin of the feedback signal . Since the temporal dynamics of compression requires a particular time course of the activity in the oculomotor map , we can use this constraint to predict the origin of the feedback signal . The frontal eye field ( FEF ) shows a continuum of saccade-related cells ranging from a strong visual to no visual response [29] . Similarly , some cells in the superior colliculus ( SC ) initially slowly build up their activity and others show a burst of activity only around saccade onset [30] . The movement fields of saccade-related cells in FEF and SC can be closed and open-ended [12 , 29 , 30] . Cells with closed movement fields fire only when the saccade amplitude is around the optimum for that cell , whereas with open movement fields a cell continues to discharge also for larger saccade amplitudes . Furthermore , saccade-related cells are clipped , partially clipped or unclipped [30] . The discharge has been classified as clipped if the activity drops close to baseline by the end of the saccade . Although several neurons with open movement fields , primarily of a build-up type , can be found in SC , the majority of burst cells has closed movement fields and a clipped activity profile [30] . We systematically varied the shape and time course and fitted the model to the data , showing the time course and spatial range of compression , by adjusting the other parameters of the model with the constraint that the model remains consistent with the spatial pattern of compression . The model predicts that the main contribution originates from cells with closed movement fields and clipped discharge ( Figures 3A and S1 ) . Open movement fields systematically reduce mislocalization of stimuli flashed beyond the saccade target , since the feedback signal now shows a weaker spatial gradient for larger eccentricities ( Figure 3B ) . The activity in the oculomotor map should exceed its half-maximum value not earlier than 30 ms prior to saccade onset , which is consistent with the firing pattern of burst cells . However , this value depends on the assumption that the gain is instantaneous , i . e . , even a low activity of the cells in the oculomotor map leads to a significant gain increase . We tested the model also with a damped gain function with little increase at the target site for low oculomotor activity ( see “Gain Modulation” in Methods ) , and observed that the half-maximum activity can occur much earlier . Thus , whereas an instantaneous gain function requires that the feedback signal primarily originates in oculomotor burst cells , a damped gain function allows that build-up , and/or visual activity contributes to the feedback signal . In both cases , however , the effective feedback signal would be primarily driven by saccade-related activity , since the early prelude activity would have little impact on the gain . The feedback signal in the model represents the contribution of many cells which , in experimental data , have to be combined with respect to their firing rate and their movement field to interpolate the activity distribution on the cortical surface [31] . Our prediction about the spatiotemporal properties of the feedback signal could be tested by calculating detailed spatiotemporal activity distributions in the SC and in the FEF for a given a saccade amplitude . Another prediction is concerned with the shape of the feedback signal in visual space . We modeled the feedback signal as a Gaussian in cortical space similar to collicular neurons with closed movement fields [12 , 30] . Under this assumption our model predicts an anisotropic magnification in early visual areas . This qualitatively resembles findings in striate cortex of monkey [32 , 33] and human V1 and V2 [34] . As a consequence of this anisotropic magnification the feedback signal appears elongated in visual space ( Figure 3C ) . This prediction could be tested by estimating the shape of the oculomotor feedback signal in spatially arranged visual maps with fMRI . Anatomical and physiological investigations revealed widespread connections from the oculomotor system to extrastriate visual areas that list these target areas as candidates for participating in compression . The SC has indirect projections to visual and frontal areas via the thalamus [35] . The FEF is linked with V2 , V3 , V3a , V4 , MT , MST , FST , VIP , LIP , V4t , TEO and TE [36–39] . The FEF projections to these areas appear to be topologically organized in terms of saccadic amplitude [37] , as required by the model . A gain increase has been observed in V4 cells after a stimulation of the FEF using currents below the level that evoke a saccade [25] . Subthreshold stimulation in the SC also induces a shift of attention and an increase in visibility at the motor field of the stimulated site [28 , 40] . In addition to these anatomical and physiological considerations we can formulate stronger constraints on the involved areas by tuning the parameters of the model to the minimal possible receptive field size and compare it to the receptive field sizes of several areas in question ( Figure 3D and Text S2 ) . For the strong compression in the spatial range of ±20° around the saccade target ( Figure 2A ) the model requires at least a receptive field size as observed in areas V4 , MT , or TEO , alternatively in V3a as well . The receptive field constraint of L2 is consistent with the receptive field sizes found in TE and LIP . Too small receptive field sizes , e . g . , at the level of V2 for layer 1 and between 10° and 20° for layer 2 , still allow to fit the data from flashed bars close to the saccade target , but effects from those flashed at a larger distance cannot be accounted for ( Figures 3E and S2 ) . The reason is that with a small receptive field size the population response becomes too narrow to be affected by the feedback signal so that the spatial range of strong compression is reduced to less than ±10° . Increasing the width of the feedback signal is not a solution . A broader feedback signal would increase the gain of the whole population to a similar degree . However , a mislocalization only occurs when the population is distorted which requires a difference in the gain across the population . Thus , a broader feedback signal would increase the range of compression , but the amount of compression would be reduced ( slope of the line through ( 10° , 10° ) in Figure 3E would approach 1 ) . We next turn to the predicted receptive field dynamics in the model and their relation to peri-saccadic receptive field changes observed in different brain areas [3 , 4 , 6 , 7] . To determine the receptive fields of model neurons we calculated the spatial borders of the half-maximum response , as is commonly done in neurophysiological experiments . We determined the receptive fields in two conditions , pre-saccadic and peri-saccadic prior to the eye movement . We find combinations of shift , shrinkage and also expansion of receptive fields ( Figure 4A ) . For receptive fields above or below fixation , locations that have been commonly used to investigate remapping , the model shows peri-saccadic shifts of receptive fields similar to those observed in V3a , LIP and FEF [4 , 6 , 7 , 41] ( Figure 4B , e . g . , a model L1 , pool cell with receptive field center at ( −7° , −16° ) ) . For cells with receptive fields located above or below the saccade target , the model predictions differ from remapping ( Figure 4B , e . g . , a model L1 , pool cell with receptive field center at ( 20° , 20° ) ) . Whereas remapping predicts a change into the direction of the saccade ( towards p2 ) , for this receptive field , our model predicts a change towards the saccade target ( p3 ) , similar to observations made in area V4 [3] . Having demonstrated that the proposed model is consistent with the essential data of peri-saccadic compression—the cost side—we now ask for the benefit of oculomotor feedback . A planned saccade increases the gain of cells with receptive fields around the saccade target , similar as observed physiologically [25 , 42 , 43] . This increase in gain enhances the sensitivity of the cells and when multiple stimuli are present within a single receptive field , it can bias the competitive interactions among stimuli to suppress the influence of the unattended ones [44 , 45] . Whereas the link of the processing in oculomotor areas to changes in gain of cells in visual areas is now well established the macroscopic effect of receptive field dynamics is unclear . In order to provide an estimate of the joint effect arising from all receptive field changes , we define visual capacity as the number of cells which process a particular , small part of the visual scene , as determined by their half-maximum response . A higher visual capacity could potentially allow us to reveal finer details of objects and thus facilitate recognition . Due to cortical magnification the capacity of visual processing is not evenly distributed , since most of the cells are devoted to process the central part of the visual scene . To estimate peri-saccadic capacity effects we compared the capacity distribution during fixation with the one around saccade onset ( Figure 4C ) . We observed a capacity increase around the saccade target of more than 100% in L1 , pool . In L2 , pool the model shows a slight increase in capacity almost across the whole visual half-field that contains the saccade target . Thus , our model predicts that areas at an intermediate level of the hierarchy tune their feature detectors to encode aspects of objects located close to the saccade target whereas higher levels are more broadly tuned to the whole visual half-field .
Our model explains peri-saccadic compression , receptive field shifts , and a visual capacity increase by the same mechanism , i . e . , a spatially selective feedback signal that encodes the saccade target . The feedback signal may be provided by the oculomotor system as a corollary discharge [7] or , more abstractly , as a plan to move the eye [45] . This attentional explanation of peri-saccadic compression appears at odds to the explanation by remapping . Although attention can be covertly shifted to locations other than the target location of an upcoming saccade , it is generally accepted that spatial attention is locked onto the saccade target just prior to saccade onset [1 , 2] . Our model predicts that the effective feedback signal is driven by saccade related activity and thus it supports a premotor view of spatial attention [46] . However , the term premotor has never been clearly defined in the literature . From our point of view , premotor does not imply that the target selection has been finalized . We now know that the FEF and the SC contain a continuum of visuomovement cells from little to strong movement related activity . If these cells are the primary source of feedback there appears sufficient room that the net-signal is movement related in tasks that require an eye movement and that visual cells in the oculomotor pathway contribute to the net-signal and feed back to mid level visual areas , as suggested by a predominantly visual-selection hypothesis of spatial attention [47 , 48] . Our study supports our previously formulated reentry hypothesis of spatial attention [45 , 49] . Indeed , the present model is an anatomically more precise implementation of our previous model while dropping some details with respect to the temporal dynamics of competitive recurrent interactions . It has been suggested that covert attention could be implemented as a planned but not executed saccade [46 , 50] . If this assumption of covert attention is true , the plan to move the eye would be already sufficient to distort the population response . Thus , our model would predict that it should be possible to observe compression when an eye movement plan is aborted prior to its execution . Moreover , our model would predict a pattern of receptive field changes similar to that shown in the peri-saccadic case . This prediction of our model is supported by the observation that also covert shifts of attention resulted in a shift of V4 and MT response profiles [51 , 52] . However , a covert attentional signal may be less strong than one immediately before saccade onset and the resulting compression may be small . Besides compression , caused by oculomotor feedback , other factors might also influence the pattern of mislocalization . In the experiments , subjects have to report the perceived position after the eye movement has taken place . Thus , they must take the saccade into account to avoid a systematic offset in their location estimate and use additional retinal or extraretinal cues , such as an extraretinal eye position [53–57] , a prior assumption about stability [58] , or the relative distance to stimuli that can be used as landmarks [59–61] . The usage of this additional information allows us to compute an eye-movement invariant stimulus position ( with respect to the limits of the additional information ) presumably in a second processing stage [62] . If compression relates to an oculomotor feedback signal encoding the saccade target , why is mislocalization in total darkness predominantly characterized by a shift into the direction of the saccade with only little compression [10 , 18 , 19] ? Whereas under normal conditions the relative distance to landmarks can be used for localization , experiments in total darkness presumably require the usage of an extraretinal eye position signal . While it appears well established that the extraretinal eye position signal does not allow for a perfect on-line correction of the retinal shift , the missing compression appears puzzling . Since there is no obvious reason to postulate the absence of the oculomotor feedback signal in darkness , one would expect compression also in total darkness . However , at least two factors reduce or even diminish compression . First , experiments in total darkness require memory guided saccades or at least saccades with less visual guidance . In memory guided saccades movement related neurons typically fire less vigorously [63] and thus , the gain increase should be reduced . An indirect link between the activity of movement cells and the amount of peri-saccadic compression is also suggested by the correlation of peri-saccadic compression with saccadic peak velocity , since the peak velocity depends on the activity of movement related cells [64] . Second , in our model the gain increase depends on the stimulus strength , consistent with the observation that the magnitude of compression decreases with increasing contrast [65] . Thus , the model predicts weak or diminished compression in total darkness , if stimulus luminance is high . Indeed , cell recordings in V4 and MT revealed that the gain enhancement due to shifts of spatial attention is limited for high contrast stimuli [66 , 67] , although the exact gain function is debatable [68] . It is not clear if our predicted oculomotor feedback signal is identical to the signal causing shifts of spatial attention in those experiments , but the mechanism of gain enhancement might be independent of the source of the modulatory signal . However , for low luminance stimuli the model predicts compression , even in total darkness . We recently tested this prediction experimentally and found compression in total darkness for stimuli with near-threshold luminance [69] . Thus , compression can also be observed in total darkness , as predicted by the model , if stimuli are presented at low visibility . The process of determining the position in world-centered coordinates may also influence the mislocalization effects observed after saccadic adaptation . After saccadic adaptation , it has been observed that peri-saccadic compression is directed to the adapted end point of the saccade [70] , whereas the activity in the SC appears to encode the initial , unadapted location of the saccade target [71] ( see however [72 , 73] ) . There are two possibilities to reconcile this observation with our model . First , it may be possible that the pre-saccadic compression is directed to the unadapted goal location and that it is subsequently shifted towards the post-saccadic gaze direction by adaptation specific spatial transformations . This is supported by observations of general shifts of perceived visual location induced by saccadic adaptation outside the time interval for compression [74] . Second , more monotonic adaptation techniques could lead to cognitive changes in the saccade plan so that the feedback signal is indeed pre-saccadically directed towards the adapted end point of the saccade . This has recently been shown by observations of mandatory pre-saccadic allocation of attention towards the adapted end point after saccadic adaptation [75] . The model makes clear and strong predictions about the putative involved areas with respect to the receptive field size . We can restrict the origin of the strong compression of ±20° around the saccade target to intermediate levels of the cortical hierarchy . The observed dissociation that much less compression is found for pointing movements with closed eyes than for verbal reports of the perceived position [76 , 77] can be explained by different pathways for perception and for pointing . Online reaching and pointing movements recruit the “dorso-dorsal stream” [78] consisting of the forward projection V1 to PO to MIP and V2/V3 to V6/V6a to MIP and further to supplementary motor areas [79 , 80] . This stream has not been reported to receive significant feedback from the lateral FEF [37] . Thus , consistent with observations in MT/MST [81] , our model predicts that the encoding of a stimulus position is already distorted in a retinocentric reference system presumably at the levels of V3a , V4 , TEO , MT/MST and LIP . Fitting the data of peri-saccadic compression predicts a specific pattern of receptive field dynamics . This linkage of psychophysical data to their underlying neural brain processes is a particular strength of our approach . The cells shown in Figure 4A and 4B exemplify the fact that the receptive field effects in our model are dependent on the relative locations of fixation , saccade target , and center of the receptive field . The similarity of our model observations with studies in different brain areas raises the fundamental question about the nature of peri-saccadic receptive field changes . We demonstrated that our model predicts the remapping of receptive fields for receptive field positions that have been commonly used to investigate remapping [4 , 6 , 7 , 41] . For other locations , however , the model is consistent with observations made in V4 where receptive fields tend to shift towards the saccade target and not along the saccade vector [3] . Do V4 receptive field dynamics differ from other areas ? For example , does remapping occur primarily in oculomotor-related areas whereas our model describes properties of areas involved in the computation of object identity ? Or is remapping not homogeneous across visual space , but a special case that applies only within a certain part of the visual field ? A non-homogeneous remapping could reconcile the different observation made in V4 . No study has yet systematically addressed this question . Such systematic investigations of 2D receptive field dynamics in different brain areas are required , specifically in those which receive oculomotor feedback , e . g . , area V4 , MT , MST , V3a , TEO , LIP and VIP . Our model suggests that the transient receptive field changes serve an increase in processing capacity around the saccade target . This phenomenon has not been an integral part of earlier attention theories and offers an alternative to the common attentional spotlight metaphor . According to the spotlight metaphor the width of the focus must be properly tuned to the size of the object to which attention is directed , since processing outside of the spotlight is weak . A change in the processing capacity may offer a more robust solution since more neurons are available to process details of the object at the saccade target . Under certain assumptions one can show that the increase of the number of cells within a population improves the accuracy of coding [82 , 83] . However , an improvement in object recognition must be investigated with more elaborated future models . Indeed , if we took into account that each cell in the model layers is sensitive for a specific feature at a certain position in the visual field , the model would predict a shift in the spatial arrangement of feature detectors at an intermediate level of recognition . This suggests that the structure of objects , as determined by the feature detectors in each brain area , remains uncompressed but the position of an object is subject to change . This is consistent with the observation that the shape of a single object is not or much less distorted [84 , 85] . What mechanism is then responsible for the perception of a stable word ? Although remapping has been suggested to lead to the perception of a stable environment [4 , 7] , a global anticipatory shift of receptive fields might not be necessary . Perhaps the brain does not even attempt to maintain a continuous retinocentric representation of visual space [86 , 87] . In this regard , compression is not used for a correction of peri-saccadic artifacts to realize a stable , spatially correct representation of the external world . We rather suggest that the anticipatory processing of the object of interest at the saccade target position leads to the perception of a stable world , since we already deal with the object of interest before we even look at it [88] . Several findings , such as saccadic suppression [21] or saccadic suppression of image displacement [89] suggest that , under normal viewing conditions , we make little use of the retinal image during eye movements , but we primarily use information in the pre- and post-saccadic scenes [59 , 61] . Thus , oculomotor feedback could be essential to link the pre-saccadic representation with the post-saccadic one . First , oculomotor feedback reactivates the pre-saccadic representation of a stable stimulus at the saccade goal which otherwise would decay close to baseline [43 , 90] . Second , a strong increase in the visual capacity around the saccade target may reveal details of the object that will otherwise only be seen when the eyes land . While the representation of space and thus the perception of a stable world remains an open issue , for now a useful working hypothesis is that the brain deals with the pre-saccadic representation while the eyes move . The peri-saccadic change in the firing pattern of many early to mid-level cells could be negligible , since the pre-saccadic representation is cognitively linked with the post-saccadic one by attention and processing is focused on specific aspects of the scene . In conclusion , each saccade is accompanied by an oculomotor feedback signal , which is conveyed to mid-level visual areas and enhances the gain of cells located around the saccade target . Such gain increases lead to an advantage for the processing of stimuli located at or near the saccade target such that they are represented more actively . Moreover , the population response for stimuli presented around the saccade target is distorted . From the viewpoint of a single cell , oculomotor feedback increases its gain and alters its receptive field . On a macroscopic level the changes in receptive field size and location dynamically increase the processing capacity around the saccade target . The spatial mislocalization occurs whenever the brain must rely on the distorted population response in this period to generate a holistic impression of an object in space .
Our model of peri-saccadic perception aims at linking psychophysical data with their underlying brain processes . Although the general idea of the model is very simple , our emphasis on certain neuroanatomical and physiological details requires some advanced techniques . The consideration of important neuroanatomical and physiological details has several advantages over more simple approaches ( compare [20 , 90] ) . This added detail does not primarily serve a better fit; it rather provides a more meaningful constraint . Moreover , the model has more predictive power in the sense , that the obtained parameters are meaningful with respect to a specific cortical function . The model consists of two visual , hierarchically organized layers L1 and L2 . The computation in each layer is divided into three stages . The first stage represents the input from earlier areas . The second stage implements a gain modulation of the input , and the third stage pools the responses to obtain increasing spatial invariance . The spatial pattern of compression obtained from a flashed small dot randomly chosen from an array of 24 dots [11] cannot be reproduced by a model in which the feedback signal is defined as a Gaussian in visual space . The consideration of cortical magnification not only allows us to quantitatively fit the data , it also provides us an estimate of the shape of the feedback signal in visual space . Moreover , it affects the direction of dynamic receptive field changes . Magnification changes along with spatial pooling to account for the fact that higher areas typically show a less pronounced magnification at the fovea . The cortical space is mathematically described as a curved surface [91] . The shape of this surface depends on the changes in cortical magnification along the horizontal meridian of the visual field Mp ( ɛ ) and along isoeccentric rings Me ( ɛ ) , where ɛ denotes eccentricity ( see section “Cortical space” for details ) . Let V be the visual space , the cortical space of the input into L1 , the cortical space of the gain modulated stage and the cortical space of L1 , pool . We use Gaussian functions to model the receptive fields . Let ∈ V be the position of the receptive field center , i . e . the point in visual space which maximally activates the cell i in L1 . determines the width of the receptive field as a linear function of eccentricity ( ( ɛ ) ) . Let ps∈V be the position of the flashed stimulus in the visual field . For simplicity , we ignore the stimulus width . The activity of a given L1 , in cell is then defined by Note that denotes the distance between the receptive field center and the stimulus position . k is a constant which relates to the contrast/luminance of the stimulus at the time of the flash . After saccade onset the retinal position of the stimulus is computed according to the position of the eye in space . Our model of eye movements is given in section “Simulation of eye movements . ” Formally , the gain modulated response can be described by a sensitivity increase of a cell i to its input dependent on the oculomotor feedback signal . As derived in section “Gain modulation , ” the activity of a given L1 , gain cell i is defined as a function of the input , the gain and a term which normalizes the activity: The weight factor w is equal for all layers . denotes the feedback signal . The feedback signal could have its origin in an oculomotor map in which an activity hill is built up around the target location of a planned eye movement . The oculomotor map is fully connected with L1 and L2 . The feedback signal from the oculomotor map to L1 and L2 is determined as a Gaussian in cortical space which changes in amplitude through time: where denotes the cortical position of the cell i in L1 , in and cST ∈ denotes the center of the feedback signal in cortical coordinates . The center of the feedback signal could be an independent variable linked to the saccade plan , but in all simulations performed , we assume that it is equal to the experimentally defined saccade target . denotes the distance between the position of a given L1 , in cell and the saccade target in cortical space . Our model of cortical space is given in section “Cortical space” and the computation of distance in cortical space is explained in section “Distance measurement on the cortical surface . ” is the saccade amplitude dependent ( SA ∈{12° , 16° , 20° , 24°} ) width of the feedback signal . The assumption of a gradual spatial decrease of the feedback signal relative to the saccade target is supported by a recent observation in V4 using below threshold microstimulation in the frontal eye field [44] . In this study , the increase of the separation between the saccade endpoint ( as determined by above threshold stimulation ) and the stimulus in the receptive field of a cell resulted in a decrease of the enhancement effect . Thus , the closer the stimulus is to the saccade target , the stronger is the gain increase by microstimulation . Please note that the predictions of the model do not depend on a retinotopic projection as long as the connections between oculomotor areas and the visual areas correspond with each other in visual space . As far as the temporal characteristics f ( t ) are concerned , the activity hill should increase for t ≤ 0 , and decrease for t > 0 , where 0 represents the onset of the saccade . Thus , the strength of the feedback signal is maximal for a stimulus flashed at saccade onset . The center of the feedback signal moves with the eye , i . e . , it remains at its original position in a retinocentric coordinate system . In order to test how far the feedback signal relates to the typical time course of movement-related cells in the frontal eye field or the superior colliculus , we systematically varied the time course and shape of the signal ( Text S1 ) . For example , for f ( t ) we used an exponential: and Gaussian function: where α determines the increase and β the decrease of the activity over time in both cases . We do not explicitly simulate the visual latency of a stimulus to reach an area of interest nor the delay of the oculomotor signal . If we assume that the latency of a stimulus to reach V4 is longer than the delay of the oculomotor signal to reach V4 , our feedback signal should have its peak after saccade onset . However , a flashed stimulus has a neural persistence of 100–150 ms , which implies that it is not necessary that the feedback signal must be present at the very first response . Thus , the time course of our feedback signal will appear still plausible if we consider visual latency and persistence . The activity of a given L1 , pool cell j is determined by pooling the gain modulated input activities of the respective L1 cells . The classical receptive fields of L1 , pool cells are incorporated into the model using Gaussian functions . The activities of L1 , pool cells are weighted with respect to the distance in the visual space between the receptive field center of the cell i in L1 input and the receptive field center of the L1 , pool cell . These weighted L1 cell activities are then spatially pooled using a max operation [92] where relates to the width of the receptive field . These receptive field kernels only indirectly define the final receptive field size of each layer . Thus , our given estimates of the receptive field size were obtained from mapping the receptive fields ( section “Mapping of receptive fields” ) . The activities in L2 are computed equivalently using Equations 1–5 . With the above methods we can calculate the population response of model neurons to a visual stimulus . In order to relate the perceived stimulus position of the model to experimental data , we decoded the population response with respect to location ( section “Decoding” ) . Our model predicts that a compression of space perception is caused by a local increase in the processing capacity . This might appear as a paradox to some readers , since if the receptive fields shift towards the saccade target , the same position in space now activates neurons with receptive fields farther from the saccade target . However , this does not lead to an expansion of space , since the change of the receptive fields is not uniform and other neurons with receptive fields closer to the saccade target still respond to the stimulus . In addition , the neurons closer to the saccade target increase their sensitivity more than the ones farther away . Thus , across the whole population the neurons closer to the saccade target vote stronger , even if the ones farther away shift their receptive field closer to the saccade target . In order to focus on the localization error predicted by the oculomotor feedback , we do not consider any additional errors due to the mapping of a retinal coordinate system into a world centered coordinate system [19 , 56] . Thus , we add the position of the eye θ ( section “Simulation of eye movements” ) to the estimated stimulus position in retinal coordinates to obtain the estimated stimulus position in the world-centered space . We simulated more than 48 , 000 cells per layer which were equally distributed in cortical space up to 70° eccentricity . We iteratively determined the parameters and the number of layers in the model ( section “Fitting procedure and parameters of the model” ) from three particularly relevant experimental data sets: the spatial range of compression [9] , the time course of compression [9] and the spatial pattern of compression [11] . In our model we use static neurons . We here derive the equation of gain modulation for static neurons from an equation used for dynamic neurons . Let us therefore assume , we have a set of gain-modulated neurons . The firing rate of each neuron can be described by a differential equation [45 , 92] Such a gain function is motivated by several electrophysiological studies which have shown that feedback signals have a modulatory influence [25 , 93] and it has successfully been applied to model the effect of feedback connections on feedforward processing [45] . The term ensures that the efficiency of the feedback signal depends on the activity of the postsynaptic cell population . If the maximal firing rate exceeds the value A , the feedback signal no longer affects the gain . This term has been shown to be consistent with a multiplicative contrast gain modulation as observed in several single cell recordings [94] . When we numerically compute the firing rate and set the weight of the dynamic inhibition among the cells to winh = 0 , the change of activity in each time step is When we ensure that and further approximate we obtain for the equilibrium a non-recursive equation for the firing rate of the gain modulated neurons ( Figure 5A ) : This equation for the firing rate of a static , gain-modulated neuron is of course not equal to the dynamic , recursive solution , but it captures the essentials as verified by simulations . We alternatively used the following damped gain function to explore the source of the feedback signal ( Figure 5B ) : The damped gain function only leads to small changes in gain for low feedback activity . Neurophysiological findings in monkeys and humans indicate that central parts of the visual field are processed by a greater amount of cortical tissue as compared to peripheral parts [95 , 96] . The amount of cortical tissue , which processes one degree of the visual field , is termed the cortical magnification factor and is usually denoted in millimeter per degree [97] . For the mapping of the visual field V into cortical space C a procedure by Rovamo and Virsu [91] has been used according to which the cortical space is a topologically isomorphic distortion of the visual space , i . e . , a transformation of a sphere . The visual space is described in spherical coordinates ( ɛ , φ ) and the cortical space is described in cylindrical coordinates ( r , z , θ ) . Note that the angle ɛ is the eccentricity generating the meridians and the angle φ generates the circles of constant eccentricity . To obtain the cortical representation C of the visual field V , the sphere is transformed according to the two cortical magnification functions Mp ( ɛ ) and Me ( ɛ ) . Mp ( ɛ ) describes the changes in cortical magnification along the meridians of the visual field and Me ( ɛ ) along the circles with constant eccentricity . If both functions are equal , the cortical magnification is isotropic , i . e . at each location in the visual field magnification along a circle of constant eccentricity is equal to magnification along a meridian . The magnification function Mp ( ɛ , φ ) along the meridian is defined by and the magnification function Me ( ɛ , φ ) along the circles of constant eccentricity is defined by It is assumed that the cortical space is rotationally symmetric ( θ = φ ) , i . e . , the magnification functions do not depend on φ . Solving Equations 6 and 7 yields the complete transformation rule and Figure 6 shows the cortical space of L1 and L2 . Please note that the difference in size ( surface ) between the isotropic and anisotropic case is not relevant for the different predictions . The main factor leading to the stronger asymmetry of the compression pattern are the longer distances along the rays compared to the ones along the circles . The degree of overrepresentation around the fovea is not crucial for the results obtained . Please note that the assumed anisotropy across the whole visual field is a simplification . We do not claim that the whole human visual field is subject to anisotropy . Since we describe the oculomotor feedback signal as a Gaussian in cortical space , the distance between the center of the signal and the cortical position of each cell is required . In order to compute the distance one has to consider that the cortical space is a curved surface and the distance between two points is the length of the geodetic line connecting the two points . The geodetic line is the solution of the following variation problem . Let s be a real number running from 0 to 1 , g ( x1 , x2 ) be the metric tensor of the surface with respect to the local coordinates ( x1 , x2 ) , and let ( x1 ( s ) , x2 ( s ) ) be a path connecting the points ( x1 ( 0 ) , x2 ( 0 ) ) and ( x1 ( 1 ) , x2 ( 1 ) ) . Finding the geodetic line is done by minimizing by variation over the possible paths ( x1 ( s ) , x2 ( s ) ) connecting the points ( x1 ( 0 ) , x2 ( 0 ) ) and ( x1 ( 1 ) , x2 ( 1 ) ) . S is the length of the path , and is called the Lagrange function of the variation problem . The solution of this variation problem is equivalent to the solution of the system of differential equations: with respect to the boundary conditions ( x1 ( 0 ) , x2 ( 0 ) ) and ( x1 ( 1 ) , x2 ( 1 ) ) . The local coordinates describing the cortical space are the eccentricity x1 = ɛ generating the meridians and the angle x2 = φ yielding the circles of constant eccentricity . The metric tensor g ( ɛ , φ ) can directly be calculated from Equations 6 and 7 as shown in the following . The infinitesimal path-length on the cortical space is in cylindrical coordinates: Using Equations 6 and 7 we obtain: Recall Equation 8 , the components of the metric tensor g ( ɛ , φ ) can be directly taken from Equation 9: In terms of the metric tensor g ( ɛ , φ ) one obtains the Lagrange function: This yields a system of differential equations of second order: where and Since this system of differential equations has no analytical solution , it was solved numerically with respect to the boundary condition , i . e . , the two points in visual coordinates ( ɛ ( 0 ) = ɛ0 , φ ( 0 ) = φ0 , ɛ ( 1 ) = ɛ1 , φ ( 1 ) = φ1 ) . When two points are in different hemispheres the shortest path through the fovea has been used . Thus , for each simulated cell we computed its distance to the saccade target on the cortical surface . Since the subjects' eye position is not always available , we took a more general approach and simulated the time course of each saccade by approximating its velocity profile using a sixth-order polynomial Given the following constraints: it is possible to find a unique solution for the seven free parameters . If the amplitude a of a saccade is given , we have to determine the duration d of a saccade , the maximal velocity vmax and the point in time where the velocity reaches its maximum . The duration of each saccade was obtained by d = d0 + d1a [98] . Consistent with Becker [98] , who reported a range of 20–30 ms for d0 and a range of 2–3 ms per degree for d1 , we set d0 = 25 ms and d1 = 2 . 5 ms per degree . Knowing the duration of a saccade , the mean velocity is given by and the peak velocity of a saccade is where c denotes the ratio of the peak velocity to the mean velocity , i . e . , . Becker [98] approximated c with a constant value of c = 1 . 65 . Finally , we have to determine . Takagi et al . [99] defined the skewness S of the velocity profile by the ratio of the acceleration phase to the duration of the saccade . For rightward saccades they estimated the following linear regression equation where . With S it is possible to determine After determining the parameters of Equation 10 with respect to the constraints for each saccade amplitude ( Table 1 ) , we obtain the velocity and the path of the eye movement ( Figure 7 ) . The angle θ the eye moves within a time interval [t1 , t2] is given by the integral θ is then used in the model to update the retinal eccentricity of a flashed stimulus during a saccade . Assuming a stimulus is flashed at time ts at position θs ( t = 0 denotes saccade onset ) , the eccentricity ɛs of the stimulus with respect to the actual eye position is then Our neurocomputational model of peri-saccadic perception has been parameterized using mathematical functions to describe the anatomy and the neural dynamics , such as the shape and timing of the feedback signal and the receptive field size over eccentricity . Nevertheless , we have unknown parameters which could not be determined by other independent investigations ( Table 2 ) . We estimated these unknown parameters to fit the model with the data . We simulated the exact time course of the perceived stimulus position given the time and position of the flashed stimulus . In order to relate the model to data taken from a particular time window , we calculated a mean time value from all data points in the time window . From the data showing the spatial pattern of compression [11] we obtained t = ms as being used for f ( t ) in Equations 3 or 4 . We considered all data points from 0–25 ms for the 12° and 16° saccade amplitude and from 0–20 ms for the 20° and 24° saccade amplitude . The size of the window was chosen to obtain a sufficient number of trials in the time bin where the effect of compression is strongest . From the data showing the spatial range of compression in the critical phase from −25 to 0 ms [9] , we obtained a mean time value of = − 11 . 38 ms . With respect to receptive field size , only the receptive field sizes in the input of each layer are constrained by the data , since the neural population in this layer provides the input for the gain modulation and thus the degree of distortion . The data provides only little constraints about the overall magnitude of magnification , as verified by simulations with different magnification factors . However , since the ratio of cortical magnification along rays ( Mp ) to magnification along circles ( Me ) has turned out to be relevant for fitting the spatial pattern of compression ( Mp > Me ) , we have to determine specific values . To reflect the input of earlier stages the cortical magnification along the rays in L1 , in was chosen similar to the magnification in area V2 of monkey and the magnification in L1 , pool similar to monkey MT and V4 . Since we do not know direct measurements of cortical magnification in higher areas we set identical to . The magnification along the rings of constant eccentricity could either be identical to ( isotropic condition ) or different ( anisotropic condition ) . In all other model parts , magnification is isotropic . In the anisotropic condition , we roughly determined to obtain a sufficient fit of the data showing the spatial pattern of compression . After running these preliminary simulations to obtain plausible initial values , the fitting procedure was performed in two steps . In the first step , we started with small receptive field sizes as well as with a small value for the strength of the feedback signal w and iteratively increased them by allowing adjustments to the initial values of the other parameters on the data from Morrone et al . [9] . Besides determining the RF sizes , this fitting process resulted in the final values of α , β , the strength w of the feedback signal and the width of the feedback signal in L2 for a 20° saccade amplitude . In the second step , we obtained the final values of the saccade amplitude dependent feedback width ( ) on the data from Kaiser and Lappe [11] . This was done by minimizing the sum of the absolute errors between model and data , i . e . , the absolute differences in the x- and y-direction for each saccade amplitude , as a robust estimation procedure [100] . The obtained value for the width of the feedback signal was then also used in the simulation of the data from Morrone et al . [9] . Thus , all data was fitted with a single parameter set . Mean errors between data and model for the spatial range of compression ( Figure 2C ) were computed as follows: for each of the eight conditions , i . e . , each model specification ( isotropy versus anisotropy ) and each saccade amplitude ( 12° , 16° , 20° , 24° ) , the differences between the vector endpoints of the perceived and the predicted flash positions were obtained for the x- and y-directions , yielding to a total of 48 differences ( 24 differences in the x-direction and 24 differences in the y-direction ) for each condition . Then , the mean error for both the x- and y-directions was obtained by computing the respective arithmetic mean . Negative values indicate an undershoot of the model , i . e . , the theoretically predicted component is smaller than the empirically obtained compression . We tested separately for the x- and y-directions for each saccade amplitude if on average each model deviates statistically significant from the data ( two-sided one-sample t-test , α = 0 . 05 , df = 23 ) . For the isotropic model all mean errors reach statistical significance ( p < 0 . 05 ) except the error in the y-direction for the 12° and 16° saccade . For the anisotropic model none of the mean errors reaches statistical significance ( p > 0 . 05 ) . In order to quantify the model fit we used the following proportional reduction in error measure E1 is simply the sum of squared error ( SSE ) of the data with respect to a particular empirical mean value and E2 is the SSE with respect to the corresponding model predictions . If E2 ≥ E1 , pre was set to zero . Since we have i = 1 . . . 13 pairs of ( E1;E2 ) ( 8 of the spatial compression pattern , 4 of the time course , 1 of the spatial range ) , aggregated pre-measures where obtained by summing up the respective E1i and E2i so that To exclude the apparent shift in baseline from the measurement , we additionally determined pre*-measures of bars flashed at 20° and −20° for which errors ( E1 and E2 ) were computed using data points only in the period before t = 40 ms . For a comparison of the model receptive fields with mapped receptive fields in cortical cells we have to apply the same methods . To approximate the receptive field size , one-dimensional activity profiles were obtained by presenting a point stimulus in steps of 1° along the horizontal meridian . Since the size of a given receptive field kernel only depends on the eccentricity of the receptive field center in visual space , only cells of one hemisphere with centers along the horizontal meridian were included into the mapping to speed up the procedure . As commonly done in electrophysiology the obtained activity profiles were fitted using a Gaussian , yielding a set of receptive field widths defined by the σ of the respective Gaussian . According to Albright and Desimone [101] who approximated the size of receptive fields by Gaussian functions , the ratio of the width ( σ ) to the manually mapped width of the receptive fields is about 0 . 5 . Thus , in order to convert the set of the estimated L2 , in receptive field width into the usually used , each entry of the set is multiplied by a factor of 2 . This converted set was then fitted with a linear function to obtain the final description of the receptive field size in L2 , in . We determined the receptive field dynamics of the L1 , pool , L2 , in and L2 , pool cells with a half-maximum response threshold in two conditions , pre-saccadic and peri-saccadic ( t = 0 ) . The half-maximum response threshold is a common method to analyze physiological data [3] . The pre-saccadic receptive field was mapped without any feedback and the peri-saccadic receptive field with maximal feedback strength using dot stimuli presented in the visual field in steps of 4° . The obtained activity profile was then normalized to the maximal activity and interpolated to a resolution of 1° . The receptive field of a given cell is then defined as the area in visual space in which the activity exceeds half of the maximal activity of this cell . The model provides us a population response with respect to a flashed stimulus . In order to compare the output of the model with the data we have to determine the perceived stimulus position by decoding the population response with regard to spatial position . Since the model is deterministic and noiseless , decoding approaches based on probability distributions such as Bayesian inference are not appropriate . Thus , we directly use the firing rates ri and assume that a number of active cells N participate in encoding the stimulus location ps . r = {r1 , … , rN} can be considered as a vector in the N-dimensional space of neural responses . The unmodulated ( = 0 ) activity distribution resulting from the presentation of a stimulus at the location ps is used as a template f = r ( ps ) to which the distorted population r is compared . The estimated position ( in retinocentric coordinates ) is the one for which the angle between the two vectors r and f is minimized [102] , which is equivalent to This measure is particularly useful , since it tolerates the absolute increase in firing rate through the gain modulation . Please note our results are not qualitatively dependent on this particular method of decoding . | Early in the vertebrate lineage fast movements of the eye , called saccades , developed . This improvement in spatial direction selectivity has been achieved at a cost to handle a sequence of different views . Recent experiments showed that the brain uses its knowledge about the upcoming eye movement to guide perception prior to the next saccade . They revealed an improved recognition of objects at the saccade target , a change of receptive fields , and a mislocalization of briefly flashed stimuli towards the saccade target . We here offer a novel , unifying explanation for these phenomena and link them to a common neural mechanism . Our model predicts that the brain uses oculomotor feedback to transiently increase the processing capacity around the saccade target by changing the receptive field structure in visual areas and thus , it links the pre-saccadic scene to the post-saccadic one . A briefly flashed stimulus probes this change in the receptive field structure and demonstrates a close interaction of object and spatial perception . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"physiology",
"neuroscience",
"homo",
"(human)",
"primates"
] | 2008 | The Peri-Saccadic Perception of Objects and Space |
Glycosylation is a fundamental modification of proteins and membrane lipids . Toxins that utilize glycans as their receptors have served as powerful tools to identify key players in glycosylation processes . Here , we carried out Clustered Regularly Interspaced Short Palindromic Repeats ( CRISPR ) -Cas9–mediated genome-wide loss-of-function screens using two related bacterial toxins , Shiga-like toxins ( Stxs ) 1 and 2 , which use a specific glycolipid , globotriaosylceramide ( Gb3 ) , as receptors , and the plant toxin ricin , which recognizes a broad range of glycans . The Stxs screens identified major glycosyltransferases ( GTs ) and transporters involved in Gb3 biosynthesis , while the ricin screen identified GTs and transporters involved in N-linked protein glycosylation and fucosylation . The screens also identified lysosomal-associated protein transmembrane 4 alpha ( LAPTM4A ) , a poorly characterized four-pass membrane protein , as a factor specifically required for Stxs . Mass spectrometry analysis of glycolipids and their precursors demonstrates that LAPTM4A knockout ( KO ) cells lack Gb3 biosynthesis . This requirement of LAPTM4A for Gb3 synthesis is not shared by its homolog lysosomal-associated protein transmembrane 4 beta ( LAPTM4B ) , and switching the domains between them determined that the second luminal domain of LAPTM4A is required , potentially acting as a specific “activator” for the GT that synthesizes Gb3 . These screens also revealed two Golgi proteins , Transmembrane protein 165 ( TMEM165 ) and Transmembrane 9 superfamily member 2 ( TM9SF2 ) , as shared factors required for both Stxs and ricin . TMEM165 KO and TM9SF2 KO cells both showed a reduction in not only Gb3 but also other glycosphingolipids , suggesting that they are required for maintaining proper levels of glycosylation in general in the Golgi . In addition , TM9SF2 KO cells also showed defective endosomal trafficking . These studies reveal key Golgi proteins critical for regulating glycosylation and glycolipid synthesis and provide novel therapeutic targets for blocking Stxs and ricin toxicity .
The plant toxin ricin is derived from castor oil plant seeds . It has been utilized as a poison in criminal cases and is classified as a potential bioterrorism agent [1] . Shiga and Shiga-like toxins ( Stxs ) are a family of bacterial toxins including the prototype Stx , produced by the bacteria Shigella dysenteriae , and related Shiga-like toxins Stx1 and Stx2 , produced by Shigatoxigenic strains such as enterohemorrhagic E . coli ( EHEC ) [2 , 3] . EHEC is a major pathogen responsible for food poisoning and causing abdominal cramps and bloody diarrhea , as well as the life-threatening complication of hemolytic uremic syndrome ( HUS ) [4] . Ricin and Stxs are structurally and evolutionarily distinct but share the same mode of action: both act as ribosomal RNA N-glycosidase and inhibit protein synthesis by cleaving the same adenine of 28S rRNA . Ricin is composed of an A chain ( 32 kDa ) and a B chain ( 34 kDa ) , connected via a disulfide bond . The A chain is an N-glycosidase and the B chain is the receptor-binding domain . Stxs are A-B5 bacterial toxins [2 , 3] , composed of an A chain ( 32 kDa ) , which is an N-glycosidase , and a receptor-binding domain consisting of five identical B chains ( about 7 . 7 kDa each ) . These B chains form a ring , and the A chain connects to the B chain by inserting its C-terminus into the center pore of the B chain ring . Stx1 has only one single amino acid difference from Stx , while Stx2 represents a distinct serotype , with about 56% sequence identity to Stx . Ricin and Stxs also share similar entry pathways into cells [2 , 3 , 5 , 6] . Once they have entered cells through endocytosis , they are sorted into retrograde trafficking routes and enter the endoplasmic reticulum ( ER ) through the Golgi apparatus . Their A chains are then released from the ER into the cytosol , utilizing the host protein translocation machinery on the ER membranes . Consistent with this trafficking pathway , inhibitors that disrupt the Golgi apparatus , such as Brefeldin A , block ricin and Stxs toxicity . Small molecule inhibitors that disrupt the specific retrograde transport pathways utilized by ricin and Stxs have also been reported ( Retro-1 and Retro-2 ) [7] , although the host targets for Retro-1/2 remain to be established . The major difference between ricin and Stxs is their receptor recognition . Ricin binds broadly to glycan moieties containing galactose and N-acetylgalactosamine [8] . In contrast , Stxs specifically recognize the glycan headgroup of Gb3 ( also known as CD77 ) , a glycosphingolipid [2 , 3 , 9] . Crystal structure studies suggest that each Stx B domain contains three potential Gb3 binding sites; thus , one Stx may simultaneously cluster up to 15 Gb3 molecules [10] . Expression of Gb3 is highly restricted in a subset of germinal center B cells , kidney tissues , vascular endothelial cells , and neurons , while the majority of other cell types do not express detectable levels of Gb3 . Species such as cattle and deer do not express Gb3 and can serve as natural reservoirs for Shigatoxigenic bacteria [11] . Glycosylation is one of the most common modifications of proteins and membrane lipids [12 , 13] . It is initiated in the ER or on the ER membranes , while the majority of the remaining steps are carried out inside the Golgi apparatus . Transfer of sugar moieties to proteins or lipids are catalyzed by various glycosyltransferases ( GTs ) [14] . Genetic defects in protein and lipid glycosylation result in congenital disorders of glycosylation ( CDG ) , a growing disease family comprising nearly a hundred disorders [15] . These defects occur not only on GTs but also on key regulatory proteins that control the specificity/activity of GTs . Toxins that utilize glycan moieties as their receptors could serve as powerful tools in mutagenesis screens to identify host proteins involved in glycosylation [12 , 13] . Indeed , ricin has undergone various genome-wide screens on mammalian cells , including small interfering ribonucleic acid ( siRNA ) -mediated knockdown ( KD ) approaches , Clustered Regularly Interspaced Short Palindromic Repeats ( CRISPR ) -Cas9–mediated KD and knockout ( KO ) approaches , and retroviral mutations in haploid cells [16–19] . These screens have yielded a list of GTs and nucleotide transporters ( NTs ) involved in N-linked protein glycosylation . Here , we identified a human bladder cancer cell line that is highly sensitive to Stxs . Utilizing this cell line , we carried out CRISPR-Cas9–mediated KO screens for Stx1 and Stx2 . In addition , we also carried out a genome-wide screen for ricin . These screens identified lysosomal-associated protein transmembrane 4 alpha ( LAPTM4A ) as a novel Golgi protein specifically required for Gb3 biosynthesis . The screens also revealed two Golgi proteins , Transmembrane protein 165 ( TMEM165 ) and Transmembrane 9 superfamily member 2 ( TM9SF2 ) , as key factors for maintaining proper glycosylation levels in cells .
We first assessed the sensitivity of a panel of human cancer cell lines to Stx1 and Stx2 using a 72-h cell viability assay . The cells were exposed to a titration of toxins for 72 h , and the percentage of surviving cells was measured using MTT assays ( Fig 1A and S1A Fig ) . The toxin dose that induced death of 50% cells is designated IC50 . Most cell lines are insensitive to Stx1 and Stx2 , with IC50 > 10 , 000 ng/mL ( S1 Table ) . The bladder carcinoma cell line 5637 was the most sensitive one to Stx1 and Stx2 , with IC50 at about 0 . 028 ( Stx1 ) and 0 . 007 ( Stx2 ) ng/mL . The human kidney adenocarcinoma cell line ACHN is also quite sensitive to Stx . This level of sensitivity is not shared by two other bladder cancer cell lines ( T24 and RT4 ) , suggesting that it is not a general feature of bladder cancer cells . Consistently , immunostaining analysis using a polyclonal anti-Stx1 antibody showed robust binding of Stx1 to 5637 and ACHN cells , whereas binding to HeLa cells was undetectable ( S1D Fig ) . Utilizing 5637 cells , we carried out genome-wide CRISPR-Cas9–mediated loss-of-function screens . Cells stably expressing Cas9 were established and transduced with a lentiviral sgRNA library ( GeCKO V2 ) , targeting 19 , 052 human genes with six single guide ribonucleic acids ( sgRNAs ) per gene [20 , 21] . These cells were then subjected to two rounds of selection with Stx1 or Stx2 ( Fig 1B ) . The sgRNA sequences in surviving cells were identified via next-generation sequencing ( NGS ) . Cells that were not treated with toxins served as controls . The identified genes were ranked based on the number of unique sgRNAs ( y axis ) and the total number of NGS reads ( x axis ) ( Fig 1C and 1D ) . The full list of screen results is shown in S1 Data . Most top-ranked hits overlapped between Stx1 and Stx2 screens . Five of the top eight genes are known factors in the Gb3 synthesis pathway ( S1G Fig ) : A4GALT , B4GALT5 , SLC35A2 , UGCG , and SPTSSA [13] . Serine palmitoyltransferase small subunit A ( SPTSSA ) is a part of the serine palmitoyltransferase complex on ER membranes , which catalyzes the first and rate-limiting step in sphingolipid biosynthesis to generate ceramide ( Cer ) . Cer is then transported to the Golgi . Ceramide glucosyltransferase ( UGCG ) catalyzes glucose onto Cer on the cytosol side of the Golgi , which generates glucosylceramide ( GlcCer ) . UGCG may also flip GlcCer into the lumen side of the Golgi , where β-1 , 4-galactosyltransferase 5 ( B4GALT5 ) then catalyzes the transfer of a galactose from UDP-galactose onto GlcCer to generate lactosylceramide ( LacCer ) , which is the precursor for both globo-series and ganglio-series of glycosphingolipids . α-1 , 4-galactosyltransferase ( A4GALT ) finally produces Gb3 by catalyzing the transfer of galactose to LacCer [22 , 23] . A4GALT is ranked number 1 in the Stx1 screen and number 2 in the Stx2 screen based on NGS reads ( S1 Data ) . SLC35A2 ( UDP-galactose translocator ) transports UDP-galactose from the cytosol into the lumen of the Golgi . The other three of the top eight genes are LAPTM4A , TMEM165 , and TM9SF2 . LAPTM4A is a poorly characterized small four-pass membrane protein with unknown function , previously proposed to be localized mainly on lysosomes [24–27] . It is ranked as high as A4GALT in our screens ( S1 Data ) . TMEM165 is a multi-pass transmembrane protein localized in the Golgi . Mutations in TMEM165 have been linked to CDG , and TMEM165 deficiency causes a defect in glycosylation , possibly because of dysregulation of Mn2+ hemostasis in the Golgi , as Mn2+ is an essential metal cofactor for many GTs [28–30] . TM9SF2 contains a large N-terminal domain and nine transmembrane domains; its function remains unknown [31] . Both TMEM165 and TM9SF2 have been previously identified as genes involved in heparan sulfate ( HS ) biosynthesis in two independent genetic screens using haploid cells , suggesting that TMEM165 and TM9SF2 deficiency affects multiple glycosylation pathways [32 , 33] . Interestingly , a recent genome-wide CRISPR-Cas9 screen , carried out by incubating EHEC with human colorectal carcinoma cell line HT-29 , also identified LAPTM4A and TM9SF2 as key factors for EHEC toxicity to cells [34] . Furthermore , LAPTM4A KO and TM9SF2 KO cells showed no binding of Stx1 , suggesting that the deficiency in LAPTM4A or TM9SF2 reduces Gb3 in cells [34] . To validate our screen results , we generated mixed ( uncloned population ) KO 5637 cells using the CRISPR-Cas9 approach for A4GALT , SLC35A2 , UGCG , B4GALT5 , LAPTM4A , TMEM165 , and TM9SF2 . Cell viability assays were carried out to determine their sensitivity to Stx1 and Stx2 ( S1 Table ) . A4GALT , SLC35A2 , UGCG , B4GALT5 , and LAPTM4A KO cells all showed > 105-fold increase in resistance to Stx1 and Stx2 compared to wild-type ( WT ) cells . TM9SF2 KO and TMEM165 KO cells showed about 10-fold and 780-fold smaller increases in resistance compared with LAPTM4A KO and A4GALT KO cells . In contrast to Stxs , most cells are sensitive to ricin ( S1B Fig and S1 Table ) . We thus utilized HeLa cells that stably express Cas9 for a genome-wide screen for ricin . The sgRNA sequences in surviving cells were identified and ranked ( Fig 1E and S2 Data ) . The majority of top-ranked genes fall into four pathways: N-linked protein glycosylation , fucosylation , membrane trafficking , and ER-associated protein degradation ( ERAD ) /quality-control pathways ( Fig 1F ) . Top hits involved in glycosylation include enzymes located in the ER for catalyzing formation and degradation of high mannose oligosaccharide ( ALG5 , ALG6 , ALG8 , MOGS ) and transferring oligosaccharides to an asparagine residue as an N-linked glycan ( OST4 ) , as well as enzymes located in the Golgi that convert oligo-mannose to galactose-containing complex N-linked glycans ( MAN1A2 , MAN2A1 , MGAT1 , and MGAT2 ) . These results are consistent with the established view that galactose on N-linked glycans is the primary receptor for ricin . Our screens also identified key players in the fucosylation pathway , including TSTA3 and GMDS , which catalyze the synthesis of GDP-fucose; SLC35C1 , which transports GDP-fucose from the cytosol into the Golgi lumen; and FUT4 , which catalyzes the transfer of fucose to N-acetyllactosamine to generate fucosylated carbohydrates , such as the non-sialylated carbohydrate antigen , Lewis X . The critical role of fucosylation in ricin toxicity has been previously reported and validated , possibly because fucosylation promotes the exposure of terminal galactoses by preventing their sialylation [19 , 35 , 36] . The screens revealed a series of proteins involved in intracellular vesicular transport , including members of the Golgi-associated retrograde protein ( GARP ) complex ( VPS51 , 52 , 53 , 54 ) , which is a tethering complex involved in retrograde transport from the early endosome to the Golgi , as well as a few proteins involved in Golgi-ER trafficking ( e . g . , GOSR1 , NBAS , STX5 , NAPG , ARL5B ) . There are also two ER proteins among the top-ranked hits: ER resident protein 44 ( ERP44 ) , which is a member of the protein disulfide isomerase family , and ubiquitin conjugating enzyme E2 G2 ( UBE2G2 ) , which is an E2 ubiquitin-conjugating enzyme involved in ERAD . Ricin has been previously subjected to shRNA-mediated KD screens and a CRISPR-Cas9–mediated loss-of-function screen utilizing a K562 cell line ( a human bone marrow lymphoblast ) and a retroviral mutagenesis screen using haploid cells [16–19] . Among our top 20 hits , 13 overlap with the top hits of at least one of the previous screens , providing a degree of validation across distinct cell lines [17 , 18] . Among the other seven newly identified hits , two ( VPS51 and MOGS ) are within well-established pathways required for ricin , as described above . The other five are GOSR1 , JTB , NBAS , TMEM165 , and TM9SF2 . Golgi SNAP receptor complex member 1 ( GOSR1 ) and neuroblastoma-amplified sequence ( NBAS ) are involved in Golgi-ER trafficking . Jumping translocation breakpoint protein ( JTB ) may play a role in regulating cell proliferation , but its role in ricin intoxication remains unknown . TMEM165 and TM9SF2 are the only two top-ranked factors shared between ricin and Stx screens . To validate our ricin screen results , we generated mixed KO HeLa cells using the CRISPR-Cas9 approach for MGAT2 , SLC35C1 , ERP44 , and TAPT1 , which are among the top hits reported from previous screens but have not been experimentally validated . We also generated mixed KO cells for the three new factors GOSR1 , JTB , and NBAS . These KO cells all showed modest increases ( about 10–200-fold ) in resistance to ricin compared with WT cells ( S2E Fig and S1 Table ) . We then focused on investigating the role of LAPTM4A , which is specific for Stxs , as well as TMEM165 and TM9SF2 , the two factors shared between Stxs and ricin . We first examined whether the requirement of LAPTM4A is limited to 5637 cells . We generated LAPTM4A KO ACHN cells using the CRISPR-Cas9 approach , which became resistant to Stx1 and Stx2 ( S2C and S2D Fig ) . To further address the concern on potential off-target effects , we generated a second line of LAPTM4A KO 5637 cells using a different sgRNA sequence ( S2 Table ) . This new KO line was resistant to Stx1 and Stx2 as well ( LAPTM4A-KO-II-Mix , S2A and S2B Fig ) . We also generated a 5637 KO cell line lacking LAPTM4B , which is a homolog of LAPTM4A . This cell line showed sensitivity to Stx1 and Stx2 similar to that of WT cells ( S2A and S2B Fig ) , suggesting that the role of LAPTM4A is not shared by its homolog . To further investigate the role of LAPTM4A , we isolated single clones from the LAPTM4A KO population . The genotype of each clone was determined by sequencing , which indicates that 5637 cells contain three sets of chromosomes ( S3 Data ) . Two lines ( KO-10 and KO-12 ) contain frameshift mutations at the target region on all three chromosomes . One line ( Mut-9 ) contains frameshift mutations on two chromosomes and a deletion of nine base pairs on the third chromosome . Consistent with these genotyping results , KO-10 and KO-12 cells are resistant to Stx1 and Stx2 in cell viability assays , while Mut-9 cells showed only reduced sensitivity to both toxins ( Fig 2A and 2B and S3 Table ) . To determine whether increased resistance is specific to Stxs , we examined the sensitivity of LAPTM4A KO cells to ricin and two other bacterial exotoxins: anthrax-diphtheria chimeric toxin ( A-Dtx ) and cholera toxin ( Ctx ) . A-Dtx is composed of the N-terminal part of anthrax toxin lethal factor ( LFn ) fused to the enzymatic domain of diphtheria toxin ( DTA ) and the receptor-binding/translocation domains of Anthrax toxin ( PA ) [37] . PA mediates binding and entry of toxins into cells and translocates DTA into the cytosol from endosomes . DTA induces death of cells by inhibiting protein synthesis—the same process disrupted by ricin and Stxs , but the entry of this chimeric toxin does not require retrograde transport into the Golgi-ER . Ctx utilizes ganglioside GM1 as its major receptor and requires retrograde transport into the Golgi-ER for its release into the cytosol [38] . Ctx then catalyzes ADP-ribosylation of the Gs alpha subunit , which elevates cAMP levels . The sensitivity of cells to ricin and A-Dtx was determined by cell viability assays , while the sensitivity to Ctx was quantified by measuring cAMP levels in cell lysates . KO-10 and KO-12 showed sensitivity to ricin , A-Dtx , and Ctx similar to that of WT cells ( Fig 2C–2E and S3 Table ) . We also examined Mut-9 , A4GALT KO , and LAPTM4B KO cells , which all showed levels of sensitivity to these toxins similar to that of WT cells . To determine which step of toxin action is affected in LAPTM4A KO cells , we first examined binding of Stx1 to cells via immunofluorescent staining . WT and Mut-9 cells showed robust binding of Stx1 , while there was no detectable binding to KO-10 or KO-12 cells ( Fig 2F ) . As expected , A4GALT KO cells showed no binding of Stx1 , while LAPTM4B KO cells showed robust binding ( Fig 2F ) . Furthermore , transfecting KO-10 and KO-12 cells with a plasmid that expresses LAPTM4A rescued binding of Stx1 , while expression of LAPTM4B in these two KO cells did not restore Stx1 binding ( Fig 2G ) . We also examined binding of Ctx utilizing a fluorescently labeled receptor-binding domain of Ctx ( CtxB ) . Unlike Stx1 , CtxB showed similar levels of binding to all these cells ( Fig 2F ) , suggesting that lack of LAPTM4A specifically affects the Gb3 branch of glycolipids . In addition to immunofluorescent staining , binding of Stx1 was also analyzed by immunoblot of cell lysates , as well as by flow cytometry , which yielded the same results as immunofluorescent staining ( S3 Fig ) . Loss of Stx1 binding implies that LAPTM4A is critical for Gb3 expression . To examine this possibility directly , we sought to quantify the levels of Gb3 and its precursors LacCer , GlcCer , and Cer in cells using mass spectrometry analysis . Total lipids were extracted from cell lysates . Samples were then analyzed using an ultra-pressure liquid chromatograph ( UPLC ) coupled to a mass spectrometer . Quantification of glycolipids was normalized based on endogenous phosphatidylcholine ( PC ) as an internal standard ( Fig 2H and S4A Fig ) . We found that Gb3 is greatly reduced in KO-10 and KO-12 cells , similar to the case in A4GALT KO cells ( Fig 2I and S4 Data ) . Ectopic expression of LAPTM4A in KO-12 cells partially restored Gb3 levels . In contrast to Gb3 , levels of LacCer are elevated in KO-10 and KO-12 cells , as well as in A4GALT KO cells . This increase in LacCer is abolished with expression of LAPTM4A in KO-12 cells . In addition , KO-12 cells also showed elevated GlcCer and Cer levels , both of which were restored with expression of LAPTM4A . These results demonstrate that LAPTM4A KO cells lack Gb3 expression . Because LacCer is elevated in LAPTM4A KO cells , it is likely that LAPTM4A is involved in Gb3 synthesis from LacCer , rather than in Gb3 degradation pathways . To understand the function of LAPTM4A , we next examined its subcellular localization . As there is no suitable antibody to detect endogenous LAPTM4A , we utilized a LAPTM4A tagged with a triple HA tag on its C-terminus . This tagged version restored binding of Stx1 when expressed in KO-10 and KO-12 cells ( Fig 2G ) . Expression of this tagged LAPTM4A in 5637 , HEK293T , and HeLa cells all showed high degrees of colocalization with both Giantin and TGN46 , two well-established markers for the Golgi , but not other organelle markers such as Rab5 ( early endosome ) , Rab7 ( late endosome ) , Sec61A ( ER ) , or Lamp1 ( lysosome ) ( Fig 3A and S5 and S12 Figs ) . We also examined localization of endogenous A4GALT using a polyclonal antibody whose specificity was confirmed using A4GALT KO cells ( S6A Fig ) . As expected , A4GALT largely colocalizes with the Golgi marker Giantin . Consistently , HA-tagged LAPTM4A colocalizes with A4GALT in 5637 , HEK293T , and HeLa cells ( Fig 3B ) . To further examine whether LAPTM4A interacts with A4GALT , we co-expressed HA-tagged LAPTM4A and FLAG-tagged A4GALT in HEK293T cells and then carried out co-immunoprecipitation ( co-IP ) assays using a FLAG tag antibody . The FLAG-tagged A4GALT is correctly localized to the Golgi and can rescue binding of Stx1 to A4GALT KO cells ( S6B–S6D Fig ) . In addition , we also utilized a FLAG-tagged B4GALT5 as a control . Both A4GALT and B4GALT5 are type II transmembrane proteins in the Golgi , each with a single transmembrane domain and a short cytoplasmic domain located on its N-terminus . We found that LAPTM4A was specifically pulled down together with A4GALT , but not B4GALT5 ( Fig 3C ) . Furthermore , a chimeric protein composed of the cytoplasmic and transmembrane domains of A4GALT and the luminal domain of B4GALT5 retained the ability to pull down LAPTM4A ( Fig 3C ) . We next analyzed whether the absence of LAPTM4A affects the Golgi localization and/or stability of A4GALT . We found that A4GALT is still localized to the Golgi in KO-10 and KO-12 cells and the expression levels of A4GALT in KO-10 and KO-12 cells are similar to those of WT and LAPTM4B KO cells ( Fig 3D ) . Immunoblot analysis of the cell lysates further confirmed that the lack of LAPTM4A did not affect the level of A4GALT ( Fig 3E ) . In addition , KO-10 and KO-12 cells also showed levels of A4GALT mRNA similar to that found in WT cells ( S6E Fig ) . LAPTM4A and LAPTM4B are small four-pass transmembrane proteins ( S7A Fig ) whose overall topology has not been established . We expressed two versions of LAPTM4A in cells , one with an HA tag on its N-terminus and the other with an HA tag on its C-terminus . We also utilized an ER membrane protein , ER membrane protein complex subunit 1 ( EMC1 ) , which has a single transmembrane domain , as a control . Cells were permeabilized with two different detergents: Digitonin , which permeabilizes only the plasma membrane but not the Golgi/ER membrane , or Saponin , which permeabilizes both the plasma membrane and the Golgi/ER membrane . Immunostaining analysis using an HA antibody labeled both the N-terminal-tagged and C-terminal-tagged LAPTM4A when cells were permeabilized with Digitonin ( S7B Fig ) . In contrast , only C-terminal-tagged EMC1 was labeled , but not the N-terminal-tagged EMC1 under Digitonin treatment . These results suggest that both the N- and C-termini of LAPTM4A are exposed in the cytosol and that only two small regions ( 33 and 32 residues , respectively ) between transmembrane domains are located within the Golgi lumen ( S7C Fig ) . Taking advantage of the observation that LAPTM4B cannot restore Stx1 binding to LAPTM4A KO cells , we generated a series of chimeric proteins between these two homologs in order to map the region responsible for their functional difference . These homologs were expressed in KO-10 and KO-12 cells and binding of Stx1 was analyzed by immunostaining , flow cytometry , and immunoblot ( Fig 3F and S8 Fig ) . The results suggest that the second luminal domain ( residues 129–159 ) is a critical region; engrafting this region to LAPTM4B generates a chimeric protein ( designated AB4 ) that restored Stx1 binding . In contrast , engrafting other regions , such as the N-terminal cytosolic region , the C-terminal cytosolic region , the two middle transmembrane domains plus the middle cytoplasmic domain , or all four transmembrane domains , did not restore Stx1 binding . Co-IP analysis showed that both AB4 and LAPTM4B can interact with A4GALT ( S6F Fig ) , suggesting that the interaction with A4GALT is mediated by regions conserved between LAPTM4A and LAPTM4B . We conclude that the second luminal domain accounts for the observed functional difference between LAPTM4A and LAPTM4B . It is likely that this luminal region of LAPTM4A contributes to Gb3 synthesis by influencing the catalytic activity/specificity of A4GALT , although the mechanism remains to be established . We next investigated the role of TMEM165 and generated single clones of TMEM165 KO 5637 cells via the CRISPR-Cas9 approach . We obtained one KO line ( TM-KO-3 ) that contains frameshift mutations on all chromosomes , and one mutation line ( TM-Mut-1 ) that contains a frameshift and an insertion of six base pairs ( S3 Data ) . The latter still expressed TMEM165 , but with two extra residues inserted at the target region ( insertion of Cys-Tyr residues between Leu279 and Cys280 ) . Cell viability assays showed that both TM-KO-3 and TM-Mut-1 cells are about 20–80-fold less sensitive to Stx1 and Stx2 compared with WT cells ( Fig 4A and 4B and S3 Table ) . To analyze the sensitivity of TMEM165-deficient cells to ricin , we initially carried out the standard 72-h cell viability assay , but the IC50 values were not significantly changed under our assay conditions ( S3 Table ) . We then analyzed the sensitivity of these cells by fixing the ricin concentration and monitoring cell viability over time ( Fig 4C ) . Both TM-Mut-1 and TM-KO-3 cells showed higher levels of surviving cells compared with WT cells when exposed to ricin for 20–30 h , and this difference disappeared by 40–50 h ( Fig 4C ) , suggesting that TMEM165 deficiency caused a rather minor reduction in sensitivity to ricin . We also found that TMEM165-deficient cells showed similar levels of sensitivity toward A-Dtx but reduced sensitivity to Ctx , compared with WT cells ( Fig 4D and 4E ) . We next examined binding of Stx1 and CtxB to TMEM165-deficient cells by immunofluorescent staining , immunoblot , and flow cytometry ( Fig 4F and S9A and S9B Fig ) . Both TM-KO-3 and TM-Mut-1 showed reduced binding of Stx1 and CtxB compared with WT cells , suggesting that a lack of TMEM165 affects the toxin binding step . Ectopic expression of TMEM165 restored binding of Stx1 and CtxB to TM-KO-3 and TM-Mut-1 cells ( Fig 4F ) . Mass spectrometry analysis confirmed that both TM-KO-3 and TM-Mut-1 cells have lower levels of Gb3 compared with WT cells ( Fig 4G and S4 Data ) . In contrast to LAPTM4A KO cells , these two lines showed lower levels of LacCer , GlcCer , and Cer . We further analyzed the levels of gangliosides in these TMEM165-deficient cells by mass spectrometry ( S4B Fig ) and found that TM-KO-3 and TM-Mut-1 cells showed low levels of gangliosides such as GM2 ( Fig 4G ) , suggesting that the absence of TMEM165 affects the biosynthesis of glycosphingolipids globally . Mutations in TMEM165 have been linked to a subtype of CDG . It has been reported that TMEM165 is localized to the Golgi and that the hypo-glycosylation defect in TMEM165-deficient cells can be rescued by supplementing the culture medium with Mn2+ [28–30] . Consistent with these reports , we found that TMEM165 colocalizes with the Golgi markers Giantin and TGN46 in 5637 , HEK293T , and HeLa cells ( Fig 5A and S9C , S9D and S12 Figs ) . Adding MnCl2 into medium partially restored binding of Stx1 and CtxB to TM-KO-3 and TM-Mut-1 cells ( Fig 5B and 5C ) . Higher concentrations of Mn2+ are toxic to cells , and TM-KO-3 and TM-Mut-1 cells both showed a lower tolerance for high concentrations of Mn2+ compared with WT cells , further supporting the role of TMEM165 in regulating Mn2+ homeostasis ( Fig 5D ) . We further screened other major metal ions and found that both NiCl2 and FeCl3 can also restore binding of Stx1 to TMEM165-deficient cells ( Fig 5E and 5F ) . To examine the role of TM9SF2 , two single KO 5637 cell lines were generated via the CRISPR-Cas9 approach . Both ( SF2-KO-8 and SF2-KO-9 ) contain frameshift mutations in all three chromosomes ( S3 Data ) . A single clone ( SF2-WT-5 ) that still contains a WT allele was also selected as an additional control . Cell viability assays showed that both KO lines became highly resistant to Stx1 and Stx2 , while SF2-WT-5 cells showed only a slight reduction in sensitivity compared with WT cells ( Fig 6A and 6B and S3 Table ) . Similar to TMEM165 KO cells , the IC50 of TM9SF2 KO cells toward ricin was not significantly changed under our standard 72-h cell viability assay ( S3 Table ) , but these cells showed greater viability after exposure to a fixed concentration of ricin for a shorter period of time ( Fig 6C ) . TM9SF2 KO cells also showed no change in sensitivity to A-Dtx , but reduced sensitivity to Ctx ( Fig 4D and 4E ) . We found that binding of Stx1 and CtxB was largely abolished in two TM9SF2 KO cells , but not in the SF2-WT-5 control cells ( Fig 6F and S10 Fig ) . Binding was restored in two KO cell lines when TM9SF2 was expressed via transient transfection ( Fig 6F ) . Mass spectrometry analysis revealed that both TM9SF2 KO cells had low levels of Gb3 , LacCer , GlcCer , Cer , and GM2 ( Fig 6G ) , indicating that TM9SF2 deficiency caused global disruption in glycosphingolipid synthesis . It has been previously reported that TM9SF2 deficiency in haploid cells reduces HS biosynthesis [32 , 33] . Consistently , we found that surface HS levels were reduced in the two TM9SF2 KO cells when analyzed by flow cytometry using an antibody that recognizes HS ( Fig 6H ) . These results confirmed that TM9SF2 deficiency affects multiple glycosylation pathways . Previous studies have showed that Myc-tagged TM9SF2 is localized mainly on endosomes [31] . More recently , Tanaka and colleagues showed that endogenous TM9SF2 , detected with a polyclonal TM9SF2 antibody , is localized in the Golgi in a haploid cell line [33] . Golgi localization was also recently confirmed for TM9SF2 in HeLa cells [34] . We first validated the specificity of this polyclonal antibody , which showed no signal on TM9SF2 KO cells ( S11A Fig ) . We found that endogenous TM9SF2 in 5637 , HEK293T , and HeLa cells are largely colocalized with the Golgi markers ( Fig 7A and S11B , S11C and S12 Figs ) , confirming that TM9SF2 is a Golgi protein across multiple cell lines . Interestingly , both SF2-KO-8 and SF2-KO-9 cells appear to form many large vacuole-like structures within the cytosol , which are labeled with Rab7 ( Fig 7B ) . To further examine endosomal trafficking in SF2 KO cells , we loaded exogenous sphingosine ( Sph ) labeled with the fluorescent dye nitrobenzoxadiazole ( NBD ) . Sph was taken up by cells , trafficked through endosomes , and dispersed throughout the cells within 4 h in WT cells ( Fig 7C ) . In contrast , trafficking of Sph was interrupted and remained within vesicular structures even at 24 h in both SF2-KO-8 and SF2-KO-9 cells ( Fig 7C ) . This trafficking defect was rescued when TM9SF2 was expressed via transient transfection in SF2-KO-8 and SF2-KO-9 cells ( Fig 7D ) . These Sph-containing vesicles are labeled with Rab5 and Rab7 , but not Lamp1 , suggesting that Sph was largely retained within endosomes in TM9SF2 KO cells ( Fig 7E ) . We further examined trafficking of NBD-labeled Cer and phosphatidylserine ( PS ) . Both lipids were taken up by cells and eventually dispersed within cells in WT cells . In contrast , both accumulated in vesicular structures in SF2-KO-8 and SF2-KO-9 cells ( S11D and S11E Fig ) . Furthermore , we generated mixed TM9SF2 KO HeLa cells , which showed similar defects in trafficking of NBD-labeled Sph , Cer , and PS ( S11F Fig ) , demonstrating that these trafficking defects exist across distinct cell types . These findings suggest that the absence of TM9SF2 not only disrupts glycosylation in Golgi but also severely disrupts endosomal trafficking in general , which may also contribute to resistance to Stxs and ricin .
Here , we found that the human bladder cancer cell line 5637 is highly sensitive to Stxs . Cell line 5637 is easy to culture and well suited for cell biology studies; thus , it can serve as a useful human cell model for studying Stxs . The top eight hits from Stx1 and Stx2 screens are all involved in Gb3 synthesis and cells lacking Gb3 are highly resistant to Stxs ( 105-fold ) , further demonstrating that binding to Gb3 is the key rate-limiting step for Stx intoxication . In contrast , top hits from the ricin screen are distributed across multiple pathways , including protein glycosylation , fucosylation , retrograde trafficking , and ERAD pathway . Mutations in these ricin host factors elicited only low levels of resistance ( <200-fold ) . These results are consistent with the view that various galactose-containing moieties and multiple trafficking pathways can mediate entry of ricin redundantly . Other notable top-ranked hits shared between Stx1 and Stx2 screens include ARCN1 , UGP2 , and SPPL3 . Coatomer subunit delta ( ARCN1 ) is a component of the coat protein complex I ( COPI ) complex , a protein complex that mediates budding of vesicles from the Golgi membranes . COPI plays a critical role in vesicular transport between the Golgi and the ER [39] , but its role for Stxs remains to be validated [17 , 40] . UDP-Glucose Pyrophosphorylase 2 ( UGP2 ) is a cytosolic enzyme that produces UDP-glucose , a substrate required for glycosylation . Signal peptide peptidase-like 3 ( SPPL3 ) is a membrane aspartic protease localized in the Golgi . It has been shown that SPPL3 cleaves many GTs and glycosidases within the Golgi , thus releasing their enzymatic domains into the Golgi lumen [41 , 42] . SPPL3 was identified in our ricin screen as well , although its exact role in relation to Stxs and ricin remains to be validated . Our screening protocol relies on the cell viability assay after 72 h incubation . Such a long incubation could mask minor reductions in toxin trafficking within cells . It is possible that a few trafficking-related factors , such as GARP and STX5 , were identified because they potentially may also affect glycosylation processes . Our Stx1 screen did not identify GPP130 , a protein previously reported to be important for retrograde transport of Stx1 ( but not Stx2 ) [43] . However , the role of GPP130 for Stxs remains to be established [44] . A recent shRNA-based KD screen for Stxs , utilizing HeLa cells that overexpress A4GALT , reported that defects in the Golgi protein UNC50 reduces the sensitivity of cells to Stx2 by about 200-fold [45] . UNC50 was not identified in our screen for Stx2 . Our major finding is identification of LAPTM4A , a poorly characterized small four-pass transmembrane protein , as a novel factor critical for Gb3 biosynthesis . Endogenous LAPTM4A protein has been detected in both the Golgi and lysosome fractions of rat liver membrane lysates [24 , 25] . Previous studies suggested that LAPTM4A is localized on late endosomes and lysosomes based on using LAPTM4A tagged with fluorescent protein mCherry , HA , or Myc on its N-terminus [25–27] . On the other hand , LAPTM4A tagged with the green fluorescent protein ( GFP ) on its C-terminus was shown to be localized in the Golgi in HeLa cells [34] . Here , we found that LAPTM4A , with a C-terminal HA tag , is primarily localized in the Golgi in three different cell lines . The exact function of LAPTM4A remains to be established . It has been shown that LAPTM4A expression in yeast confers a multidrug-resistance phenotype and altered subcellular distribution of steroids , suggesting that LAPTM4A is involved in the transport of nucleosides or some small molecules [24 , 25 , 46] . LAPTM4B is up-regulated in many human cancer cells and has been suggested to be involved in recruitment of the leucine transporter to lysosomes , regulation of epidermal growth factor receptor lysosomal sorting and degradation , and facilitating Cer removal from late endosomes [47–49] . Despite the homology between LAPTM4A and LAPTM4B , only LAPTM4A is required for Gb3 synthesis . By switching different domains between LAPTM4A and LAPTM4B , we showed that the specificity is encoded within the second luminal domain of LAPTM4A . Many GTs require other proteins for their stability/trafficking/activity . For instance , the core 1 β3galactosyltransferase ( C1GALT1 ) requires the ER-localized membrane protein Cosmc ( C1GALT1C1 ) as a molecular chaperone , which is essential for its proper folding and trafficking [50–52] . However , LAPTM4A does not appear to be required for stability/trafficking of A4GALT , as the Golgi-localization and expression levels of A4GALT are not affected in LAPTM4A KO cells . The soluble enzymatic domain of GTs involved in glycolipid biosynthesis often showed no activity in vitro , suggesting that their activity requires a proper membrane environment and/or “activator” proteins to present the lipid acceptor to GTs . This requirement of an “activator” might be because the headgroups of lipid acceptors are difficult for the soluble enzymatic domain of GTs to reach . This is analogous to the well-established finding that degradation of sphingolipid requires cofactor proteins in addition to glycosidases . These cofactors are known as sphingolipid activator proteins , which are small enzymatically inactive proteins that present glycolipid substrates to glycosidases [53] . This activator function is similar to the previously proposed “add-on” domain for GTs [54] . For instance , it has been shown that α-lactalbumin can serve as an “add-on” domain for β4-galactosyltransferase , and formation of the complex changes the specificity of the acceptor [55] . We propose that LAPTM4A may act as an activator specifically for A4GALT , and that its second luminal domain facilitates recognition of the glycolipid substrates by A4GALT . A similar case is that of Drosophila β-1 , 4-N-acetylgalactosaminyltransferase B , which is responsible for synthesis of glycolipids and requires a six-pass membrane protein GABPI for its activity [56 , 57] . Furthermore , the two short luminal domains of GABPI ( 12 and 35 amino acids long ) are required for the activity of this GT . GABPI has no mammalian homologs . LAPTM4A thus represents the first potential activator for glycolipid GTs identified in mammals , suggesting that the requirement of an activator could be a conserved feature for glycolipid GTs . GABPI is a member of the DHHC protein family , which are palmitoyltransferases , but it does not possess palmitoyltransferase activity , suggesting that it is utilized as a GT “activator” independent of its evolutionary origin [56] . It is possible that LAPTM4A has other functions independent of serving as an “activator” for A4GALT , as LAPTM4A is widely expressed across different tissues . Our screens also identified TMEM165 and TM9SF2 as two host factors required for both Stx and ricin intoxication . TMEM165 has been proposed to maintain Mn2+ homeostasis within the Golgi , although its exact function remains to be fully established . Mn2+ is an essential cofactor for many Golgi GTs involved in both glycoprotein and glycolipid synthesis . It has been shown that TMEM165 deficiency leads to severe reduction in galactose and GalNAc in both glycoproteins and glycolipids [29] . This reduction can be rescued fully by Mn2+ and partially by galactose supplementation [29 , 58] . Consistently , we found that Mn2+ supplementation restored binding of Stx to TMEM165 KO cells . Furthermore , TMEM165 KO cells become more sensitive to toxicity from high concentrations of Mn2+ . Together , these findings further support that TMEM165 plays a critical role in Mn2+ homeostasis in the Golgi . The function of TM9SF2 remains a mystery . TM9SF2 is expressed in all tissues and is evolutionarily conserved . It belongs to a family of highly conserved membrane proteins , with three members in yeast ( Tmn1 to Tmn3 ) , Drosophila ( TM9SF2 to TM9SF4 ) , and Dictyostelium discoideum ( Phg 1A to 1C ) and four members in mammals ( TM9SF1 to TM9SF4 ) . The sequence features of this family suggest that they could function as channels or small-molecule transporters . Defects in TM9SF2 family members have been associated with two types of defects: ( 1 ) disruption in phagocytosis and endosomal trafficking: it has been shown that the absence of Phg 1A strongly impaired both cell adhesion and phagocytosis of bacteria in Dictyostelium , and both TM9SF4 and TM9SF2 are required for phagocytosis in Drosophila as well [59 , 60] . Furthermore , deletion of Tmn family members in yeast has also resulted in disruption of endosomal sorting [61] . Consistent with these results , we found that TM9SF2 KO cells showed defective endosomal trafficking . ( 2 ) Altering metal ion homeostasis: it has been reported that deletion of all three yeast homologs resulted in a 75% reduction in cellular copper ( Cu ) and a 50% increase in cellular Mn2+ [62] . This disruption in metal ion homeostasis may affect the glycosylation process in the Golgi , reducing the amount of both glycoproteins and glycolipids . Consistently , we found that TM9SF2 KO cells showed low levels of glycosphingolipids . The Golgi location of TM9SF2 suggests that the observed defects in endosomal trafficking might be an indirect effect , although it remains possible that TM9SF2 is also distributed at low levels on endosomes and is required for endosomal trafficking directly . Globo-series glycosphingolipids are unique in that their expression is highly restricted . Our studies revealed that biosynthesis of Gb3 not only requires its GT but also a membrane protein , LAPTM4A , which may provide an additional layer of regulation to ensure the restricted expression of Gb3 . Our studies also revealed two Golgi-localized factors required for maintaining a proper environment for GT activities within the Golgi . These findings further indicate that the glycosylation process in the Golgi is tightly controlled and regulated . Further research into the function of these factors will contribute to our understanding of the glycosylation process and CDG . Finally , our findings have also provided new potential therapeutic targets for treating Stxs and ricin intoxication .
The following reagents were purchased from commercial vendors: ricin ( Vector Laboratories , L-1090 , Burlingame , CA ) , Ctx ( List Biological Laboratories , #101C , Campbell , CA ) , CtxB Alexa Flour 555 conjugate ( Invitrogen , C34776 , Waltham , MA ) , MTT ( Sigma , M5655 , St . Louis , MO ) , NBD-Sph ( Avanti , 810205 , Alabaster , AL ) , NBD-C6-Cer ( Abcam , ab144090 , Cambridge , MA ) , and NBD-PS ( Avanti , 810192 , Alabaster , AL ) . Antibodies for the following antigens were obtained from the indicated vendors: the A domain of Stx1 ( Stx1A , List Biological Laboratories , #761L , Campbell , CA ) , Actin ( Aves Labs , ACT-1010 , Tigard , OR ) , HA-tag ( BioLegend , 901502 , San Diego , CA ) , FLAG-tag ( Sigma , F3165 , St . Louis , MO ) , Rab5 ( Abcam , ab13253 ) , Rab7 ( Abcam , ab137029 ) , Giantin ( Abcam , ab80864 and ab37266 ) , TGN46 ( Abcam , ab50595 ) , Sec61A ( Abcam , ab183046 ) , Lamp1 ( Abcam , ab24170 ) , and A4GALT ( Abcam , ab98998 ) . A chicken polyclonal anti-TM9SF2 antibody was generously provided by Dr . Yusuke Maeda ( Osaka , Japan ) [33] . The following cell lines were all originally obtained from ATCC with their catalog number noted: ACHN ( CRL-1611 ) , HeLa ( CCL-2 ) , HT-29 ( HTB-38 ) , Caco-2 ( HTB-37 ) , U2OS ( HTB-96 ) , A498 ( HTB-44 ) , A549 ( CRM-CCL-185 ) , 5637 ( HTB-9 ) , T24 ( HTB-4 ) , RT4 ( HTB-2 ) , and HEK293T ( CRL-3216 ) . Shiga toxin clones ( pLPSH3 for Stx1 and pJES120 for Stx2 ) were generously provided by Dr . Alison O’Brien ( Bethesda , MD ) . Protective antigen ( PA ) of Anthrax toxin was cloned into pET21a vector ( Novagen ) with a His6-tag at C-terminal . pET15b-LFn-DTA was obtained from Addgene ( #11075 ) . Recombinant PA and LFn-DTA were expressed in E . coli ( BL21 strain ) and purified as His-tagged proteins . The cDNAs of LAPTM4A ( 2958870 ) , A4GALT ( 3913851 ) , LAPTM4B ( 5264567 ) , TMEM165 ( 8143981 ) , TM9SF2 ( 40146324 ) , EMC1 ( 4831005 ) , and B4GALT5 ( 30915234 ) were purchased from GE Dharmacon . SgRNA-resistant full-length LAPTM4A , LAPTM4B , TM9SF2 , and EMC1 with triple-HA tag at their C-termini ( with EFGSGSGS as linker ) ; full-length LAPTM4A , TMEM165 , and EMC1 with triple-HA tag at their N-termini ( with GSGSGSEF as linker ) ; full-length A4GALT and B4GALT5 with triple-FLAG tag at their N-termini ( with GSGSGSEF as linker ) were cloned into ether pcDNA3 . 1 vector ( Invitrogen , V800-20 ) or pLenti-Hygro vector ( Addgene , #17484 ) . The LAPTM4A ( LA ) -LAPTM4B ( LB ) chimeric proteins were constructed as the following: AB1 ( LB1–173–LA181–233 ) , AB2 ( LB1–153–LA161–233 ) , AB3 ( LB1–120–LA129–233 ) , AB4 ( LB1–120–LA129–160–LB154–226 ) , AB5 ( LA1–29–LB26–226 ) , AB6 ( LB1–25–LA30–49–LB47–71–LA83–102–LB93–99–LA109–128–LB121–152–LA161–180–LB174–226 ) , AB7 ( LB1–25–LA30–180–LB174–226 ) , AB8 ( LB1–46–LA50–160–LB153–226 ) , and AB9 ( LB1–71–LA83–128–LB121–226 ) , with triple-HA tag at their C-termini . These constructs were cloned into pLenti-Hygro vector via Gibson Assembly ( NEB , E2621 ) . The A4GALT ( A4 ) -B4GALT5 ( B4 ) chimera construct A43B352 ( A41–43–B436–388 ) with triple-FLAG tag at the N-termini ( with GSGSGSEF as linker ) was cloned into pcDNA3 . 1 vector via Gibson Assembly . Lentiviral sgRNA plasmid libraries were generated using the human GeCKO-V2 sgRNA library ( Addgene , #1000000049 ) . The GeCKO-V2 library is composed of two sub-libraries ( sub-library A and sub-library B ) . Each sub-library contains three unique sgRNA per gene and was independently packed into lentiviral libraries . The titer of lentiviral libraries was calculated as colony-forming units ( CFU ) per mL . The cell representation was 500 , so each sgRNA on average is distributed into 500 cells . The multiplicity of infection ( MOI ) was 0 . 2 . Polybrene ( 8 μg/mL , Santa Cruz , sc-134220 ) was also added to the medium to increase viral transduction efficiency . Cells were cultured in virus-containing medium for 2 d . Infected cells were selected with Puromycin ( 5 μg/mL , Thermo Scientific , A1113830 ) ; 3 . 3 × 107 ( for sub-library A ) or 2 . 9 × 107 ( for sub-library B ) cells were either saved as Round 0 ( R0 ) samples , or seeded onto four 15-cm culture dishes and exposed to toxins . The survival cells were reseeded and cultured with normal medium without toxins until about 70% confluence . Cells were then subjected to the next round of selection . The remaining cells were harvested . The genomic DNA was extracted using a commercial kit ( Qiagen , 13323 , Gaithersburg , MD ) . DNA fragments containing the sgRNA sequences were amplified by PCR using primers lentiGP-1_F ( AATGGACTATCATATGCTTACCGTAACTTGAAAGTATTTCG ) and lentiGP-3_R ( ATGAATACTGCCATTTGTCTCAAGATCTAGTTACGC ) . NGS was performed by a commercial vendor ( Genewiz , Illumina MiSeq , South Plainfield , NJ ) . The selected sgRNA sequences ( S2 Table ) were cloned into LentiGuide-Puro vectors ( Addgene , #52963 ) . ACHN and 5637 cells that stably express Cas9 were generated using LentiCas9-Blast construct ( Addgene , #52962 ) and were selected using Blasticidin S ( 10 μg/mL , RPI , B12150 . 01 ) . HeLa-Cas9 was generously provided by Dr . Abraham Brass ( Worcester , MA ) . These Cas9-expressing cells were transduced with lentiviruses that express selected sgRNAs . The mixed stable cell lines were selected using Puromycin . Single clone of KO cells were generated by diluting the mixed KO cells at about 0 . 8 cell per well in 48-well plates . The genotype of single-cell clones were determined by amplifying the DNA fragments containing the sgRNA targeting region using the following primers: LAPTM4A_F ( CCACAGGTAGTCTCCCACTATTTTATATCTTTGTTACTTC ) , LAPTM4A_R ( GCCTAGGAGTCTCTACCACATCGC ) , TMEM165_F ( TCCGTTGGCCACCATTTTTAGTGCTTCTGAA ) , TMEM165_R ( ATGACTTTGTATTTTGCTAACTTCTACACAGGTG ) , TM9SF2_F ( GAGGAGAGATGTGGTACATAGGACTTGGAG ) , TM9SF2_R ( CTGCCCTTTAAGCCACCGTCTTAAC ) , followed by ligating the PCR product into T-vectors ( Promega , A3600 , Madison , WI ) . The ligation products were transformed into E . coli ( DH5α strain ) and plated onto agar plates . Twenty single clone colonies were selected , and their plasmids were extracted and sequenced . Cells were seeded in 96-well microplates and incubated overnight until about 70% confluence . The medium was replaced with 100 μL toxin-containing medium , with a total of 10 serial dilutions ( 10-fold ) . The cells were further incubated for 72 h in toxin-containing medium and then 10 μL MTT ( 5 mg/mL in PBS ) was added to each well and incubated at 37 °C for 4 h . A total of 100 μL solubilization solution ( 10% SDS in 0 . 01 M HCl ) was then added to each well and incubated overnight . The absorbance of formazan formed was then measured at 580 nm by a microplate reader ( BMG Labtech , FLUOstar Omega ) . A vehicle control without toxins and a blank of PBS were analyzed in parallel . The cytotoxicity curves were analyzed and fitted using Origin software ( version 8 . 5 ) . Cells were seeded onto glass coverslips in 24-well plates and incubated for 24 h until about 70% confluence . Cells were washed three times with ice-cold PBS , fixed with 4% paraformaldehyde ( PFA ) for 20 min at room temperature ( RT ) , permeabilized with 0 . 3% Triton X-100 for 30 min , and blocked with 10% goat serum for 40 min , followed by incubation with primary antibodies ( 1 h ) and fluorescence-labeled secondary antibodies ( 1 h ) . Slides were sealed within DAPI-containing mounting medium ( SouthernBiotech , 0100–20 ) . We note that for labeling endogenous TM9SF2 , cells were fixed in methanol at −20 °C for 10 min , as this antibody did not work on samples fixed by PFA . For the toxin surface binding assay , cells were incubated with toxin-containing medium ( 4 . 8 μg/mL Stx1 , or 2 ng/mL Ctx-555 ) on ice for 60 min . Cells were washed , fixed , and subjected to immunostaining without permeabilization . Stx1 polyclonal antibody was used to recognize surface-bound Stx1 . Fluorescent images were captured with the Olympus DSU-IX81 Spinning Disk Confocal System . Images were pseudocolored and analyzed using ImageJ . Cells were harvested and washed three times with PBS . The cell pellets were lysed with RIPA buffer ( 50 mM Tris , pH 7 . 5 , 1% NP40 , 150 mM NaCl , 0 . 5% sodium deoxycholate , 1% SDS , protease inhibitor cocktail ) . Cell lysates were centrifuged and the protein amounts in supernatants were measured by BCA assay ( Thermo Scientific , 23225 , Waltham , MA ) . The supernatants were heat denatured for 5 min , subjected to SDS-PAGE , and transferred onto a nitrocellulose membrane . The membrane was blocked with a TBST buffer ( 10 mM Tris , pH 7 . 4 , 150 mM NaCl , 0 . 1% Tween-20 ) containing 5% skim milk at RT for 1 h . Then , the membrane was incubated with the primary antibodies for 1 h , washed , and incubated with secondary antibodies for 1 h . Signals were detected using the enhanced chemiluminescence method by a Fuji LAS3000 imaging system . For Stx and Ctx binding , cells were incubated with toxin-containing medium ( 4 . 8 μg/mL Stx1 , or 2 ng/mL Ctx-555 ) on ice for 60 min , washed three times with ice-cold PBS , and collected . Cell pellets were fixed with 4% PFA for 20 min at RT and blocked with 10% goat serum for 40 min . Stx1 polyclonal antibody was used to recognize surface-bound Stx1 ( 1 h ) . Cells were washed and incubated with Alexa488-labeled secondary antibody for 1 h , washed twice , and subjected to single-cell sorting using a Canto II FACS system ( BD Biosciences ) . Cells not treated with toxins were used as controls . For cell surface HS , cells were collected with 1 mM EDTA in PBS and subsequently resuspended in PBS with 1% BSA . Cells were incubated with either the 10E4 monoclonal antibody against HS ( 1:400 , amsbio , 370255 , Cambridge MA ) or mouse IgM ( 1:200 , abcam , ab18401 , Cambridge , MA ) for 1 h on ice . Cells were washed twice with PBS and incubated with Alexa488-labeled secondary antibody for 1 h on ice and washed twice , followed by single-cell sorting . Single cells were gated and analyzed using FlowJo software ( version 10 , FlowJo , Ashland , OR ) . Cellular concentrations of cAMP were examined using the Direct Immunoassay Detection Kit ( Abcam , ab138880 , Cambridge , MA ) . Briefly , cells were treated with Ctx ( 50 μg/mL , 4 h ) , lysed , and the protein amounts were measured . The cell lysates were loaded into anti-cAMP antibody-coated plates . The HRP-linked cAMP was added to compete with the cellular cAMP . The activity of HRP-cAMP conjugate was measured using a microplate reader . A calibration curve of free cAMP was analyzed in parallel . The cAMP concentrations in different cells were normalized by the total protein amount in cell lysates . HEK293T cells were co-transfected with plasmids encoding FLAG-tagged A4GALT , B4GALT5 , or a chimeric protein A43B352 , together with HA-tagged LAPTM4A , LAPTM4B , or AB4 at a 1:1 ratio . Cells were harvested 48 h later and cell lysates subjected to co-IP assays using anti-FLAG magnetic beads ( Sigma , M8823 , St . Louis , MO ) . Briefly , cells were lysed with RIPA buffer and incubated with anti-FLAG beads overnight at 4 °C . Beads were then washed ( 0 . 1% Triton X-100 in PBS ) , pelleted , and boiled in SDS sample buffer ( 100 mM Tris , pH 8 . 0 , 4% SDS , 10% glycerol , 0 . 1% bromophenol blue ) . After centrifugation , the supernatant ( Pull-down ) as well as the whole cell lysates ( Input ) were subjected to immunoblot analysis . Briefly , 3 . 5×106 cells from each cell line were washed with 0 . 1% ammonium acetate and suspended in 0 . 35 mL of deionized water . A mixture of methanol ( 1 . 25 mL ) and chloroform ( 0 . 675 mL ) was then added , together with 3 , 000 ng lipid standards d3-GM2 ( d18:1/18:0 , Matreya , #2051 , State College , PA ) . The sample was vortexed for 1 min , followed by incubation for 30 min and centrifugation at 2 , 000g for 30 min at RT . Supernatant ( the methanol/chloroform fraction ) was collected . The remaining cell pellet was resuspended in 250 μL of water , and lipids were extracted again using a methanol/chloroform mixture ( 2:1 ratio , 1 mL ) . The supernatant fractions were combined , dried under nitrogen gas , and stored at −20 °C until analysis . Lipid extracts were resuspended in 100 μL of 60% methanol and 40% water . For each sample , a total of 5 μL was injected into a Kinetex column ( C18 , 1 . 4 μm , 100 Å , 2 . 1×50 mm; Phenomenex , Torrance , CA ) using a UPLC ( Waters Corporation , Milford , MA ) coupled to a Waters Synapt G2-si quadrupole time-of-flight mass spectrometer fitted with an electrospray ionization source operating in positive ion mode . LC separation was carried out at a flow rate of 0 . 27 mL/min using mobile phase A: 0 . 1% formic acid , 5 mM ammonium formate in water , and mobile phase B: 0 . 1% formic acid , 5 mM ammonium formate in methanol , using the gradient conditions as follows: 0–1 min ( 60% B ) , 1–2 min ( 60%–70% B ) , 2–40 min ( 70%–100% B ) , 40–43 min ( 100% B ) , 43–43 . 1 min ( 100%–60% B ) , and 43 . 1–50 min ( 60% B ) . Gangliosides were analyzed in negative ion mode . For each sample , a total of 10 μL was injected onto a Kinetex column and LC-MS system described above . Mobile phase A , 0 . 1% ( v/v ) formic acid in water , and mobile phase B , 0 . 1% ( v/v ) formic acid , methanol , isopropanol ( 5/47 . 5/47 . 5 ) at a flow rate of 0 . 23 mL/min were used for elution using the following gradient conditions: 0–1 min ( 60% B ) , 1–12 min ( 100% B ) , 12–14 min ( 100% B ) , 14–14 . 1 min ( 60% B ) , and 14 . 1–20 min ( 60% B ) . Mass calibration and external lock mass correction were carried out using Glu-1-Fibrinopeptide B . For each lipid detected in positive ion mode ( Gb3 , LacCer , GlucCer , and Cer ) , protonated , sodium adduct , and dehydrated ions were detected . Extracted ion chromatograms obtained using a 15-ppm window centered on the theoretical ionic mass of glycosphingolipids were integrated using TargetLynx XS ( Waters Corporation ) ; summing of peak areas of the corresponding adducts and further data processing was carried out in Excel ( Microsoft , Redmond , WA ) . Endogenous PC was used as an internal standard to obtain peak area ratios for Gb3 , LacCer , GlcCer , and Cer . The corrected peak areas and that of the internal standard d3-GM2 were used to obtain peak area ratios of GM2 . Relative standard deviations ( RSDs ) for the replicate analyses were within 15%–20% . Cells were precooled on ice and exposed to fluorescently labeled lipids ( NBD-Sph , NBD-C6-Cer , and NBD-PS , 5μM ) in serum-free medium containing DF-BSA ( 5 μM ) for 40 min at 4 °C . Cells were washed twice with PBS and incubated for the indicated time in serum-free medium containing DF-BSA ( 5 μM ) . Cells were then washed three times with ice-cold PBS and fixed with 4% PFA for 20 min at RT . Slides were sealed within DAPI-containing mounting medium . HEK293T cells were transfected with HA-tagged LAPTM4A or EMC1 . After fixation in 4% PFA , cells were permeabilized with either Saponin buffer ( 0 . 1% Saponin , 0 . 1% BSA in PBS ) for 30 min at RT or Digitonin buffer ( 5 μg/mL Digitonin , 0 . 3 M Sucrose , 0 . 1 M KCl , 2 . 5 mM MgCl2 , 1 mM EDTA , 10 mM HEPES , pH 6 . 9 ) for 15 min at RT . Cells were then subjected to immunofluorescent staining analysis . The total cellular RNA was extracted by TRIzol ( Invitrogen , 15596026 , Waltham , MA ) , quantified , and subjected to a reverse transcription reaction ( Applied Biosystems , 4375575 , Foster City , CA ) . The cDNA was quantified and subjected to a qPCR reaction . The reactions were run in triplicate on 96-well plates with an ABI Prism 7700 Sequence Detection System ( Applied Biosystems ) . SYBR Green ( Roche , 04913850001 ) was used to monitor dsDNA synthesis . Glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) was used as a housekeeping control to normalize the relative mRNA level . The primers used were as follows: A4GALT_RT_F ( GAGACTTCAGACCGGACCAA ) , A4GALT_RT_R ( AAGCCCTTTCATCAGGACCA ) , LAPTM4A_RT_F ( CAAGTGGGTTGGCTGATTCC ) , LAPTM4A_RT_R ( AGGCCAGGAGGTCATCTTTG ) , GAPDH_RT_F ( AGGGCTGCTTTTTAACTCTGGT ) , and GAPDH_RT_R ( CCCCACTTGATTTTGGAGGGA ) . | Shiga and Shiga-like toxins ( Stxs ) are a family of bacterial toxins and key virulence factors for the bacteria Shigella dysenteriae and enterohemorrhagic Escherichia coli ( EHEC ) , which cause food poisoning throughout the world . Ricin is a plant toxin and a potential bioterrorism agent . Stxs recognize the host receptor glycolipid Gb3 , while ricin utilizes a broad range of glycans as receptors . Here , we carried out genome-wide loss-of-function CRISPR-Cas9 screens using human cells to identify factors required for Stxs and ricin . Besides host factors previously known to be involved in the action of these toxins , our screens revealed three previously poorly characterized Golgi proteins: LAPTM4A , which is specifically required for Stxs , and TMEM165 and TM9SF2 , which are required for both Stxs and ricin . Further characterization demonstrates that LAPTM4A is specifically required for biosynthesis of the glycolipid Gb3 , potentially acting as an “activator” protein for the glycosyltransferase that synthesizes Gb3 , whereas TMEM165 and TM9SF2 are likely required to maintain a proper environment within the Golgi for optimal activity of glycosyltransferases . These findings provide mechanistic insights to glycolipid biosynthesis and regulation of glycosylation levels in the Golgi and also reveal novel therapeutic targets for preventing Stxs and ricin intoxication . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"binding",
"glycolipids",
"cell",
"physiology",
"medicine",
"and",
"health",
"sciences",
"methods",
"and",
"resources",
"toxins",
"cell",
"viability",
"testing",
"pathology",
"and",
"laboratory",
"medicine",
"sphingolipids",
"toxic",
"agents",
"toxicology",
"ricin",
"bioassays",
"and",
"physiological",
"analysis",
"glycosylation",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"lipids",
"cell",
"analysis",
"proteins",
"biochemistry",
"immunostaining",
"cell",
"biology",
"post-translational",
"modification",
"biology",
"and",
"life",
"sciences",
"glycobiology"
] | 2018 | Genome-wide CRISPR screens for Shiga toxins and ricin reveal Golgi proteins critical for glycosylation |
Bacteria can arrest their own growth and proliferation upon nutrient depletion and under various stressful conditions to ensure their survival . However , the molecular mechanisms responsible for suppressing growth and arresting the cell cycle under such conditions remain incompletely understood . Here , we identify post-transcriptional mechanisms that help enforce a cell-cycle arrest in Caulobacter crescentus following nutrient limitation and during entry into stationary phase by limiting the accumulation of DnaA , the conserved replication initiator protein . DnaA is rapidly degraded by the Lon protease following nutrient limitation . However , the rate of DnaA degradation is not significantly altered by changes in nutrient availability . Instead , we demonstrate that decreased nutrient availability downregulates dnaA translation by a mechanism involving the 5' untranslated leader region of the dnaA transcript; Lon-dependent proteolysis of DnaA then outpaces synthesis , leading to the elimination of DnaA and the arrest of DNA replication . Our results demonstrate how regulated translation and constitutive degradation provide cells a means of precisely and rapidly modulating the concentration of key regulatory proteins in response to environmental inputs .
The ability of cells to arrest their growth and proliferation in response to nutrient depletion or stressful conditions is typically critical for their survival . Growth arrest requires global changes in protein synthesis leading to a decline in the production of cellular mass . Importantly , growth arrest also usually demands a concomitant cessation of cell cycle processes , including DNA replication . The mechanisms that modulate cell cycle progression following nutrient limitation remain poorly understood . A promising candidate for transducing information about nutritional status to the bacterial cell cycle is DnaA , the conserved replication initiator protein . DnaA is a AAA+ ATPase required for replication initiation in most bacteria [1] . It directly binds to and unwinds the origin of replication and subsequently recruits replisome components . Several mechanisms have been reported in different bacteria that modulate DnaA activity to ensure the correct timing of DNA replication initiation [2] . One major mechanism , first elucidated in Escherichia coli and generally referred to as RIDA ( regulatory inactivation of DnaA ) , involves ATP binding and hydrolysis [3 , 4] . Binding to ATP favors an active conformation that allows for the assembly of DnaA into an oligomeric structure that promotes duplex unwinding [1 , 5] . After initiation , ATP hydrolysis by DnaA can be stimulated ( in E . coli ) by the protein Hda bound to the DNA-loaded replicase clamp [6] . ATP hydrolysis inactivates DnaA and thereby helps prevent the re-initiation of DNA replication [4 , 7] . RIDA likely operates in other proteobacteria; additionally DnaA activity or its access to the origin can be regulated in some bacteria by interacting proteins or sequestration mechanisms [2] . Although there has been considerable progress in understanding how DnaA and replication initiation is coordinated with other cell cycle events , much less is known about how DnaA activity and DNA replication initiation are coordinated with changes in growth rate . One long-standing hypothesis posits that in steady-state growing cultures replication initiation is triggered at a constant cell-mass-to-origin ratio such that growth rate is intrinsically coupled to replication [8] . However , the precise mechanism responsible for this phenomenon remains unclear , and a number of studies have challenged this model [9–12] . Furthermore , it remains largely unexplored how DNA replication initiation is controlled during the transition from exponential growth to an arrested state , for example at entry into stationary phase , or during the onset of starvation upon nutrient limitation . The α-proteobacterium Caulobacter crescentus is an important model system for understanding the bacterial cell cycle . Caulobacter cells are inherently asymmetric such that each cell division yields two distinct daughter cells , which differ with respect to their morphological and reproductive fates [13] . While one daughter , the stalked cell , initiates DNA replication immediately after cell division , the other daughter , the motile swarmer cell , is arrested in G1-phase and cannot initiate until after differentiation into a stalked cell . The replicative asymmetry of Caulobacter daughter cells ultimately stems from the asymmetric activation of CtrA , a response regulator that directly binds to and silences the origin of replication in swarmer , but not stalked cells [14 , 15] . CtrA is not critical , however , for preventing the re-initiation of DNA replication before cell division; like E . coli , and most other bacteria , the periodicity of replication initiation is dictated primarily by DnaA [16–18] . Similar to E . coli , a major mechanism controlling DnaA activity in C . crescentus is the stimulation of ATP hydrolysis upon initiation [17 , 19] . In contrast to E . coli and Bacillus subtilis , both of which possess multi-fork replication under fast-growth conditions [20 , 21] , Caulobacter daughter cells are both born with one chromosome that replicates once-and-only-once per cell cycle [22] . Hence , C . crescentus does not require mechanisms to trigger multi-fork replication upon shift to nutrient-rich conditions . Nevertheless , Caulobacter must control the timing of replication initiation and cell division in response to nutritional changes or stress conditions to maintain genomic integrity . Prior studies have shown that the abundance of DnaA decreases rapidly following glucose starvation and on entry into stationary phase [23 , 24] , although the mechanisms responsible are unclear . One study suggested that DnaA proteolysis is stimulated by glucose starvation [23] , with a subsequent study demonstrating that the small signaling molecule ( p ) ppGpp is somehow involved in regulating DnaA stability following nutrient limitation [24] . In contrast to DnaA , CtrA is maintained upon carbon starvation and it was shown that ( p ) ppGpp and inorganic polyphosphate ( polyP ) , another signalling molecule , are required for CtrA stability [25] . Although the mechanism ( s ) regulating DnaA during stationary phase and following carbon starvation remain unclear , recent work has provided insight into how DnaA abundance is adjusted following perturbations to the global state of cellular protein folding [26] . This work showed that the Lon protease degrades DnaA in Caulobacter in vivo and in vitro [26] . Degradation of DnaA by Lon occurs even in optimal growth conditions , but is stimulated even more upon the depletion of the DnaK chaperone or thermal stress , when unfolded proteins accumulate and the heat shock response is induced . Lon synthesis is upregulated as part of the heat shock response and , in addition , unfolded proteins appear to directly stimulate Lon to degrade DnaA [26] . Thus , the induction and stimulation of Lon blocks DNA replication initiation in proteotoxic stress conditions . In E . coli the activity of Lon in degrading ribosomal proteins and the antitoxins of toxin-antitoxin systems is stimulated by ( p ) ppGpp and polyP [27 , 28] . Whether Lon is required to modulate DnaA abundance in Caulobacter during nutrient starvation or stationary phase , and whether ( p ) ppGpp or polyP affect this degradation , remain unexplored . Here , we investigated the mechanisms that drive a decrease in DnaA and DNA replication upon entry to stationary phase and following glucose exhaustion in C . crescentus . Our data demonstrate that Lon-mediated degradation is required in both conditions , but that the rate of proteolysis does not change significantly . Instead , we show that DnaA translation decreases as nutrients become scarce; this decrease in synthesis , combined with constitutive degradation by Lon rapidly eliminates DnaA and prevents DNA replication initiation . This mechanism depends on the 5´-untranslated leader region of the dnaA transcript , but does not depend on ( p ) ppGpp , indicating that another signal produced by nutrient limitation ultimately controls DnaA synthesis .
The Lon protease degrades DnaA in C . crescentus and ensures a G1-arrest in conditions that lead to proteotoxic stress [26] . To investigate whether Lon also eliminates DnaA on entry to stationary phase , we first assessed changes in the steady-state levels of DnaA during the transition from exponential to stationary phase . We took samples from a culture of wild-type cells grown in rich medium to optical densities ( OD600 ) of 0 . 4 , 0 . 8 , 1 . 2 and 1 . 4 , and then measured DnaA levels by semi-quantitative Western blotting . DnaA protein abundance decreased as the culture reached higher optical densities , with the biggest decrease occurring between OD600 0 . 8 and 1 . 2 , when growth had slowed but not fully arrested ( Fig 1A ) . In contrast to DnaA , the abundance of the response regulator CtrA , a negative regulator of DNA replication , remained relatively constant . Consistent with a reduction of DnaA and concomitant maintenance of CtrA , flow cytometry analysis indicated that most stationary phase cells ( 57% ) contained a single chromosome indicating that cells were able to complete on-going rounds of DNA replication and the cell cycle but were blocked for initiating a new round of DNA replication ( Fig 1B ) . Note that cells were analyzed 24 hours after reaching the maximal OD600 of 1 . 5 when cell size and morphology are similar to exponential phase cells ( Fig 1B ) ; prolonged periods in stationary phase of seven days result in the formation of elongated helical cells [29] . We also measured steady-state levels of DnaA at increasing cell densities in a strain containing a deletion of lon ( Δlon ) . The growth-phase dependent downregulation of DnaA was largely abolished in this mutant ( Fig 1A ) ; CtrA was not significantly affected . Flow cytometry analysis indicated that the majority of Δlon cells ( 77% ) were also not able to arrest the cell cycle with a single chromosome . Instead , a considerable number of Δlon cells ( 39% ) grown in stationary phase contained more than two chromosomes and were somewhat filamentous ( Fig 1B ) . Consistent with these flow cytometry data , using the fluorescence repressor-operator system ( FROS ) , which fluorescently marks origins of replication [30] , we observed that Δlon cells grown to stationary phase often contained two or more origins per cell ( Fig 1C ) . By contrast , most wild type cells in stationary phase have only one origin . To investigate whether DNA replication is ongoing in the Δlon mutant in stationary phase , we monitored the localization of the replisome by expressing an ectopic copy of dnaN-YFP . Wild-type cells rarely habor more than one DnaN-YFP focus ( Fig 1D ) . By contrast , a significant number of Δlon cells contained more than one DnaN-YFP foci at OD600 1 . 4 , indicating that in these cells multiple replisomes replicate the DNA . Notably , after 24 hours of growth in stationary phase , in both wild type and Δlon cells DnaN-YFP foci were no longer detectable , suggesting that at this time point DNA replication no longer takes place and that cells arrest with the number of chromosomes that they accumulated early on in stationary phase . Altogether , these data demonstrate that the Lon protease is required to eliminate DnaA at the entry to stationary phase and to ensure that cells contain a single fully replicated chromosome when entering stationary phase . ( p ) ppGpp was previously suggested to affect DnaA accumulation in C . crescentus and E . coli [24 , 25 , 31 , 32] . Furthermore , in E . coli ( p ) ppGpp and polyP can trigger Lon to degrade ribosomal proteins and antitoxins [27 , 28] . To analyze the contribution of ( p ) ppGpp and polyP to the regulation of DnaA abundance and DNA replication during entry to stationary phase , we constructed strains containing a deletion of spoT , which encodes the only ( p ) ppGpp synthase in C . crescentus [24] , or both a spoT deletion and a deletion of ppk1 , which encodes a polyphosphate kinase that drives polyP synthesis [25 , 33 , 34] . Both strains grew to a higher OD600 ( ~2 . 0 ) than the wild type ( OD600 ~ 1 . 5 ) ( Fig 2A ) , supporting earlier reports that an inability to produce ( p ) ppGpp promotes a proliferative mode [25 , 35] . To test if ( p ) ppGpp is responsible for the downregulation of DnaA in stationary phase , we compared DnaA levels in the ΔspoT and ΔspoT/Δppk1 mutants with the wild type . DnaA was eliminated from the mutant cells in a similar manner as in wild-type cells during entry to stationary phase ( Fig 2B ) and in vivo degradation assays showed that the stability of DnaA was nearly identical in wild-type and ΔspoT cells ( S1 Fig ) . The levels of CtrA were also not strongly affected by the deletion of spoT and ppk1 in the conditions tested , suggesting that ( p ) ppGpp does not play a major role in adjusting DnaA and CtrA levels during the entry to stationary phase ( Fig 2C ) . Consistent with the clearing of DnaA and maintenance of CtrA , flow cytometry analysis indicated that cells lacking spoT arrested DNA replication initiation , although many cells arrested with two chromosomes rather than one ( Fig 2D ) . This may indicate a disruption of a later cell cycle step in spoT mutants during the entry to stationary phase . A reduction in DnaA levels during the entry to stationary phase may result from increased proteolysis or decreased synthesis , or both . Previous work showed that proteotoxic stress , resulting from chaperone depletion or acute heat shock , can increase the rate of DnaA proteolysis [26] . To investigate whether the rate of DnaA degradation is affected by growth phase , we measured DnaA stability in vivo by adding chloramphenicol to cells to stop protein synthesis and then assessed DnaA decay rates over time by immunoblotting . As documented previously [26] , DnaA stability decreases significantly , from ~48 min . to ~13 min . , in cells depleted of the chaperone DnaK due to increased degradation by Lon ( Figs 3A and S2 ) . In contrast , the half-life of DnaA at OD600 ~ 1 . 0 was only slightly shorter ( 20 min . ) than at OD600 ~ 0 . 4 ( 23 min . ) ( Figs 3B and S2 ) . Using a simple model for DnaA abundance ( see Methods ) , we determined that a difference in protein half-life of three minutes could produce at most a 25% decrease in DnaA abundance between an OD600 of 0 . 4 and an OD600 of 1 . 5 , 400 min later ( Fig 3C ) . To generate the observed 90% drop in DnaA abundance over the same OD600 range would require that the half-life decreases to ~9 . 1 min by OD600 ~ 1 . 0 ( in comparison to the measured value of 20 min ) . We also measured DnaA stability at OD600 ~ 1 . 2 and did not detect a difference in half-life greater than three minutes when compared to OD600 ~ 0 . 4 ( S3 Fig ) . Thus , we conclude that a change in protein half-life cannot explain the change in DnaA abundance that occurs at the onset of stationary phase . Although growth phase had little effect on DnaA stability , deleting the Lon protease had a strong effect on DnaA stability . In Δlon cells , DnaA had a half-life >120 min . in both exponential and early stationary phase cells ( Figs 3D and S2 ) , reinforcing previous results that DnaA degradation depends strongly on Lon [26] . The stabilization of DnaA in Δlon cells agrees with the finding that Δlon cells fail to timely eliminate DnaA at the entry to stationary phase ( Fig 1A ) . Taken together , our data indicate that Lon is required to efficiently clear DnaA at the onset of stationary phase , but , importantly , that the rate of Lon-dependent degradation of DnaA is not substantially changed upon stationary phase entry . Because regulated degradation does not explain the growth phase-dependent decrease in DnaA abundance , we thought that changes in DnaA levels likely stem from changes in DnaA synthesis . To test this possibility , we used a strain in which the promoter of dnaA and its 5´ untranslated leader region ( 5'UTR ) were replaced by Plac , an IPTG-regulated promoter , and its native leader . Addition of 1 mM IPTG to the growth medium resulted in constitutive dnaA expression with DnaA levels comparable to those seen in wild-type cells grown to exponential phase in rich medium ( S4 Fig ) . We then followed DnaA abundance in this strain from exponential phase into stationary phase . In contrast to the wild type , the Plac-dnaA strain was unable to clear DnaA upon entry to stationary phase , with DnaA levels remaining relatively constant up to an OD600 of 1 . 2 ( Figs 3E and S5 ) . DnaA levels dropped by ~40% once cells were at OD600 ~ 1 . 5 , although DnaA levels were decreased by nearly 90% at the same density in wild-type cells . These data demonstrate that constitutive expression of dnaA is sufficient to bypass the downregulation of DnaA at high cell density , in agreement with our finding that DnaA degradation is not significantly changed upon entry to stationary phase ( Fig 3B ) . Moreover , flow cytometry analysis demonstrated that the number of cells harboring a single chromosome in stationary phase was reduced in the Plac-dnaA strain compared to wild type ( Fig 3F ) . Notably , however , Plac-dnaA cells did not accumulate extra chromosomes as seen with Δlon cells grown to stationary phase ( Fig 1B ) , suggesting that the Δlon phenotype likely results from an increased stability of DnaA and other Lon substrates . Our results with Plac-dnaA strongly suggest that changes in DnaA synthesis cause DnaA levels to drop during entry to stationary phase . Knowing DnaA steady-state levels , DnaA half-life , and cell growth rates at different optical densities allowed us to infer how the rate of DnaA synthesis changes as a function of culture density using a mathematical model . Our modeling predicted that DnaA synthesis drops approximately 20-fold between OD600 0 . 4 and OD600 1 . 5 ( Fig 4A ) . To test if this change in DnaA synthesis results from changes in dnaA transcription or mRNA stability , we measured dnaA mRNA levels using quantitative real-time RT-PCR ( qPCR ) on samples from a culture grown to an OD600 of 0 . 2 , 0 . 4 , 0 . 8 , 1 . 2 , and 1 . 6 . Unexpectedly , dnaA mRNA levels did not vary significantly as a function of culture density ( Fig 4B ) . Even at an OD600 of 1 . 6 , dnaA mRNA levels did not fall below 65% of transcript levels measured in exponential phase . By contrast , katG , a known stationary phase-induced gene [36] was upregulated more than 60-fold at OD600 1 . 6 , and l13p , encoding a ribosomal protein that is repressed during stationary phase , was downregulated more than 50-fold ( Fig 4B ) . Consistent with our qPCR results , DNA microarray analysis showed that dnaA transcript levels were not substantially changed in stationary phase ( 70% of exponential phase levels ) , again in contrast to katG and l13p , which showed significant induction and repression , respectively ( Fig 4C ) . These results show that although the rate of DnaA synthesis strongly declines at the onset of stationary phase , dnaA mRNA abundance does not , implying that DnaA translation is likely the growth-phase regulated step in DnaA synthesis . Incorporating the qPCR results into our mathematical model , we inferred that the rate of dnaA translation during the transition to stationary phase must decline to approximately 5% of the rate during exponential phase growth ( Fig 4A ) . Changes in translation often involve the 5' untranslated region , or leader , of bacterial mRNAs . In C . crescentus , dnaA contains a relatively long 5' leader of 155 nt [37 , 38] , which was previously shown to affect dnaA expression during exponential phase [39] . To test if the 5' leader also plays a role in modulating DnaA synthesis at the onset of stationary phase , we placed the coding region of dnaA under the control of the native dnaA promoter , but without 140 nt of the leader , retaining only the region of the leader containing the native Shine-Dalgarno sequence ( Fig 4D ) . This construct was cloned into a low-copy vector and transformed into a strain in which the chromosomal copy of dnaA could be depleted by growing cells in the absence of xylose . As a control , we used a plasmid in which dnaA is controlled by the entire upstream region of the native dnaA locus , including the promoter and the 5' UTR ( Fig 4D ) . With the control plasmid , DnaA was eliminated when the culture reached high optical density as growth rate starts to decline , as in wild-type cells ( Figs 1A and 4E ) . Strikingly however , with the plasmid lacking the 5' leader , DnaA was no longer downregulated upon entry to stationary phase ( Figs 4E and S6 ) , demonstrating that the 5' leader is required for the growth-phase-dependent decrease in DnaA abundance . In contrast to wild type cells , which arrest the cell cycle with a single chromosome in stationary phase ( Fig 1B ) , the strain carrying the plasmid lacking the 5'UTR was not able to arrest in G1-phase ( Fig 4F ) . To test if the 5' leader of dnaA is sufficient to induce a downregulation of protein abundance at the entry to stationary phase , we placed dnaA , with its 5' leader , under control of a Plac promoter ( Fig 4D ) . A strain harboring this construct grown in the presence of IPTG showed a significant downregulation of DnaA upon entry into stationary phase , similar to that seen in wild type cells ( Figs 4E and S6 ) . This downregulation of DnaA was sufficient to allow cells to arrest the cell cycle with a single chromosome in G1-phase in stationary phase ( Fig 4F ) . In sum , our data suggest that as cells transition from exponential to stationary phase , translation of the dnaA mRNA decreases significantly; because DnaA has a relatively short half-life , due to constitutive degradation by Lon , this drop in translation leads to a relatively rapid decrease in the abundance of the replication initiator and a consequent G1-arrest . A reduction in growth rate during the entry into stationary phase may result from the exhaustion of nutrients or the accumulation of inhibitory waste products , cellular stress , or some combination thereof [40] . We hypothesized that a decrease in nutrient availability might be the signal that ultimately modulates dnaA translation . To test this idea we analyzed DnaA accumulation in growth media containing different amounts of nutrients . M2G , a minimal medium , in which the sole carbon source is glucose , was used as the most nutrient poor medium . We supplemented M2G with increasing amounts of peptone , a pepsin digest consisting of polypeptides and amino acids used in rich media such as PYE . DnaA protein levels during mid-exponential phase were clearly correlated with the complexity of the growth medium ( Fig 5A ) . Increases in the amount of peptone added to M2G were mirrored by increases in DnaA steady-state levels , as measured by Western blotting . In contrast to DnaA , levels of the Lon protease were relatively unaffected by the growth medium ( S7 Fig ) . We performed the same experiment using a Plac-dnaA strain in which dnaA lacks its native promoter and 5' leader . Growing this strain in the presence of 1 mM IPTG caused DnaA levels to be relatively constant and independent of the growth medium ( Fig 5A ) , demonstrating that nutrient-dependent changes in DnaA protein levels likely depend on changes in DnaA synthesis , not proteolysis . Moreover , in a strain containing the construct PdnaA-ΔUTRdnaA-dnaA ( Fig 4D ) in which dnaA is regulated by its native promoter but lacks its usual long leader sequence , DnaA levels did not differ between M2G and PYE ( Figs 5B and S7 ) , strongly suggesting that the 5'UTR leader of DnaA is responsible for nutrient-dependent changes in protein levels . Consistent with this conclusion , we found that the construct Plac-UTRdnaA-dnaA showed a growth-medium-dependent accumulation of DnaA , similar to wild type ( Figs 5B and S7 ) . Neither a deletion of spoT nor a lower temperature , which decreases growth rate , had a significant effect on DnaA abundance in the two different media ( Fig 5C ) . To further examine the correlation between nutrient availability , growth rate and changes in DnaA abundance , we grew C . crescentus in PYE medium , which contained either higher ( 2x PYE ) or lower ( 0 . 5x PYE ) amounts of nutrients , respectively , and followed DnaA abundance along the growth curve . In 2x PYE medium cultures reached stationary phase at OD600 2 . 8; by contrast , growth in 0 . 5x PYE led to a growth arrest at OD600 0 . 8 ( Fig 5D ) . DnaA levels dropped in both conditions concomitantly with the cessation of growth , consistent with the hypothesis that changes in DnaA abundance coincide with nutrient exhaustion and a slowdown of the growth rate ( Figs 5D and S7 ) . Next , we wanted to test if cells that have already reached stationary phase can accumulate DnaA after adding nutrients to the culture . To address this question we added concentrated nutrients ( final concentration 1x , 5x or 10x of nutrients in PYE medium ) to a culture grown for two hours in stationary phase at OD600 ~ 1 . 5 , and then monitored subsequent changes in growth rate , DnaA levels and DNA replication . Addition of 1x PYE led only to a slow increase in growth rate , likely because the fresh nutrients are quickly consumed by the high-density culture leading to a re-entry into stationary phase ( Figs 5E and S7 ) . In this condition DnaA levels remained relatively low . By contrast however , addition of 5x or 10x PYE nutrients allowed cells to resume rapid growth , which was nearly as fast as the growth of a culture that was backdiluted from stationary phase into fresh PYE medium . In these conditions DnaA levels increased within two hours and flow cytometry analysis showed that cells were able to initiate DNA replication ( Figs 5E and S8 ) . These data show that cells that have reached a high OD are able to upregulate DnaA and initiate DNA replication when sufficient amounts of fresh nutrients are added to allow for rapid growth . Altogether these results reinforce our model that DnaA levels and DNA replication are tightly linked to nutrient availability and cellular growth rate . Previous work demonstrated that rapid carbon starvation can also lead to the elimination of DnaA [23–25] . To monitor DnaA levels in starvation conditions , we performed a glucose exhaustion assay in which cultures grown in M2G were shifted to M2G1/10 , which contains 10% of the glucose in M2G . Initially , cells continued growing; however , as glucose in the medium was consumed , the density of the culture stopped increasing , leveling off at OD600 ~ 0 . 25 , approximately 4 hours after the shift to M2G1/10 ( Fig 6A ) [25] . Concomitant with this growth arrest , DnaA levels dropped and cells arrested in G1-phase ( Fig 6A and 6B ) . We measured the stability of DnaA in M2G and four hours after shifting cells to M2G1/10 , the time point when growth arrested and DnaA abundance drfopped most strongly . However , we did not detect changes in DnaA stability ( S9 Fig ) . Cells harboring a Δlon mutation contained approximately 2–3 fold higher DnaA levels and were not able to efficiently clear DnaA upon glucose exhaustion ( Fig 6A ) . Likewise , constitutive expression of DnaA from the Plac promoter abolished downregulation of DnaA during glucose exhaustion . Furthermore , by using the different plasmid-borne constructs that either contain or lack the 5´UTR of dnaA ( Fig 4D ) , we found that the downregulation of DnaA upon carbon exhaustion strongly depended on the presence of this region of the dnaA mRNA ( Figs 6C and S10 ) . Together , these findings suggest that , as in stationary phase , the adjustment of DnaA abundance is mediated by the combined effects of regulated translation and constant proteolysis . In the ΔspoT mutant , DnaA was eliminated as in wild-type cells despite growth to a higher final OD600 ( Fig 6A ) , similar to the situation in stationary phase . Flow cytometry analysis showed that glucose starvation led to a G1-arrest in the wild type , but not in the Δlon mutant ( Fig 6B ) . Notably , the Plac-dnaA strain still exhibited a G1-arrest in most cells despite the maintenance of DnaA levels , demonstrating that in this condition the availability of DnaA is not sufficient for DNA replication initiation and that another mechanism exists that blocks replication . One possibility is that DnaA is not active for DNA replication in this condition . Alternatively , CtrA binding to the origin might block DnaA's access to the origin [41] . Indeed , it has been shown previously that CtrA is stabilized in starved swarmer cells [23 , 25] . Consistent with this finding , we observed a significant increase in CtrA levels upon glucose exhaustion in wild type and Plac-dnaA cells ( Fig 6A ) . By contrast , CtrA was not upregulated in Δlon and ΔspoT strains ( Fig 6A ) . Together these data demonstrate that both DnaA and CtrA are tightly , and reciprocally , regulated to ensure that DNA replication does not initiate upon carbon starvation . In addition , the nucleotide bound state of DnaA might be affected under starvation conditions . We tested if the addition of glucose to a carbon-starved culture can restore DnaA levels and DNA replication , performing a similar nutrient re-addition experiment as above ( Fig 5E ) . In this case we shifted wild-type cells from M2G to M2G1/10 to deplete glucose . After growth had been arrested for two hours and DnaA was no longer detectable , we added glucose back to the culture at a final concentration of 0 . 2% . Growth of the culture quickly resumed , with a rate similar to a culture that was kept in M2G throughout the experiment ( Fig 5F ) and DnaA levels were rapidly upregulated after glucose addition ( Figs 5F and S7 ) . Moreover , the number of cells in S-phase quickly increased after glucose addition , indicating that cells started to initiate DNA replication ( S8 Fig ) . Hence , in carbon-starved cells the lack of nutrients is the only reason for low DnaA protein levels; the addition of nutrients rapidly restores DnaA levels and DNA replication .
Our new results demonstrate that DnaA abundance is tightly regulated by two complementary post-transcriptional mechanisms , which adjust the levels of DnaA in response to nutrient depletion . First , decreasing levels of nutrients slow down DnaA synthesis by affecting its rate of translation; second , Lon-dependent degradation allows DnaA concentration to rapidly drop following the changes in translation ( Fig 7 ) . Importantly , the rate of degradation is not significantly affected by changes in nutrient availability . This stands in contrast with proteotoxic stress conditions , which were previously shown to induce the transcription of lon and to cause an accumulation of unfolded proteins that can directly stimulate Lon activity and DnaA degradation [26] ( Fig 7 ) . Although the rate of DnaA proteolysis does not change upon nutrient exhaustion , a fast constitutive rate of proteolysis is still critical for adjusting the level of DnaA in this condition . Cells containing a deletion of lon were unable to clear DnaA and had severe cell cycle defects . Likely , DnaA has evolved a relatively short half-life to allow dynamic changes in its abundance upon environmental inputs through the modulation of the rate of synthesis . Previous proteome-wide studies showed that only a minority of proteins ( approx . 4% in C . crescentus ) are proteolytically unstable [45] , many of which have important regulatory functions , including CtrA , SciP , FtsZ , CcrM and GcrA [45–47] . In other bacteria , well-studied examples of regulatory proteins with short half-lives include the alternative sigma factors σ32 and σS [48] . Intriguingly , the nutrient-dependent regulation of DnaA synthesis does not act at the level of transcription but instead at the post-transcriptional level by a mechanism involving the 5´UTR of the dnaA transcript . A previous study showed that this 5´UTR had a repressing effect on dnaA expression during exponential growth [39] . However , a physiological role has not been elucidated until now . Our new data show that the 5´UTR ensures the downregulation of DnaA synthesis in response to nutrient exhaustion . We hypothesize that a small non-coding RNA or a metabolite , produced in a nutrient-dependent manner , may bind to this leader and thereby induce changes in the mRNA secondary structure , which in turn make ribosome binding and translation either more or less efficient depending on nutrient conditions . Alternatively , a regulatory protein might associate with dnaA mRNA and affect translational efficiency in a nutrient-dependent manner . In particular , when post-transcriptional control is paired with a short half-life of the target protein , as demonstrated here , the regulation of protein abundance is rapid and precise . We propose that such dynamic control of protein abundance could also be utilized for the better design and construction of synthetic circuits , which so far mainly depend on transcriptional mechanisms [49] . Our study investigated the dynamics of DnaA production and degradation under two conditions: at the entry to stationary phase of cultures grown in complex medium and upon exhaustion of the carbon source glucose in cells grown in minimal medium . In both conditions DnaA synthesis was controlled post-transcriptionally in response to nutrient exhaustion and Lon-mediated proteolysis was required to eliminate the protein . Nevertheless , there are also differences between these conditions . In carbon starvation conditions , the inhibition of DNA replication initiation might also depend on upregulation of CtrA , which directly silences the origin of replication [14] , reinforcing previous models that a reciprocal regulation of CtrA and DnaA ensures a coordinated cell cycle block upon growth arrest . Consistent with previous results [25] , our data demonstrate that SpoT plays a role in controlling CtrA abundance . Additionally , our data also now indicate a possible role of Lon in this pathway . How exactly these factors affect CtrA abundance and thereby ensure precise regulation of CtrA abundance in response to changing environmental conditions remains to be studied . In other bacteria DNA replication initiation is likely regulated at the onset of stationary phase and carbon starvation as well . In E . coli it has been observed that cells grown to stationary phase arrest the cell cycle with two or four whole chromosomes [50] , indicating that DNA replication initiation is blocked in this condition . The underlying mechanisms remain unclear [51] . Future studies will help to elucidate if the environmental control of DnaA is conserved among bacteria . An earlier study proposed that ( p ) ppGpp regulates DnaA abundance during carbon starvation [24] . In that previous study , synchronized swarmer cells were transferred to M2 minimal medium without any carbon source . In contrast , we investigated DnaA levels and stability in mixed cultures during a less abrupt nutrient exhaustion , which likely better represents the situation in natural environments . We found that the regulation of DnaA abundance at the onset of nutrient exhaustion was not affected in strains which are unable to produce ( p ) ppGpp , indicating that this signaling molecule is not required for DnaA proteolysis and starvation-induced elimination of DnaA under the conditions tested . In agreement with our data , another recently published study showed that the artificial overproduction of ( p ) ppGpp does not impact DnaA stability [31]; prolonged ( p ) ppGpp overproduction affected DnaA synthesis only moderately and indirectly [31] . Although ΔspoT cells were still able to clear DnaA at the onset of stationary phase or starvation , we observed deficiencies in arresting cell growth and the cell cycle . The inability of ΔspoT cells to arrest the cell cycle might stem from a misregulation of CtrA under starvation conditions . Alternatively , or in addition , other cell cycle processes or replication proteins might be affected by ( p ) ppGpp . In B . subtilis and E . coli , ( p ) ppGpp is known to affect DNA replication elongation by directly inhibiting DNA primase [52 , 53] . Altogether , our results highlight the importance of tightly regulating DNA replication at the onset of adverse conditions demanding growth arrest . The modulation of growth and proliferation is well known to affect bacterial fitness and survival . For instance , entering a non-growing and non-proliferating state has been demonstrated to enhance bacterial drug tolerance and intracellular persistence of pathogenic bacteria [54] . Understanding the regulation of fundamental processes like DNA replication under conditions that require growth arrest is thus important for developing strategies for bacterial growth control .
Wild type C . crescentus NA100 and its mutant derivatives were grown in PYE ( complex medium ) , M2G medium ( minimal medium containing 0 . 2% glucose ) , M2G1/10 medium ( minimal medium containing 0 . 02% glucose ) or in M2G with varying amounts of peptone and yeast extract as indicated in Fig 5 . When necessary , growth medium was supplemented with 0 . 3% xylose , 0 . 2% glucose , 3% sucrose or 1 mM IPTG . For addition of nutrients to a stationary phase grown culture , 10x or 50x stock solutions of PYE were prepared and added as 1:5 or 1:10 dilutions to the stationary-phase grown culture . Note , that the 50x stock contained only the nutrient ingredients ( peptone and yeast extract ) of PYE medium , but not the salts ( MgSO4 , CaCl2 ) . Cultures were grown at 30°C at 200 rpm . Antibiotics were added in the following concentrations as needed for solid and liquid media , respectively: oxytetracycline ( 2 μg ml−1 or 1 μg ml−1 ) , kanamycin ( 25 μg ml−1 or 5 μg ml−1 ) , chloramphenicol ( 1 μg ml−1 or 2 μg ml−1 ) or spectinomycin ( 200 μg ml−1 or 25 μg ml−1 ) . E . coli strains were routinely grown in LB medium at 37°C , supplemented with chloramphenicol ( 30 μg ml−1 or 20 μg ml−1 ) , kanamycin ( 50 μg ml−1 or 30 μg ml−1 ) , oxytetracycline ( 12 μg ml−1 ) , or spectinomycin ( 50 μg ml−1 ) as required . Strains used in this study are listed in S1 Table . Deletions of spoT and spoT/ppk1 in strains ML2389 and ML2390 were created by using the two-step recombination procedure [55] . To generate the spoT deletion , plasmid pNPTS-spoT was introduced into C . crescentus CB15N by electroporation . Clones that had integrated the vector at the spoT locus were selected on PYE plates containing kanamycin . A second recombination step was performed to select for plasmid excision . Single colonies of the first integrants were grown overnight in PYE without kanamycin . After overnight growth , 1 μl was plated for counter-selection on PYE containing sucrose . Sucrose resistant clones were restreaked to test for loss of kanamycin resistance and hence plasmid excision . The resulting clones have either regenerated the wild-type allele or generated the desired in-frame deletion . To distinguish between the two outcomes , PCR was performed to verify deletion of the spoT gene . To generate the spoT/ppk1 strain , the same two-step recombination procedure was performed , except plasmid pNPTS-ppk1 was introduced into the ML2389 ( ∆spoT ) background . To generate strain KJ743 plasmid pNPTS-lon::tetr was introduced into strain KJ300 by electroporation . Integrants were selected on plates containing tetracycline and kanamycin . A second recombination step was performed for plasmid excision . PCR and Western blotting were performed to confirm the deletion of the lon gene . Strains KJ741 and KJ742 were generated by electroporating plasmid pRVYFPC-5:Pvan-dnaN-YFP::tetr into strains CB15N or LS2382 , respectively . Strain ML2000 was generated by introducing a PlacI-lacI cassette 73 bp upstream of the hfaA promoter using the two-step recombination procedure outlined above . Next , 400 bp upstream of dnaA was replaced with the 122 bp Plac promoter also using two-step recombination . Dependence of dnaA expression on IPTG was then confirmed by growing the strain in PYE lacking IPTG , verifying replication arrest by flow cytometry , and then verifying cellular filamentation by phase microscopy . Strains KJ729 , KJ730 and KJ731 were generated by electroporating plasmids pCT133-PdnaA-UTRdnaA-dnaA , pCT133-PdnaA-ΔUTRdnaA-dnaA and pCT133-Plac-UTRdnaA-dnaA into strain GM2471 . Samples from C . crescentus cultures grown in the appropriate conditions were fixed in 70% ethanol . Fixed cells were pelleted at 4000 rpm , resuspended in 50 mM sodium citrate buffer containing 2 μg/ml RNase and incubated at 50°C for 4 hrs or overnight to digest RNA . Samples were diluted and stained with 2 . 5 μM SYTOX green before being analyzed by flow cytometry using a BD LSRII or a LSRFortessa flow cytometer ( BD Biosciences ) . Flow cytometry histograms were processed with FlowJo software . To quantify the number of cells in G1 phase ( 1N ) , with 2N or with a chromosome content >2N , respectively , we used FlowJo . Flow cytometry profiles within one figure were recorded in the same experiment , on the same day with the same settings . The scales of y- and x-axes of the histograms within one figure panel are identical . Each experiment was repeated independently and representative results are shown . Cells were fixed with 0 . 5% paraformaldehyde , pelleted , and resuspended in an appropriate volume of PBS . Fixed cells were mounted onto PYE 1 . 2% agarose pads and phase contrast images taken using a Ti eclipse inverted research microscope ( Nikon ) with a 100x/1 . 45 NA objective ( Nikon ) . For the analysis of fluorescently marked origins or DnaN-YFP foci , YFP emission/excitation filters were used . ImageJ and Adobe Photoshop were used for image processing . Pelleted cells , normalized to the optical density of the culture , were resuspended in 1X SDS sample buffer and heated to 95°C for 10 min . Total protein samples were then subjected to SDS-PAGE for 60 min at 130 V at room temperature on 11% Tris-HCl gels and transferred to PVDF or nitrocellulose membranes . Proteins were detected using primary antibodies against DnaA ( Jonas et al . 2011 ) , DnaK , RpoA , CtrA or E . coli Lon ( kindly provided by R . T . Sauer ) in appropriate dilutions , and a 1:5000 dilution of secondary HRP-conjugated antibody . The primary antibody against C . crescentus DnaA was affinity purified to enhance specificity and to prevent cross-reactivity with C . crescentus Hsp . SuperSignal Femto West ( Thermo Scientific ) was used as detection reagent . Blots were scanned with a Typhoon scanner ( GE Healthcare ) or a Chemidoc ( Bio-rad ) system . Images were processed with Adobe Photoshop , and the relative band intensities quantified with ImageJ software . To measure protein degradation in vivo , cells were grown under the desired conditions . Protein synthesis was blocked by addition of 100 μg/ml chloramphenicol . Samples were taken every 10 min and snap frozen in liquid nitrogen before being analyzed by Western blotting . The effect of varying degradation rates on DnaA abundance was investigated using the following equation: dP ( t ) dt=ks−kd ( t ) P ( t ) where ks is the rate of protein synthesis ( assumed constant ) and kd ( t ) =at+ln223min is a linearly increasing degradation rate , having a value corresponding to a half-life of 23 min at t = 0 ( OD600 0 . 4 ) ( Fig 3B ) . As the half-life is always much shorter than the doubling time , we can safely ignore the effects of dilution due to growth ( or the lack thereof ) . To generate the solid red curve in Fig 3C , we choose a such the half-life at t = 160min ( OD600 1 . 0 ) is 20min ( Fig 3B ) . We fixed ks by assuming that protein levels are in steady state at t = 0 , a reasonable assumption during exponential phase growth . The equation was solved using the ode45 solver of MATLAB ( The MathWorks Inc . ) . To find out how fast degradation at OD600 1 . 0 would have to be in order to explain the data , we used the MATLAB’s constrained non-linear optimization algorithm , fmincon , to find the values of ks and a that result in the best fit to the observed relative DnaA abundance ( Fig 3C , blue line ) as measured by relative least square . This best fit is plotted as the dashed red line in Fig 3C . The value of a found results in a half-life of 9 . 1min at t = 160min ( OD600 1 . 0 ) . In order to estimate the rate of DnaA protein synthesis , we allow ks to vary with time and write ks ( t ) =dP ( t ) dt+kd ( t ) P ( t ) . We can then calculate ks ( t ) point-wise by using the measured protein abundance ( Fig 3E ) and half-lives ( Fig 3B ) . We estimate dP ( t ) dt from a linear fit through the data excluding the overnight time point and take the half-life to be 23 min for the first two time points and 20 min for the last two time points . The resulting ( normalized ) values of the synthesis rate ks , expressed as a function of OD600 , are presented in Fig 4A ( blue line ) . We convert this synthesis rate into an estimated translation rate by dividing each time-point by the normalized mRNA abundance , as measured by qPCR ( red line ) . RNA was collected from bacteria that were grown under the appropriate conditions and extracted using the RNeasy mini kit ( Qiagen ) . The generation of labeled cDNA and hybridization of custom Agilent arrays was performed as earlier described [56] . RNA was collected from bacteria that were grown under the appropriate conditions as described above . Equal amounts of isolated RNA were reverse transcribed into cDNA using the iScript cDNA synthesis kit ( Bio-rad ) . The cDNA was used as template for the real-time PCR reaction using the iTaq universal SYBR Green Supermix ( Bio-rad ) and primers as listed in S2 Table . Analysis was performed with a qTower instrument ( Analytik Jena ) using the standard run mode . For detection of primer dimerization or other artefacts of amplification , a dissociation curve was run immediately after completion of the real-time PCR . Individual gene expression profiles were normalized against 16S RNA , serving as an endogenous control . Relative expression levels were determined with the comparative Ct method . Each qPCR reaction was performed in triplicates . The data shown represent means of at least two independent biological replicates . | The duplication of genetic material is a prerequisite for cellular growth and proliferation . Under optimal growth conditions , when cells strive to grow and divide , DNA replication must be initiated with high frequency . However , under nutrient limiting conditions cells stop initiating DNA replication to ensure cellular integrity . Here , we identify mechanisms responsible for blocking DNA replication initiation under nutrient limitation in Caulobacter crescentus . In this bacterium nutrient limitation results in a strong downregulation of DnaA , the conserved replication initiator protein , which is required for DNA replication in nearly all bacteria . Our data demonstrate that the downregulation of DnaA depends on a reduction in DnaA synthesis in combination with fast degradation by the protease Lon . The changes in DnaA synthesis are mediated by a post-transcriptional mechanism , which adjusts DnaA translation in response to nutrient availability . The constitutively high rate of DnaA degradation then ensures the rapid clearance of the protein following the changes in translation . Our work exemplifies how regulated protein synthesis and fast degradation of key regulatory proteins allow for the precise and dynamic control of important cellular processes in response to environmental changes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Nutritional Control of DNA Replication Initiation through the Proteolysis and Regulated Translation of DnaA |
Faithful execution of developmental gene expression programs occurs at multiple levels and involves many different components such as transcription factors , histone-modification enzymes , and mRNA processing proteins . Recent evidence suggests that nucleoporins , well known components that control nucleo-cytoplasmic trafficking , have wide-ranging functions in developmental gene regulation that potentially extend beyond their role in nuclear transport . Whether the unexpected role of nuclear pore proteins in transcription regulation , which initially has been described in fungi and flies , also applies to human cells is unknown . Here we show at a genome-wide level that the nuclear pore protein NUP98 associates with developmentally regulated genes active during human embryonic stem cell differentiation . Overexpression of a dominant negative fragment of NUP98 levels decreases expression levels of NUP98-bound genes . In addition , we identify two modes of developmental gene regulation by NUP98 that are differentiated by the spatial localization of NUP98 target genes . Genes in the initial stage of developmental induction can associate with NUP98 that is embedded in the nuclear pores at the nuclear periphery . Alternatively , genes that are highly induced can interact with NUP98 in the nuclear interior , away from the nuclear pores . This work demonstrates for the first time that NUP98 dynamically associates with the human genome during differentiation , revealing a role of a nuclear pore protein in regulating developmental gene expression programs .
In eukaryotes , the nuclear envelope ( NE ) forms a membrane barrier around the nuclear genome . All molecular trafficking in and out of the nucleus is mediated by nuclear pore complexes , large multiprotein channels composed of ∼30 different nuclear pore proteins ( Nups ) that span the NE [1]–[3] . In addition to mediating transport , nuclear pore complexes have also been implicated in genome organization and transcriptional regulation [4] . Initial electron microscopy studies suggested that nuclear pore complexes specifically associate with decondensed , transcriptionally active euchromatin in an otherwise highly condensed , heterochromatic nuclear periphery [5]–[7] . Based on these observations , it has been proposed that nuclear pore complexes may interact with active genes to promote the export of their transcripts [7] . Consistent with this hypothesis , several reports have demonstrated that Nups bind active regions of the genome in Saccharomyces cerevisiae and more recently in Drosophila melanogaster [4] , [8]–[17] . In yeast , all Nup-genome interactions identified so far are thought to occur at nuclear pore complexes at the nuclear periphery ( i . e . ‘on-pore’ interaction ) . However , the organization of the nuclear pore complexes is highly dynamic [18] and a subset of mobile Nups has been shown to shuttle on and off nuclear pore complexes , thereby potentially extending the functional reach of Nups . Interestingly , evidence of intranuclear Nups that bind specific regions of the genome has been found in Drosophila suggesting that Nups can also bind chromatin away from the nuclear pores ( i . e . ‘off-pore’ interaction ) [8] , [13] , [17] . In Drosophila embryonic culture cells , Nups predominantly interacted with active genes inside the nucleoplasm , whereas the nuclear pore complexes at the nuclear periphery was associated with repressed genes [17] . Limited studies have been carried out to address whether Nups play an important role in transcription in the mammalian genome . In neonatal rat ventricular cardiomyocytes , NUP155 was found to interact with the histone deacetylase HDAC4 and nuclear pore components associate with a number of HDAC4-target genes [19] . The only study that addressed the potential role of Nups in gene regulation in human cells has shown that nuclear pore complexes preferentially associate with repressive chromatin domains [20] . Combined with studies from fungi and flies , it appears that Nups can interact with both active and silent loci , depending on the cell type or the type of Nups investigated . Therefore , it is tempting to speculate that Nups may dynamically associate with the genome according to developmental stages during differentiation . Accumulating evidence suggests that the organization of the genome is highly dynamic during development [21]–[23] . For example , on a global level , the hyperdynamic and open chromatin organization has been correlated to the differentiation potential of pluripotent cells , and this property is lost upon differentiation . Moreover , on the single-gene level , repositioning of developmental genes and tissue-specific promoters relative to the nuclear periphery during differentiation has been well documented [24]–[30] . The potential involvement of Nups in chromatin-related aspects of developmental regulation is further supported by the findings that mutations in multiple Nups caused specific developmental defects rather than a global deficiency that would have been predicted if the sole role of Nups was to mediate transport in all cell types [31] . Several studies suggest that Nups are involved in developmental gene regulation in lower organisms . In yeast , the mating pheromone alpha factor induces alterations in the association between Nups and specific genomic regions [9] . In Drosophila salivary glands , a subset of Nups including the mobile NUP98 can dissociate from nuclear pores and activate a number of ecdysone-induced genes in the nuclear interior ( i . e . ‘off-pore’ Nup-gene interaction ) . These findings raise several key questions regarding the chromatin-related function of Nups during development . For instance , are Nups involved in establishing gene expression programs in diploid cells of mammalian organisms , especially human , during differentiation of pluripotent cells and establishment of cell fate ? Do Nups relocate to developmentally induced genes on a genome-wide level in human cells ? What are the differences between ‘on-pore’ and ‘off-pore’ Nup-gene interactions in the context of development , and do nuclear pores at the nuclear periphery have a role in developmental gene regulation ? We decided to determine if NUP98 , a nuclear pore complex component that is located on both the cytoplasmic and the nucleoplasmic faces of the nuclear pore complex and has the capacity to move on and off the nuclear pore [18] , [32] , interacts with the human genome . Using chromatin immunoprecipiation sequencing ( ChIP-seq ) we show that NUP98 associates with developmentally regulated genes in stem cells and progenitor cells . In neural progenitor cells , overexpression of full-length NUP98 increases expression levels of a subset of its binding targets , and overexpression of a dominant negative fragment of NUP98 decreases mRNA levels of NUP98-associated genes . In addition , we found that developmental NUP98-gene interactions occur both on nuclear pore complexes and in the nuclear interior . The ‘on-pore’ interactions seem to be enriched for genes involved in the initial stage of developmental induction , whereas the ‘off-pore’ ( i . e . intranuclear ) targets are comprised of genes mediating later stages of developmental induction . We concluded that during human stem cell differentiation , NUP98 associated with specific regions of the genome in a manner that was tightly coupled to the developmental stage . In addition , the nuclear pores appeared to function as a transient platform that supported the initial induction of developmental genes .
To test whether NUP98 can bind to the mammalian genome during cell differentiation , we performed ChIP-Seq experiments on cultured human embryonic stem cells ( ESCs ) , neural progenitor cells ( NeuPCs ) that were differentiated in vitro from ESCs , and neurons that were differentiated in vitro from NeuPCs . We also determined the presence of chromatin-bound NUP98 in lung fibroblast IMR90 cells as an example of another differentiated cell type . As expected , in all four cell types , NUP98 was found both on nuclear pores at the nuclear periphery and intranuclear sites , consistent with its reported capacity to move on and off the nuclear pores ( Figure 1A ) [18] , [32] . We first validated the ChIP-Seq method using IMR90 cells with two independent antibodies against human NUP98 . As expected , both antibodies stained nuclear pores and a few intranuclear sites in IMR90 cells ( Figure S1A ) . Additionally , both proved efficient and specific in western blot and immunoprecipitation experiments ( Figure S1B , S1C ) . Since Nups were not expected to bind directly to DNA , we employed two cross-linking conditions for the ChIP-Seq experiment , formaldehyde single cross-linking and formaldehyde-disuccinimidyl glutarate double-crosslinking in order to more efficiently crosslink indirect Nup-chromatin contacts . After crosslinking , we immunoprecipitated NUP98 using the two antibodies , purified DNA that was immunoprecipitated , and had DNA amplified and subjected to deep sequencing . Sequencing reads were then mapped to the human genome ( Figure S2 ) . The results from the four ChIP-Seq experiments , using two antibodies and two cross-linking conditions , were highly consistent ( Figure S1D–S1F ) , with 73% NUP98-binding regions from pull-down using the first antibody overlapping with 78% NUP98-binding regions using the second antibody . We further validated our results by randomly selecting several NUP98-binding regions called from the ChIP-Seq experiment and confirming the interaction between NUP98 and these regions by ChIP-qPCR ( Figure S1G ) . After validation of the ChIP-Seq method , we extended the ChIP-Seq analysis to human embryonic stem cells , human embryonic stem cell-derived NeuPCs that were ∼90% positive for the neural progenitor cell marker Nestin ( Figure S3 ) , and postmitotic neurons . Interestingly , the genome-binding pattern of NUP98 varied greatly depending on the developmental stage of the cells . NUP98 occupied more genomic regions in ESCs and NeuPCs than in differentiated cells . Further analysis revealed that 71% of NUP98-chromatin sites in ESCs and 74% in NeuPCs were specific for the respective cell-type ( i . e . not found in the other cell types ) ( Figure 1B , 1C ) . The most dramatic difference was found in neurons where essentially no significant enrichment for NUP98 binding could be identified ( Figure S4 and data not shown ) . Together , these findings suggest that Nup98's ability to interact with the human genome is developmentally regulated . We further analyzed whether NUP98-DNA interaction occurred on gene regulatory elements and/or coding regions in ESCs and NeuPCs by assigning NUP98 binding regions to promoters , exons , introns , and intergenic regions . In both ESCs and NeuPCs , NUP98 bound to 500–600 genes ( Figure 1D ) and exhibited a significant enrichment in promoter regions ( Figure 1E ) . It is important to note that the few NUP98 binding sites in IMR90 cells were preferentially found in intergenic regions ( Figure 1F ) , providing additional evidence for a dynamic and developmentally-controlled association of NUP98 with the human genome . Although we cannot rule out that NUP98 binding in IMR90 has functional significance , we decided to focus our analysis on NUP98-bound genes in ESCs and NeuPCs . In order to identify potential DNA sequence motifs and/or potential NUP98-interacting transcription factors that direct NUP98-DNA binding , we analyzed the transcription factor motifs overrepresented in NUP98-binding sequences found in ESCs and NeuPCs . We found that GA-boxes were an evolutionarily conserved NUP98-associated motif . This motif was not only overrepresented in human NUP98-binding genomic regions , but also in published Drosophila NUP98 binding sequences ( Figure 2A , Figure S5A and S5B ) [8] , [17] . In Drosophila , GA-boxes are recognized by GAGA factor , which is a transcriptional activator that is crucial for the proper expression of several homeotic genes [33] . This suggests that the interaction between NUP98 and GA-box motifs , potentially related to the regulation of developmental genes , is evolutionary conserved and further validates our ChIP-Seq results . We also identified YY1 binding site as NUP98-associated motif in ESCs and NeuPCs ( Figure S5C ) . Both GAGA factor and YY1 have been linked to boundary activities , in line with the potential role of Nups in the compartmentalization of chromatin into active and silent domains [31] , [34]–[36] . The binding motif of nuclear DEAF-1 related ( NURD ) /homolog to Drosophila DEAF-1 is also a NUP98-associated motif enriched in both ESCs and NeuPCs ( Figure S5C ) . NURD displays homology to the protein SP100 , a component of the promyelocytic leukemia-associated nuclear body , implying that NUP98 might be involved in the regulation of nuclear bodies and is consistent with the reported link of NUP98 to leukemia [37]–[39] . Moreover , we have found that in ESCs specifically , NUP98 binding sequences were enriched for motifs recognized by GC-Box factors SP1 , C2H2 zinc finger transcription factors and SMAD ( Figure S5C ) . These findings raise the exciting possibility that NUP98 is linked to the core transcription circuitry that is crucial for the maintenance of pluripotency in ESCs [40] , [41] . To further understand the dynamic DNA-binding behavior of NUP98 , we investigated the functional categories of genes bound by NUP98 in ESCs and NeuPCs by gene ontology analysis . In ESCs , the top functional category enriched in NUP98 targets was found to be cytoskeleton organization ( Figure 2B ) . This is consistent with recent reports showing that in Drosophila embryonic culture cells NUP98 binding targets were also enriched for cytoskeleton genes [17] . As discussed later ( Figure S7 ) , NUP98 targets in ESCs could be divided into two groups , one associated with active histone marks and one associated with silent histone marks . The active group of NUP98 targets in human ESCs was enriched for genes in the functional categories of cell cycle regulation , cell communication and metabolism . Such genes were also enriched in Drosophila NUP98 targets in embryonic cells [17] ( and data not shown ) . Interestingly , NUP98 targets were specifically enriched for neurogenesis genes in NeuPCs , including genes in functional categories of nervous system development , neuron projection development , and neuron development ( Figure 2B ) . Examples of NUP98-interacting neurogenesis genes include NRG1 , ERBB4 , SOX5 , and ROBO [42]–[44] . Furthermore , analysis of disease terms enriched in NUP98 targets in NeuPCs revealed that NUP98 is linked to genes involved in multiple diseases of the nervous system ( Figure S6 ) . Such diseases include neurodegenerative disorders such as Alzheimer disease and tumors such as glioma and neoplasms of the nerve tissue . The latter finding might be relevant for the previously reported role of NUP98 in tumorigenesis [39] . These results suggest that NUP98 is recruited to neural developmental genes in a developmentally controlled manner . The specific association between NUP98 and neurogenesis genes in NeuPCs raised the possibility of a positive correlation between NUP98 binding and the activation of these genes during neural differentiation . To test this possibility , we compared the expression levels of genes bound by NUP98 to those of the same number of randomly selected genes in ESCs and NeuPCs using published RNA-Seq datasets [42] , [43] ( Figure 3A , 3B ) . We found that genes bound by NUP98 had higher expression levels in NeuPCs compared to randomly selected gene sets , suggesting that NUP98-binding was associated with elevated gene expression levels . As an independent test , we correlated the genomic localization of NUP98-binding regions to that of expressed mRNA in NeuPCs ( Figure 3C ) . We were able to detect a positive correlation between the location of NUP98 binding on the genome and the location of mRNA production , indicating the positive correlation between NUP98 binding and mRNA expression . Having established a link between NUP98 binding and active gene expression in NeuPCs , we asked if NUP98 binding to its target genes in NeuPCs would coincide with their transcriptional induction during neural differentiation . We found that NUP98-bound loci in NeuPCs had higher expression levels than either ESCs or IMR90 cells ( Figure 3D ) . By contrast , for randomly selected genes , there was no statistically significant difference in expression levels in any of the analyzed cell types . Together , these findings support the notion that NUP98 gains association with developmental genes as they are undergoing transcriptional activation during development . Considering all genes in the human genome , from published RNA-Seq datasets , there are a total of 8388 genes activated during differentiation of ESCs into NeuPCs . They were defined as genes whose expression levels were not detectable in ESCs but detectable in NeuPCs or were upregulated by more than two-folds in NeuPCs compared to ESCs [42] , [43] . 2 . 7% of these genes gained NUP98 binding in NeuPCs compared to ESCs , suggesting that NUP98 is associated with specific subset of developmentally regulated genes . In addition , we found 138 genes that lost NUP98 binding and also became inactivated in terms of expression levels upon differentiation from ESCs to NeuPCs . The expression levels of these genes were detectable in ESCs but undetectable in NeuPCs or were downregulated more than two-fold in NeuPCs compared to ESCs from published RNA-Seq datasets [42] , [43] . This suggests that NUP98 might also be linked to active gene expresison in pluripotent cells . In contrast to the direct correlation between NUP98 binding and gene activation in NeuPCs , the scenario in ESCs appears more complicated . To gain additional insight into the type of chromatin environment that NUP98 interacts with , we compared NUP98 binding to the levels of different histone modifications by comparing our ChIP-Seq datasets to published ChIP-Seq datasets of histone modifications in ESCs [42] . Specifically , we examined H3K79me2 and H3K36me3 that are linked to active transcription , as well as H3K27me3 and H3K9me3 that are linked to repressed chromatin domains [44] . We compared histone modification levels for NUP98-binding regions and randomly selected regions as negative controls . We found that , in ESCs , NUP98 binding showed positive correlation with both active and silent histone marks . In contrast , NUP98 binding in IMR90 cells , which does not target promoter regions , was exclusively linked to high H3K9me3 levels ( Figure 4 ) . This observation is consistent with the idea that NUP98 is preferentially , if not exclusively , involved in developmental gene regulation in pluri-/multi-potent cells whereas in differentiated cells either associates with repressive chromatin ( e . g . IMR90 cells ) or lacks chromatin association altogether ( e . g . neurons ) . The finding that NUP98 associates with both active and silent chromatin domains in ESCs could be due to two reasons: 1 ) NUP98 is directed to bivalent domains that exhibit both active and silent histone marks or 2 ) there are two subsets of NUP98 targets , one active and one silent . To distinguish between these two possibilities , we determined the extent to which pairs of histone marks were found at NUP98 binding regions by Pearson's Correlation analysis ( Figure S7A ) . Specifically , we examined the extent of correlation between 4 pairs of histone marks , H3K36me3 ( active histone mark ) - H3K27me3 ( silent histone mark ) , H3K36me3 ( active ) -H3K9me3 ( silent ) , H3K79me2 ( active ) - H3K27me3 ( silent ) , and H3K79me2 ( active ) - H3K9me3 ( silent ) . The result showed that the correlation between active and silent histone marks for NUP98 targets was low , suggesting NUP98-binding regions can be divided into at least two distinct subgroups , the group with active histone marks and the group with silent marks . In order to examine the types of genes included in each group , for each histone mark we ranked the genes bound by NUP98 by the levels of the histone mark found at that loci , selected the top 40% of the genes and performed gene ontology analysis on those genes ( Figure S7B–S7D ) . We found that NUP98 targets with high levels of active histone marks ( H3K79me2 or H3K36me3 ) were uniquely enriched for genes involved in macromolecule and nucleic acid metabolism and various aspects of the cell cycle such as nuclear division and mitosis . On the other hand , NUP98 targets , which were characterized by high levels of repressive histone mark H3K27me3 , were uniquely enriched for genes involved in transmembrane transport . Furthermore , we confirmed that NUP98 targets with high levels of active histone marks were actively transcribed , whereas the ones with high levels of silent histone marks were repressed ( Figure S7E–S7H ) . These observations are reminiscent of the findings in Drosophila embryonic culture cells in which NUP98 associates with both active and repressed genes as well as cell cycle and nucleic acid metabolism genes ( [17]; ( data not shown ) . Combining the observations in Drosophila and human cells , it is possible that NUP98 exhibits an evolutionally conserved role in associating with and potentially regulating cell cycle and nucleic acid metabolism genes . Together these data suggest that in undifferentiated ESCs , NUP98 associates with one subgroup of active genes including cell cycle and nucleic acid metabolism genes as well as with one group of silent chromatin regions . Since NUP98 associated with neural development genes during neural differentiation , we asked if this nuclear pore complex component plays a role in their expression . We randomly selected 24 genes from the 54 genes in the ‘nervous system development’ gene ontology category that showed specific enrichment in NeuPCs ( Figure 2B ) together with GAPDH as well as additional genes that did not bind NUP98 as negative controls , and examined how their expression levels were affected by NUP98 overexpression in neural progenitor cells using qRT-PCR ( Figure 5A , 5B , Figure S8A ) . To do this , NeuPCs were transfected with NUP98 and the overexpressed NUP98 localized to both nuclear pores and nucleoplasm ( Figure S9 ) . Strikingly , we found that 12 NUP98-associated neural developmental genes showed statistically significant increase in expression levels upon NUP98 overexpression , indicating that NUP98 regulates the transcription of these genes . Since not all genes responded to NUP98 overexpression , we suspect that NUP98 might not be rate-limiting in all its target genes . We then overexpressed a fragment of NUP98 ( amino acid 1–504 ) in NeuPCs , which lacks a C-terminal domain of NUP98 that is no longer capable of binding to the nuclear pore complex ( Figure S9 ) . We were interested in this region of NUP98 because this is the same fragment as appeared in multiple NUP98-involved leukemic fusions and this fragment has been found to interfere with the differentiation of haematopoietic progenitor cells [39] . Given reported evidences for a role of NUP98 in gene regulation [8] , [17] and our observation of the association between NUP98 and developmental genes at the progenitor cell stage , we hypothesized that this NUP98 fragment might interfere with the expression of NUP98 targets required for neural differentiation . We found that overexpression of this fragment of NUP98 had a dominant negative effect on the expression of NUP98-binding neural developmental genes , and 20 of the 24 genes exhibited statistically significant decrease in expression levels ( Figure 5C , 5D ) . No significant effects on gene expression have been observed for GAPDH as well as additional genes that did not bind NUP98 ( Figure 5C , 5D , Figure S8B ) . This suggests that the C-terminal domain of NUP98 is required for the expression of NUP98 target genes because the fragment lacking this domain could not stimulate expression of target genes as the full length NUP98 protein did . As an additional negative control , we overexpressed NUP35 using the same vector and found no effects on the expression of NUP98-binding genes ( Figure S10 ) . We did not examine the effect of NUP98 knockdown on gene expression because NUP98 is encoded on the same mRNA with a core component of the nuclear pore , NUP96 , which is essential to nuclear pore biogenesis [32] . Knockdown of NUP98 causes simultaneous knockdown of NUP96 and a failure in nuclear pore formation and cell death ( data not shown ) . Therefore , it was not possible to specifically analyze the gene regulatory function of NUP98 from such knockdown experiments . Collectively , these results indicate that NUP98 is functionally relevant for the expression of neural developmental genes it associates with in NeuPCs . To obtain further insights into the role of NUP98 during differentiation we monitored the mRNA levels of 24 NUP98 target genes that were in the neural development gene ontology category through differentiation from ESCs to NeuPCs , and subsequently to postmitotic neurons in which Nup98 does not seem to bind the genome ( Figure 6 ) . We found that all 24 genes were upregulated when ESCs were differentiated to NeuPCs , consistent with the genome-wide correlation analysis and supporting a role of NUP98 in the induction of transcription ( Figure 3D ) . When NeuPCs were further differentiated to neurons , the majority of genes ( 20 genes ) showed continued transcriptional induction . Among those 20 genes , we focused on 6 genes that exhibited the most dramatic increase in expression in neurons . We observed that these genes could be largely divided into two groups ( Figure 6 ) . Group I genes ( GRIK1 , NRG1 , and MAP2; colored in red ) showed modest transcriptional induction in NeuPCs compared to ESCs . However , this cohort of genes underwent a robust increase in expression during the transition from NeuPCs to neurons . Group II genes ( GPM6B , SOX5 , and ERBB4; colored in green ) underwent a dramatic activation in the initial commitment of ESCs to NeuPCs and only slight upregulation during subsequent neuronal differentiation . This suggests that NUP98 associates with both genes starting to be developmentally induced ( Group I genes ) and genes that are at a later stage of induction ( Group II genes ) in NeuPCs . As a mobile nuclear pore complex component , NUP98 can act both at the nuclear pore complexes and inside the nucleus at sites that are not attached to the nuclear envelope ( NE ) [8] , [17] . Therefore , we wondered if either of the two classes of genes is specifically associated with nuclear pore complexes at the NE . We examined the localization of the group I and group II NUP98 targets by immunofluorescence-fluorescence in situ hybridization ( IF-FISH ) experiments . We used lamin ( LMNB ) staining as a marker for the NE , and only counted FISH signals whose center overlaid with the NE ( corresponding to <0 . 5 µm distance from the NE ) as ‘periphery’ localization ( Figure 7A ) . We found that the two groups of genes also showed distinct intranuclear localization at the progenitor cell stage . In NeuPCs , group I genes that will become transcriptionally active were localized to the periphery , whereas group II genes that were already expressed at high levels were in the interior of the nucleus ( Figure 7B–7D , Figure S11 ) . Upon differentiation into neurons , group I genes moved into the nuclear interior whereas group II genes maintained their interior localization ( Figure 7B–7D , Figure S11 ) . In order to further confirm the association of group I genes with the nuclear pore complexes in NeuPCs , we tested the interaction of these genes with an additional nuclear pore component NUP133 by ChIP-qPCR . NUP133 is a scaffold component of the nuclear pore complexes that associates stably with the nuclear pores at the nuclear periphery [18] . It has not been observed at nuclear pore-free lamina sites or intranuclear sites at endogenous levels . We found that NUP133 bound the group I genes at the nuclear periphery , but not group II genes in the nucleoplasm ( Figure S12A ) . As additional controls , for each group I gene , we selected two neighboring genes for a total of 6 genes ( USP16 , CLDN17 , DCTN16 , WRN , KCF7 , and PTH2R ) and observed no interaction between these genes and NUP98 or NUP133 by ChIP-qPCR ( Figure S12A ) , further supporting the idea that the group I genes interacted with nuclear pores at the nuclear periphery in NeuPCs . We also examined the intranuclear localization of the 6 neighboring genes ( USP16 , CLDN17 , DCTN16 , WRN , KCF7 , and PTH2R ) to study how far the peripheral localization extended from the group I genes . We found that the 6 genes exhibited large range of percentages of peripheral localization ( from 10% to 60% ) ( Figure S12B ) . This suggests that NUP98 binding to a given gene at the nuclear periphery could not predict peripheral localization of flanking genes . Given the association between NUP98 and neural developmental genes , we decided to test if overexpression of full length NUP98 and its dominant negative fragment in neural progenitor cells affected efficiency of neuronal differentiation . We examined the efficiency of neuronal differentiation by measuring the expression levels of markers for differentiated neurons ( RBFox3 , TUBB3 , and Syn1 ) at the end of 1 month's neuronal differentiation from NeuPCs . We observed that overexpression of full length NUP98 increased expression of those neuronal markers , whereas overexpression of the dominant negative fragment decreased their expression levels ( Figure S13 ) . This is consistent with the findings that overexpression of full length NUP98 increased expression of neural developmental genes , whereas overexpression of the fragment reduced expression of such genes ( Figure 5 ) . Collectively these results suggest that NUP98 regulates the efficiency of neuronal differentiation from neural progenitor cells . Based on these observations , we conclude that at the neural progenitor stage , there are at least two modes of gene regulation by NUP98 , 1 ) the ‘gene to pore’ model where genes relocate to the nuclear pore at the initial stage of transcriptional induction associated with neurogenesis; and 2 ) the ‘Nup to gene’ model where NUP98 acts away from the nuclear pore to interact with genes that are highly activated ( Figure 7E ) .
In addition to their well established role in mediating transport across the NE , nuclear pore proteins have been implicated in directly regulating gene expression in organisms as diverse as yeast and Drosophila [4] , [8]–[13] , [16] , [17] , [20] . However , the functions of Nups during development , especially their roles in gene regulation and in higher organisms such as humans , remain largely unexplored . Here we provide evidence that in human cells , the nuclear pore protein NUP98 binds the nuclear genome in a manner that is tightly linked to differentiation status and developmental gene expression . In embryonic stem cells , NUP98 bound genes include an active subgroup such as genes involved in cell cycle and nucleic acid metabolism regulation and a silent subgroup . In neural progenitor cells , NUP98 shows distinct association with genes activated during neural development , and NUP98 is functionally important for the expression of these genes . In the lung fibroblast IMR90 cells NUP98 mainly interacts with silent chromatin domains . This suggests that besides controlling nucleo-cytoplasmic exchange , NUPs can dynamically interact with the human genome during differentiation , providing an additional layer of genome regulation during development . From a cell biological point of view , there are at least two modes of developmental gene regulation by NUP98 , the ‘on-pore’ regulation and the ‘off-pore’ regulation . Our findings suggest that at least one of the distinctions of the two modes of regulation might be related to the temporal gene expression dynamics of NUP98 targets . Specifically , during the differentiation of human embryonic stem cells along the neural lineage , nuclear pore-tethered NUP98 acts as a short-term anchoring point for certain developmental genes at the beginning stages of transcription induction . In progenitor cells , anchorage at the nuclear pores could be especially important for genes at the initial stages of developmental induction because for these genes the activation status may not be stable yet and therefore require the microenvironment of the nuclear pores to maintain chromatin decondensation and gene transcriptional status , especially through repeated cell cycles such as in neural progenitor cells ( discussed below ) . On the other hand , for genes that are at later stages of developmental induction , the chromatin is entirely open and thus does not require the nuclear pore-tethering mechanism to maintain transcription . Under such circumstances , the nuclear interior might be a more optimal microenvironment for those genes that supports robust transcription compared to the nuclear pores which are in proximity to the nuclear lamina which can mediate transcriptional repression [45] , [46] . The rationale for the involvement of the nuclear pores in developmental gene regulation , especially at the progenitor stage , probably relates to the necessity of re-establishing chromatin organization after nuclear envelope breakdown and reformation in mitosis . During M phase of the cell cycle , chromatin is condensed , transcription activities are largely diminished and most transcription factors are absent from mitotic chromosomes , which composes a window that allows for cell fate reprogramming [47]–[49] . Therefore , in progenitor cells , upon mitosis exit , chromatin has to be decondensed in a manner that faithfully restores the ‘open’ or ‘closed’ states for different chromatin domains to ensure that corresponding developmental genes can be activated or repressed correctly . Nups are prime candidates to regulate transcription re-initiation of developmental genes based on ‘transcriptional memory’ from previous cell cycles because during mitosis exit , Nups are among the first proteins to establish contacts with chromatin and it has been found that proper chromatin decondensation requires the functioning of Nups [50]–[53] . Furthermore , association with Nups in yeast has been shown to convey a ‘gene memory’ function so that genes can be rapidly re-induced for repeated transcription stimulation cycles [12] , [54] . Along these lines of evidence , NUP98 in Drosophila is involved in the re-initiation of transcription after heat shock [8] and our study has shown that in the cycling human neural progenitor cells NUP98 associates with and regulates expression of neural development genes . Together these observations point to the role of Nups in the rapid and faithful re-initiation of expression of developmental genes after each mitosis cycle . In the search for DNA sequences that might direct NUP98-chromatin interaction , we identified a conserved DNA binding motif , the GA boxes . This motif is overrepresented in NUP98-binding sequences not only in human cells from our study , but also in Drosophila cells from published ChIP-chip and Dam-ID datasets . In Drosophila , GA-boxes are recognized by the GAGA factor , which is encoded by the Trithorax-like gene and is required for the proper development of the organism [33] . Interestingly , GAGA factor has been related to the yeast factor Rap1 because of their similarities in binding to both repetitive sequences and transcriptionally active genes as well as exhibiting boundary activity [33] , and the Rap1 binding site has been identified as the nuclear-pore recognizing DNA motif in yeast [10] . Together these lines of evidence suggest that the DNA recognition activity of Nups or Nup-interacting partners is evolutionarily conserved . Finally , the involvement of NUP98 in developmental regulation sheds light on its involvement in multiple types of leukemia where it is fused to various transcription regulators [39] . Such oncogenic NUP98-fusion proteins have been shown to promote the self-renewal of hematopoietic progenitor cells and inhibit their differentiation [55] . We found that NUP98 is connected to the regulation of genes implicated in neoplasm formation especially at the progenitor stage . In addition , overexpression of the NUP98 fragment as appeared in the fusion proteins disrupted the expression of endogenous NUP98 targets which , during normal differentiation processes , were activated . Therefore , the misregulation of developmental genes in hematopoietic cells due to genomic fusion of NUP98 with transcription regulators may be a potential mechanism driving the transformation events in NUP98-fusion protein associated leukemias .
Work involving embryonic stem cells was carried out in accordance with the policies set by the Salk Institute . Human embryonic stem cell line HUES6 were grown under feeder-free conditions in mTeSR1 medium . HUES6-derived neural progenitor cells were grown in DMEM/F12 supplemented with N2/B27 . Early passage IMR90 cells were grown in DMEM , 15% FBS and MEM nonessential amino acids . Culture and differentiation conditions were detailed in Text S1 . Primary antibodies used include rabbit anti-human NUP98 polyclonal antibody ( Cell Signaling 2292; ‘NUP98Ab1’ specified in the experiment ) , rabbit anti-human NUP98 monoclonal antibody ( Cell Signaling 2598 ) , mAb414 ( Covance MMS-120R ) , normal rabbit IgG ( Cell Signaling 2729 ) , anti-human Nestin antibody ( Chemicon ) , anti-Sox2 antibody ( Chemicon ) , and rabbit-anti-LMNB antibody ( Aviva ARP46357-P050 ) . Cells were fixed in 1% formaldehyde ( Polysciences ) for 10 min . Fixation was stopped by adding glycine to a final concentration of 125 mM . Fixed cells were lysed and sonicated . DNA was immunoprecipitated , eluted , de-crosslinked , treated with RNase and protease , and purified . Procedures were detailed in Text S1 . Library was constructed using Illumina ChIP-Seq DNA sample prep kit and sequencing was done on Illumina GAII . Mapping and peak calling of ChIP-Seq data , annotation of NUP98-binding regions , mapping and expression level analysis of RNA-Seq data , transcription factor motif analysis , gene ontology analysis , positional correlation of ChIP-Seq and RNA-Seq were conducted using the Genomatix software . Peak calling was based on Audic-Claverie algorithm for NGSAnalyzer . Chromosomal views of ChIP-Seq data were generated using Affymetrix Integrated Genome Browser and correlation of NUP98 binding with gene expression levels and histone modification levels was performed using the R package for statistical computing . FISH probes were DIG-labelled using the DIG-Nick translation mix for in situ probes ( Roche ) . Cells were fixed , immuno-stained , permeablized , denatured in 50% formamide/2xSSC for 30 min at 80°C , hybridized to DIG-labeled FISH probes overnight at 42°C , stained with anti-DIG antibody ( Roche ) and Hoechst and mounted . Procedures were detailed in Text S1 . Three-dimensional image stacks were recorded with Zeiss LSM710 scanning scope using a 63× objective , 512×512 resolution , 2× averaging and optimal interval ( 0 . 31 µm ) between stacks in Z-direction and three-dimensional images were reconstructed from the Z-stack images . For NUP overexpression , plasmids were electroporated into NeuPCs using rat neural stem cell Nucleofector solution ( Lonza Amaxa , VPG-1005 ) or ( in differentiation assays ) packaged into lentiviruses that were used subsequently to infect NeuPCs . NUP98 was knocked down by siRNA ( oligo sequence: GAG AGA GAT TTA GTT TCC TAA GCA A ) in IMR90 cells using Dharmafect 1 siRNA transfection reagent according to the manufacturer's instructions . | Development of multicellular organisms such as humans requires appropriate activation of gene expression programs according to stages of differentiation . Many proteins that directly regulate this process have been identified , including histone-modifying enzymes and transcription factors . It is not clear whether nuclear pore proteins , proteins that form the only channels in the nuclear envelope that mediate nuclear transport , regulate developmental gene regulation in higher organisms such as humans . Here we show that one nuclear pore protein has a role in gene regulation during human cell differentiation , providing insight into the development-related and transport-independent function of nuclear pore proteins . We have found that the nuclear pore protein interacts with the human genome in a dynamic manner that is tightly linked to the developmental stage . In addition , manipulating the functional levels of the nuclear pore protein can disrupt expression of the developmental genes it associates with . Our results suggest that the nuclear pore protein functionally interacts with the genome during cell differentiation , uncovering an additional layer of developmental gene regulation in humans . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2013 | Dynamic Association of NUP98 with the Human Genome |
Social distancing practices are changes in behavior that prevent disease transmission by reducing contact rates between susceptible individuals and infected individuals who may transmit the disease . Social distancing practices can reduce the severity of an epidemic , but the benefits of social distancing depend on the extent to which it is used by individuals . Individuals are sometimes reluctant to pay the costs inherent in social distancing , and this can limit its effectiveness as a control measure . This paper formulates a differential-game to identify how individuals would best use social distancing and related self-protective behaviors during an epidemic . The epidemic is described by a simple , well-mixed ordinary differential equation model . We use the differential game to study potential value of social distancing as a mitigation measure by calculating the equilibrium behaviors under a variety of cost-functions . Numerical methods are used to calculate the total costs of an epidemic under equilibrium behaviors as a function of the time to mass vaccination , following epidemic identification . The key parameters in the analysis are the basic reproduction number and the baseline efficiency of social distancing . The results show that social distancing is most beneficial to individuals for basic reproduction numbers around 2 . In the absence of vaccination or other intervention measures , optimal social distancing never recovers more than 30% of the cost of infection . We also show how the window of opportunity for vaccine development lengthens as the efficiency of social distancing and detection improve .
Epidemics of infectious diseases are a continuing threat to the health of human communities , and one brought to prominence in the public mind by the 2009 pandemic of H1N1 influenza [1] . One of the key questions of public health epidemiology is how individual and community actions can help mitigate and manage the costs of an epidemic . The basic problem I wish to address here is how rational social-distancing practices used by individuals during an epidemic will vary depending on the efficiency of the responses , and how these responses change the epidemic as a whole . Social distancing is an aspect of human behavior particularly important to epidemiology because of its universality; everybody can reduce their contact rates with other people by changing their behaviors , and reduced human contact reduces the transmission of many diseases . Theoretical work on social distancing has been stimulated by studies of agent-based influenza simulations indicating that small changes in behavior can have large effects on transmission patterns during an epidemic [2] . Further research on agent-based models has argued that social distancing can arrest epidemics if started quickly and maintained for a relatively long period [3] . Compartmental epidemic models have also been used to study social distancing by including states that represent individuals employing specific behaviors . For instance , Hyman and Li [4] formulate and begin the analysis of flu disease transmission in SIR models where some individuals decrease their activity levels following infection . Reluga and Medlock [5] uses this approach to show that while social distancing can resemble immunization , it can generate hysteresis phenomena much more readily than immunization . Rather than treating behaviors as states , some models treat behaviors as parameters determined by simple functions of the available information . Reluga et al . [6] studies dynamics where contact rates can depend on the perceived disease incidence . Buonomo et al . [7] investigates the impact of information dynamics on the stability of stationary solutions in epidemic models . Chen [8] considers a similar system but allows individuals to learn from a random sample of neighbors . Funk et al . [9] considers the information dynamics associated with social distancing in a network setting by prescribing a reduction in contacts based on proximity to infection . Related work by Epstein et al . [10] explicitly considers the spatial and information dynamics associated in response to an ongoing epidemic . Building on the ground-breaking work of Fine and Clarkson [11] , there has been substantial recent interest in the application of game theory to epidemiology [12]–[17] . The games studied so far have primarily considered steady-state problems , and have not allowed for dynamic strategies . One notable exception to this is the work of Francis [18] , which determines the time-dependent game-theoretical solution of a vaccination problem over the course of an epidemic . In another , van Boven et al . [19] studies the optimal use of anti-viral treatment by individuals when they take into account the direct and indirect costs of treatment . To study the best usage of social distancing , we apply differential-game theory at a population-scale . Differential games are games where strategies have a continuous time-dependence; at each point in time , a player can choose a different action . For instance , a pursuit-game between a target and a pursuer is a two-player differential game where each player's strategies consist of choosing how to move at each successive time until the target is caught by the pursuer or escapes . Geometrically , one might think of differential games as games where strategies are represented by curves instead of points . Two-player differential-game theory was systematically developed by Isaacs [20] as an extension of optimal control theory [21]–[23] . Here , we employ an extension of differential game theory to population games of the form described by Reluga and Galvani [24] . The analysis in this paper will be limited to the simplest case of the Kermack–McKendrick SIR model with strong mixing [25] . In the Model section , we formulate an epidemiological-economics model for an epidemic , accounting for the individual and community costs of both social distancing practices and infection . We then use differential game theory and numerical methods to identify the equilibrium strategies over the course of an epidemic . Numerical methods are used to investigate the finite-time problem where vaccines become available after a fixed interval from the start of the epidemic and the infinite-horizon problem without vaccination . Fundamental results on the value and timing of social distancing are obtained .
We now formulate a differential game for individuals choosing their best social distancing practices relative to the aggregate behavior of the population as a whole . The following game-theoretic analysis combines the ideas of Isaacs [20] and Reluga and Galvani [24] . The premise of the game is that at each point in the epidemic , people can choose to pay a cost associated with social distancing in exchange for a reduction in their risk of infection . The costs of an epidemic to the individual depend on the course of the epidemic and the individual's strategy of social distancing . The probabilities that an individual is in the susceptible , infected , or removed state at time evolve according to the Markov process ( 7 ) where is the individual's daily investment as a function of the epidemic's state-variables and the transition-rate matrix ( 8 ) Note that both and change over time . Along the lines discussed above , and represent different quantities in our analysis; represents one individual's investment strategy and the population strategy represents an aggregated average of all individual investments . We also note that there are several different ways and can be parameterized . They may be parameterized in terms of time , as and , or in implicit feedback form and , or in explicit feedback form and . The form used will be clear from the context . Since the events in the individual's life are stochastic , we can not predict the exact time spent in any one state or the precise payoff received at the end of the game . Instead , we calculate expected present values of each state at each time , conditional on the investment in social distancing . The expected present value is average value one expects after accounting for the probabilities of all future events , and discounting future costs relative to immediate costs . The expected present values of each state evolve according to the adjoint equations ( 9 ) where . The components , , and represent the expected present values of being in the susceptible , infected , or removed state at time when using strategy in a population using strategy . The expected present values depend on the population strategy through the infection prevalence . The adjoint equations governing the values of each state are derived from Markov decision process theory . They are ( 10a ) ( 10b ) ( 10c ) with the constraints that for all time . Solution of ( 10 ) b and ( 10 ) c gives ( 11 ) If it is impossible to make a vaccine , the equations must be solved over an infinite horizon . Over an infinite horizon , , assuming becomes constant . In the case of no discounting ( ) , we still have provided for sufficiently large . In the case where a perfect vaccine is universally available at terminal time , the value of the susceptible and removed states differs by the cost of vaccine for . To avoid complications with the choice of whether-or-not to vaccinate , we take so . This is reasonable in scenarios where the cost of the vaccine is covered by the government . The dynamics are independent of , so we need not consider removed individuals further . Taking and , we need only study the reduced system ( 12a ) ( 12b ) ( 12c ) with boundary conditions ( 12d ) The other conditions must be calculated from the solution of the boundary-value problem and provide useful information . will be the expected total cost of the epidemic to the individual . The final size of the epidemic is given by . Solving a game refers to the problem of finding the best strategy to play , given that all the other players are also trying to find a best strategy for themselves . In some games , there is a single strategy that minimizes a player's costs no matter what their opponents do , so that strategy can very reasonably be referred to as a solution . In many games , no such strategy exists . Rather , the best strategy depends on the actions of the other players . Any strategy played by one player is potentially vulnerable to a lack of knowledge of the strategies of the other players . In such games , it is most useful to look for strategies that are equilibria , in the sense that every player's strategy is better than the alternatives , given knowledge of their opponent's strategies . A Nash equilibrium solution to a population game like that described by System ( 12 ) is a strategy that is a best response , even when everybody else is using the same strategy . i . e . given , is a Nash equilibrium if for every alternative strategy , . A Nash equilibrium strategy is a subgame perfect equilibrium if it is also a Nash equilibrium at every state the system may pass through . I will not address the problem of ruling out finite-time blowup of the Hamilton–Jacobi equation and establishing existence and uniqueness of subgame perfect equilibria . But numerical and analytical analyses strongly support the conjecture that the stategies calculated here are the unique global subgame perfect equilibria to the social distancing game . The equilibria of System ( 12 ) can be calculated using the general methods of Isaacs [20] . The core idea is to implement a greedy-algorithm; at every step in the game , find the investment that maximizes the rate of increase in the individual's expect value . We represent strategies as functions in implicit feedback form . is the amount an individual invests per transmission generation when the system is at state . If is a subgame perfect equilibrium , then it satisfies the maximum principle ( 13 ) when everywhere . So long as behaves well , in the sense that it is differentiable , decreasing , and strictly convex , then is uniquely defined by the relations ( 14 ) Figure 1 shows the interface in phase space separating the region where the equilibrium strategy will include no investment in social distancing ( ) from the region where the equilibrium strategy requires investment in social distancing ( ) . Two cases are immediately interesting . The first is the infinite-horizon problem – what is the equilibrium behavior when there is never a vaccine and the epidemic continues on until its natural end ? The second is the finite-horizon problem – if a vaccine is introduced at time generations after the start of the epidemic , what is the optimal behavior while waiting for the vaccine ? In both of these cases , it is assumed that all players know if and when the vaccine will be available . The infinite-horizon and finite-horizon problems are distinguished by their boundary conditions . In the finite-horizon case , we assume all susceptible individuals are vaccinated at final time , so , , , while and are unknown . In the limit of the infinite-horizon case ( ) , we solve the two-point boundary value problem with terminal conditions , , and initial conditions , while and are unknown . But these conditions are insufficient to specify the infinite-horizon problem . The plane is a set of stationary solutions to Eq . ( 12 ) , so we need a second order term to uniquely specify the terminal condition when we are perturbed slightly away from this plane . Using Eq . ( 12 ) , we can show solutions solve the second-order terminal boundary condition ( 15 ) for as . Most of the equilibria we calculate are obtained numerically . Some exceptions are the special cases where , . Under these conditions , solutions can be obtained in closed-form . First , . While , and ( 16 ) When matched to the terminal boundary condition , we find that if we write in feedback form as a function of rather than , ( 17 ) is a solution so long as for all . Inspecting the inequality condition , we find that this holds as long as .
A problem with solving Eq . ( 12 ) under Eq . ( 14 ) is that it requires to be known from past time and to be known from future time . This is a common feature of boundary-value problems , and is resolved by considering all terminal conditions . Using standard numerical techniques , identifying an equilibrium in the described boundary-value problem reduces to scalar root finding for to match the given . The special form of the population game allows the solution manifold to be calculated directly by integrating backwards in time , rather than requiring iterative approaches like those used for optimal-control problems [23] . Code for these calculations is available from the author on request . Before presenting the results , it is helpful to develop some intuition for the importance of the maximum efficiency of investments in social distancing . Given for an arbitrary relative risk function , then in the best-case scenarios , where diminishments on returns are weakest , one would have to invest atleast of the cost of infection per disease generation to totally isolate themselves . The units here are derived from dimensional analysis . This could be invested for no more than generations , before one's expenses would exceed the cost of becoming infected . When returns are diminishing , fewer than generations of total isolation are practical . Thus , the dimensionless efficiency can be thought of as an upper bound on the number of transmission generations individuals can afford to isolate themselves before the costs of social distancing outweigh the costs of infection . For the infinite-horizon problem , an example equilibrium strategy and the corresponding dynamics in the absence of social distancing are shown in Figure 2 . We can show that if social distancing is highly inefficient ( the maximum efficiency ) , then social distancing is a waste of effort , no matter how large . If social distancing is efficient , then there is a threshold value of below which social distancing is still impractical because the expected costs per day to individuals is too small compared to the cost of social distancing , but above which some degree of social distancing is always part of the equilibrium strategic response to the epidemic ( Figure 3 ) . The exact window over which social distancing is used depends on the basic reproduction number , the initial and terminal conditions , and the efficiency of distancing measures . The feedback form of equilibrium strategies , transformed from coordinates to the coordinates of the phase-space is represented with contour plots in Figure 1 . Among equilibrium strategies , social distancing is never used until part-way into the epidemic , and ceases before the epidemic fully dies out . The consequences of social distancing are shown in Figure 4 . The per-capita cost of an epidemic is larger for larger basic reproduction numbers . The more efficient social distancing , the more of the epidemic cost can be saved per person . However , the net savings from social distancing reaches a maximum around , and never saves more than % of the cost of the epidemic per person . For larger 's , social distancing is less beneficial . We can also calculate solutions of the finite-time horizon problem where a vaccine becomes universally available at a fixed time after the detection of disease ( Figure 5 ) . If mass vaccination occurs soon enough , active social distancing occurs right up to the date of vaccination . Using numerical calculations of equilibria over finite-time horizons , we find that there is a limited window of opportunity during which mass vaccine can significantly reduce the cost of the epidemic , and that social distancing lengthens this window ( Figure 6 ) . The calculations show that increases in either the amount of time before vaccine availability or the basic reproduction number increase the costs of the epidemic . Smaller initial numbers of infections allow longer windows of opportunity . This is as expected because the larger the initial portion of the population infected , the shorter the time it takes the epidemic to run its full course .
Here , I have described the calculations necessary to identify the equilibrium solution of the differential game for social distancing behaviors during an epidemic . The benefits associated with the equilibrium solution can be interpreted as the best outcome of a simple social-distancing policy . We find that the benefits of social distancing are constrained by fundamental properties of epidemic dynamics and the efficiency with which distancing can be accomplished . The efficiency results are most easily summarized in terms of the maximum efficiency , which is the percent reduction in contact rate per percent of infection cost invested per disease generation . As a rule-of-thumb , is an upper bound on the number of transmission generations individuals can isolate before the costs of social distancing outweigh the costs of infection . Social distancing is not practical if this efficiency is small compared to the number of generations in the fastest epidemics ( ) . While social distancing can yield large reductions in transmission rate over short periods of time , optimal social-distancing strategies yield only moderate reductions in the cost of the epidemic . Our calculations have determined the equilibrium strategies from the perspective of individuals . Alternatively , we could ask what the optimal social distancing practices are from the perspective of minimizing the total cost of the epidemic to the community . Determination of the optimal community strategy leads to a nonlinear optimal control problem that can be studied using standard procedures [23] . Yet , practical bounds on the performance of the optimal community strategy can be obtained without further calculation . The optimal community strategy will cost less than the game-theoretic solution per capita , but must cost more than , as that is the minimum number of people who must become sick to reduce the effective reproduction number below the epidemic threshold . Preliminary calculations indicate that optimal community strategies and game equilibrium strategies converge as grows , and significant differences are only observable for a narrow window of basic reproduction numbers near . The results presented require a number of caveats . I have , for instance , only considered one particular form for the relative risk function . Most of the analysis has been undertaken in the absence of discounting ( ) , under the assumption that the epidemic will be fast compared to planning horizons . Discounting would diminish importance of long term risks compared to the instant costs of social distancing , and thus should diminish the benefits of social distancing . The benefits of social distancing will also be diminished by incorporation of positive terminal costs of vaccination ( ) . Realistically , mass vaccination cannot be accomplished all-at-once , as we assume . It's much more likely that vaccination will be rolled out continuously as it becomes available . This could be incorporated into our analysis , for instance , by including a time-dependent forcing . Other approaches include extending the model to incorporate vaccination results of Morton and Wickwire [28] , or to allow an open market for vaccine purchase [18] . The simple epidemic model is particularly weak in its prediction of the growths of epidemics because it assumes the population is randomly mixed at all times . We know , however , that the contact patterns among individuals are highly structured , with regular temporal , spatial , and social correlations . One consequence of heterogeneous contact structure is that epidemics proceed more slowly than the simple epidemic model naively predicts . Thus , the simple epidemic model is often considered as a worst-case-scenario , when compared with more complex network models [29] , [30] and agent-based models [31]–[33] . In the context of social distancing , it is not immediately clear how weaker mixing hypotheses will affect our results . Weakened mixing will prolong an epidemic , increasing the window over which social distancing is needed . But under weakened mixing , individuals may be able to use local information to refine their strategies in ways analogous to the ideas of Funk et al . [9] and Perisic and Bauch [34] . In general , the analysis of aggregate games with stochastic population dynamics require a significant technical leaps , and are the subjects of active research . One of the fundamental assumptions in our analysis is that there are no cost-neutral behavior changes that can reduce contact rates . In fact , life-experience provides good evidence that many conventional aspects of human behavior are conditional on cultural norms , and that different cultures may adopt alternative conventions . The introduction of a new infectious disease may alter the motivational pressures so that behavioral norms that were previously equivalent are no longer , and that one norm is now preferred to the others . In such cases , there are likely to be switching costs that retard the rapid adoption of the better behaviors that conflict with cultural norms . The rate of behavior change , then , would be limited by the rate of adoption of compensatory changes in cultural norms that reduce the cost of social distancing . Another deep issue is that behavior changes have externalities beyond influencing disease incidence , but we have not accounted for these externalities . People's daily activities contribute not just to their own well-being but also to the maintenance of our economy and infrastructure . Social distancing behaviors may have serious negative consequences for economic productivity , which might feed back into slowing the distribution of vaccines and increasing daily cost-of-living expenses . We can extend our analysis to include economic feedbacks by incorporating capital dynamics explicitly . Individuals may accumulate capital resources like food , water , fuel , and prophylactic medicine prior to an epidemic , but these resources will gradually be depleted and might be difficult to replace if social distancing interferes with the economy flow of goods and services . Further capital costs at the community and state scales may augment epidemic valuations . These factors appear to have been instrumental in the recent US debate of school-closure policies . One feature of a model with explicit capital dynamics is the possibility of large economic shocks . This and related topics will be explored in future work . These calculations raise two important mathematical conjectures which I have not attempted to address . The first is that the social distancing game possesses a unique subgame-perfect Nash equilibrium . There is reasonable numerical evidence of this in cases where the relative risk function is strictly convex , and stronger unpublished arguments of this in cases of piecewise linear . I believe this will also be the case for non-convex but monotone relative risks under some allowances of mixed-strategies . A second conjecture , not yet addressed formally , is that increases in the efficiency of social distancing always lead to greater use of social distancing , all other factors being equal . This seems like common sense , but the precise dependence of Figure 1 on the efficiency has yet to be determined mathematically . As with all game-theoretic models , human behavior is unlikely to completely agree with our equilibria for many reasons , including incomplete information about the epidemic and vaccine and strong prior beliefs that impede rational responses . On the other hand , our approach is applicable to a large set of related models . We can analyze many more realistic representations of pathogen life-cycles . For instance , arbitrary infection-period distributions and infection rates can be approximated using a linear chain of states or delay-equations [24] . Structured populations with metapopulation-style mixing patterns may also be analyzed . I hope to apply the methods to a wider variety of community-environment interactions in the future . | One of the easiest ways for people to lower their risk of infection during an epidemic is for them to reduce their rate of contact with infectious individuals . However , the value of such actions depends on how the epidemic progresses . Few analyses of behavior change to date have accounted for how changes in behavior change the epidemic wave . In this paper , I calculate the tradeoff between daily social distancing behavior and reductions in infection risk now and in the future . The subsequent analysis shows that , for the parameters and functional forms studied , social distancing is most useful for moderately transmissible diseases . Social distancing is particularly useful when it is inexpensive and can delay the epidemic until a vaccine becomes widely available . However , the benefits of social distancing are small for highly transmissible diseases when no vaccine is available . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases"
] | 2010 | Game Theory of Social Distancing in Response to an Epidemic |
DNA topology has fundamental control over the ability of transcription factors to access their target DNA sites at gene promoters . However , the influence of DNA topology on protein–DNA and protein–protein interactions is poorly understood . For example , relaxation of DNA supercoiling strongly induces the well-studied pathogenicity gene ssrA ( also called spiR ) in Salmonella enterica , but neither the mechanism nor the proteins involved are known . We have found that relaxation of DNA supercoiling induces expression of the Salmonella pathogenicity island ( SPI ) -2 regulator ssrA as well as the SPI-1 regulator hilC through a mechanism that requires the two-component regulator OmpR-EnvZ . Additionally , the ompR promoter is autoregulated in the same fashion . Conversely , the SPI-1 regulator hilD is induced by DNA relaxation but is repressed by OmpR . Relaxation of DNA supercoiling caused an increase in OmpR binding to DNA and a concomitant decrease in binding by the nucleoid-associated protein FIS . The reciprocal occupancy of DNA by OmpR and FIS was not due to antagonism between these transcription factors , but was instead a more intrinsic response to altered DNA topology . Surprisingly , DNA relaxation had no detectable effect on the binding of the global repressor H-NS . These results reveal the underlying molecular mechanism that primes SPI genes for rapid induction at the onset of host invasion . Additionally , our results reveal novel features of the archetypal two-component regulator OmpR . OmpR binding to relaxed DNA appears to generate a locally supercoiled state , which may assist promoter activation by relocating supercoiling stress-induced destabilization of DNA strands . Much has been made of the mechanisms that have evolved to regulate horizontally-acquired genes such as SPIs , but parallels among the ssrA , hilC , and ompR promoters illustrate that a fundamental form of regulation based on DNA topology coordinates the expression of these genes regardless of their origins .
Salmonella enterica is a facultative intracellular pathogen of the mammalian gut . After passing through the diverse environments of the stomach and digestive tract , S . enterica can invade host epithelial cells to gain access to internal tissues where it can persist inside macrophage [1] . The Salmonella pathogenicity islands 1 and 2 ( SPI-1 and SPI-2 ) encode type three secretion systems ( T3SS ) and effector proteins that enable the manipulation and invasion of host tissues [2] , [3] . SPI-1 genes are expressed primarily in the intestine during the early stages of invasion , followed by a decrease in SPI-1 expression and an increase in SPI-2 expression inside epithelial cells , and finally SPI-2 expression predominates once S . enterica has crossed the epithelium and resides in macrophage vacuoles [4] . Despite this apparently reciprocal pattern of expression over the course of invasion , both gene islands are co-regulated by many of the same global regulatory proteins . For example , SPI-1 and SPI-2 genes are strongly repressed by the nucleoid-associated protein H-NS , a highly-abundant protein that blocks and traps RNA polymerase at gene promoters by forming repressive nucleoprotein complexes [5] , [6] . SPI-1 and SPI-2 also share the transcriptional activators FIS and OmpR . FIS is required for full activation of both SPI-1 and SPI-2 genes in laboratory conditions [7] , and Δfis mutants are attenuated for virulence in mice [8] and show reduced survival in macrophage [9] . OmpR is a well-characterized direct transcriptional activator of the SPI-2 ssrAB promoter [10] , and ΔompR mutants are attenuated [11] , but the role of OmpR in SPI-1 gene expression has remained ambiguous [2] . It has been recently discovered that together OmpR and FIS drive low-level transcription of SPI-2 in the intestinal lumen , an environment classically thought to be the exclusive domain of SPI-1 [12] . By regulating expression of SPI-encoded transcription factors , H-NS , FIS , OmpR , and other global regulators sit atop a hierarchical network that integrates diverse environmental and physiological cues . SPI-encoded transcription factors fine-tune these global inputs to control precisely the dosage of T3SS and effector protein production [13] . SPI-1 encodes four AraC-like transcription factors: HilA , HilC , HilD , InvF . Through a complex feedback and feedforward mechanism , HilC and HilD control their own and each other's transcription , and together activate transcription of hilA [14] , [15] , [16] ( Figure 1A ) . HilA in turn activates invF and the genes encoding the T3SS and effector proteins [17] . Additionally , there is crosstalk between SPI-1 and SPI-2 through which HilD induces expression of ssrAB [18] ( Figure 1A ) . Unlike SPI-1 , SPI-2 encodes a single cognate regulator . Here , an unidentified signal causes the sensor kinase SsrA to phosphorylate the DNA binding protein SsrB , which in turn activates transcription of SPI-2 T3SS and effector genes [3] . SPI-1 and SPI-2 are among the best-studied genetic systems in bacteriology , yet their complex regulation has meant that the mechanisms that integrate the myriad of regulatory signals have remained enigmatic . Even less clear are the contributions made by DNA topology to the interactions and architecture of the nucleoprotein complexes that form at SPI promoters . Several lines of evidence implicate altered DNA supercoiling in coordinating SPI gene expression during invasion . The invA gene in SPI-1 , which encodes an effector protein , is repressed by relaxed DNA supercoiling [19] . Conversely , ssrA expression is induced by relaxation of DNA supercoiling [9] . S . enterica DNA is highly supercoiled in low oxygen environments but is more relaxed in oxygenated conditions , and this may reflect the DNA supercoiling dynamics that occur as S . enterica approaches the aerobic region immediately adjacent to the intestinal epithelium [20] , [21] . In tissue culture , S . enterica DNA supercoiling appears to remain static in epithelial cells but is dynamic when the bacterium resides inside macrophage [9] , which demonstrates the complexity of S . enterica's interactions with host environments . Our investigation of the links between environment , DNA supercoiling , and gene expression has uncovered a fundamental mechanism of SPI-1 and SPI-2 regulation in which relaxation of DNA supercoiling recruits OmpR to the ssrA , hilC , and hilD promoters , and this level of control functions independently of the fine-tuning effected by SPI-encoded transcription factors .
The ssrAB promoter ( PssrA ) is induced by novobiocin , an aminocoumarin antibiotic that specifically inhibits the DNA supercoiling activity of the DNA gyrase subunit B ( GyrB ) ( Figure 1B ) . In contrast , the SPI-2 T3SS and effector gene promoters , ssaB-E , sseA-G , ssaG-L , and ssaM-R , are only very slightly induced by altered DNA topology and the presumed increase in SsrA and SsrB concentrations brought about by novobiocin treatment ( Figure S1A ) . Thus , the ability of DNA relaxation to activate SPI-2 is channeled through the cognate SsrA/B two-component regulator . PssrA induction was reduced in cells lacking FIS , a master regulator of DNA supercoiling ( Figure 1B ) , possibly because novobiocin has a reduced effect on DNA supercoiling in Δfis mutants compared to wild type cells [20] . Unlike FIS , OmpR and its phospo-donor EnvZ were both absolutely required for induction of PssrA , suggesting that relaxed DNA supercoiling alone cannot activate PssrA in the absence of OmpR's ability to recruit RNAP . The requirement for EnvZ indicates that OmpR must be phosphorylated in order to stimulate these promoters , and also indicates that other phospho-donors do not activate OmpR in these conditions . The SPI-encoded regulators HilD and SsrA/B played no detectable role in PssrA induction . The alternate sigma factor RpoS is better at transcribing relaxed DNA than is the primary housekeeping sigma factor RpoD [22] , and the elevated level of RpoS during stationary phase correlates with ssrA expression in standard laboratory conditions , but deletion of rpoS did not reduce PssrA induction by novobiocin ( Figure S1B ) . Previous experiments in S . enterica have shown that the ompR-envZ promoter ( PompR ) is induced by high concentrations of novobiocin at late stages of growth , and that OmpR is an auto-regulator of this induction [23] . We found that PompR is also activated by low concentrations of novobiocin during exponential growth , and FIS and OmpR-EnvZ contribute to this induction ( Figure 1B ) . Deletion of hilD did not affect PompR induction . However , PompR activity was unexpectedly elevated in the ΔssrA/B mutant , suggesting that SsrA/B may directly or indirectly regulate ompR expression . Because SPI-1 and SPI-2 genes are usually observed to have inverse expression patterns , we expected SPI-1 genes to be insensitive or repressed by DNA relaxation . We tested the effects of novobiocin treatment on expression of the master regulators hilA , hilC , and hilD , and were surprised to find that both the hilC and hilD promoters ( PhilC and PhilD ) were induced by DNA relaxation ( Figure 1B ) . PhilA was insensitive to DNA relaxation ( Figure S1A ) , suggesting that the inducing signal is limited to PhilC and PhilD ( Figure 1A ) . Like PssrA , PhilC required both FIS and OmpR-EnvZ for induction; yet unlike PssrA , the absence of FIS was not compensated by increasing concentrations of novobiocin . Consistent with its role as a transcriptional activator , HilD was required for full activation of PhilC ( Figure 1B ) . SsrA/B did not contribute to PhilC induction . PhilD was unique among the four promoters in having higher expression in the absence of ompR and envZ , but it nevertheless required FIS for full activation ( Figure 1B ) . PhilD induction was unaffected by the absence of SsrA/B or HilD . Quantitative PCR measurement of ssrA , ompR , hilC , hilD , and hilA mRNA levels confirmed the results obtained from the reporter gene fusions ( Figure S1C ) . Having found that OmpR and relaxed DNA supercoiling work in concert to stimulate transcription from PssrA , PhilC , and PompR , we wished to test the relative contributions of OmpR and DNA topology to promoter function . To this end , the ompR-envZ operon ( ompB ) was cloned under the control of the arabinose-inducible PBAD promoter in a ΔompB mutant . PssrA expression increased only very slightly when ompB was overexpressed ( 0 . 2% arabinose ) in the absence of DNA relaxation ( Figure 2 ) . In contrast , DNA relaxation in the complete absence of OmpR ( empty pBAD vector ) had a stimulatory effect on PssrA; this activation was higher in cells carrying pBADompB , likely due to leaky transcription of ompB in the absence of arabinose . The combination of ompB over-expression and DNA relaxation had the strongest stimulatory effect , confirming that OmpR and DNA relaxation work in concert to activate PssrA . The combination of ompB over-expression and DNA relaxation also resulted in maximal expression of PhilC and PompR , however the effect was more subtle for PhilC ( Figure 2 ) . Consistent with the results presented in Figure 1B , PhilD was repressed by ompB expression , and this repression occurred in both the absence and presence of DNA relaxation . Repression of PhilD was strongest at the lower concentration of novobiocin ( 15 µg/ml ) , raising the possibility that a high degree of DNA relaxation reduces repression by OmpR , perhaps through elevated HilC levels brought about by DNA relaxation ( Figure 1A ) . The control of hilC and hilD expression by OmpR suggested that OmpR may regulate these genes through direct interactions . Electrophoretic mobility shift ( bandshift ) assays confirmed that OmpR binds specifically to both PhilC and PhilD , with OmpR demonstrating an affinity for PhilC similar to that for the positive control PompC ( Figure 3A ) . A negative control bait DNA ( kan ) was not bound specifically by OmpR at the concentrations tested . In these equilibrium binding assays , the rapid appearance of OmpR-DNA complexes over a small range of protein concentrations was evidence of cooperative OmpR binding to the bait DNA . Moreover , OmpR-DNA complexes demonstrated slower migration at higher OmpR concentrations , indicating that multiple OmpR molecules were bound to a single bait DNA molecule . Cooperative DNA binding is a feature common among NAPs— like FIS and H-NS— that bind with low-specificity to multiple proximal DNA sites [24] , [25] . Indeed , OmpR monomers are thought to first bind cooperatively to form a nucleating dimer that recruits additional OmpR dimers in a cooperative fashion [26] . DNase footprinting revealed that PhilC and PhilD each have a single region protected by OmpR ( Figure 3B ) . The PhilC region bound by OmpR is located over 100 bp upstream of the hilC start codon , consistent with OmpR's function as a transcriptional activator of this promoter . Conversely , the OmpR-protected region of PhilD is downstream of the hilD start codon , where OmpR binding is likely to have a repressive effect on hilD transcription . If OmpR function is enhanced by relaxation of DNA supercoiling , does DNA relaxation result in increased DNA binding by OmpR ? This was tested by quantifying OmpR binding to gene promoters in vivo using chromatin immuno-precipitation ( ChIP ) . A fusion of the 3×-Flag epitope tag to the C-terminus of OmpR was used for these experiments . The epitope tag added 22 amino acids adjacent to OmpR's DNA-binding domain and created a new ribosome binding site for the envZ open reading frame; nonetheless , cells with OmpR:Flag showed only a slight reduction in promoter activation by novobiocin ( Figure S1C ) . Alternatively , a 3×Flag tag at the N-terminus of OmpR could not be used because it generated a ΔompR phenotype at promoters ( not shown ) . Novobiocin treatment caused a significant increase in OmpR occupancy at PssrA , PompR , PhilC , and PhilD ( Figure 4A ) . Increased promoter occupancy was due solely to a change in binding activity as OmpR levels were observed to decrease after novobiocin treatment ( Figure 4B ) . Because OmpR requires DNA relaxation for it to be fully active at SPI promoters ( Figure 2 ) , we predicted that OmpR is an ineffective antagonist of H-NS binding and thus requires novobiocin-induced changes in DNA topology to assist in H-NS displacement . Our H-NS ChIP results confirm earlier studies that have found H-NS to occupy SPI promoters , but demonstrates a low affinity for PompR [5] , [27] ( Figure 4A ) . Surprisingly , at all four promoters H-NS abundance was not affected by DNA relaxation nor by increased OmpR binding . It is important to note however that while ChIP quantifies protein abundance at genomic regions at a resolution around 500 bp , ChIP does not resolve changes to higher-order protein complexes if protein abundance remains constant . Therefore , we cannot rule out that although H-NS is not displaced , promoter activity may increase because H-NS oligomers are restructured by OmpR binding as well as by changes in DNA topology . Chromatin immuno-precipitation of FIS fused to a 8×Myc epitope tag has been used previously to examine genome-wide FIS binding in E . coli [28] . We constructed an identical FIS:Myc fusion protein in S . enterica and used this to measure FIS occupancy of gene promoters in our experimental conditions . This revealed a high abundance of FIS at the SPI promoters , with slightly less FIS bound to PompR ( Figure 4A ) . At all loci tested , FIS occupancy decreased when cells were treated with novobiocin . This reduced FIS occupancy can be explained mostly by the ∼50% decrease in FIS levels in novobiocin-treated cells ( Figure 4B ) . Although FIS contributes to transcriptional activation of these promoters , the finding that transcriptional activation occurs even when FIS is depleted suggests that FIS may act in part through its global control of DNA topology . Because FIS transitions from a filamentous DNA-binding mode to an ordered dimer as its concentration decreases [24] , it is also possible that the depletion of FIS coupled with changes in DNA topology restructures FIS complexes into forms that favour transcription activation . We next tested whether the decrease in promoter activity observed in a Δfis mutant ( Figure 1B ) was due to a reduced ability by OmpR to access gene promoters . The ChIP data suggest that both OmpR and H-NS have less access to promoter DNA in a Δfis mutant ( Figure 4A ) . We have previously found that the Δfis mutant is resistant to relaxation of DNA supercoiling by novobiocin [20] . It may be that H-NS and OmpR require DNA relaxation to gain full access to PssrA , PompR , PhilC , and PhilD , and the degree of relaxation is too modest in the Δfis mutant . Nevertheless , novobiocin treatment caused a small increase in OmpR occupancy in Δfis mutants , indicating that OmpR binding does not absolutely require the topological constraints imposed on DNA by FIS binding . The same experiment was conducted in a ΔompR mutant . In the absence of OmpR , novobiocin treatment caused a reduction in FIS binding ( Figure 4A ) , again consistent with a reduction in FIS levels in these cells ( Figure 4B ) . Although less H-NS bound to SPI promoters in the ΔompR mutant , significantly more H-NS bound to PompR , suggesting that OmpR is an effective H-NS antagonist at its own promoter . Surprisingly , the reduced H-NS levels observed in the ΔompR mutant ( Figure 4B ) , along with the further reduction in H-NS levels upon novobiocin treatment , implicates OmpR as a regulator of hns expression . Because DNA relaxation does not appear to displace H-NS from gene promoters , we tested how removing H-NS from the system affects promoter function . Although all four test promoters had a similar pattern of induction by DNA relaxation in wild type cells , contrasting responses were observed in the absence of H-NS . As expected , all three SPI promoters were strongly upregulated ( 20 to 200-fold ) in the Δhns mutant ( Figure 4C ) . In the absence of H-NS , ssrA was induced , hilC was repressed , and hilD was unaffected by DNA relaxation . These contrasting responses may result from the different and complex regulatory inputs acting at each promoter , and further confirm that promoter induction by DNA relaxation is not due simply to antagonism of H-NS repression . Transcriptional output from PompR was the same in wild type and Δhns mutant cells in normal growth conditions ( Figure 4C ) . This finding that H-NS does not repress PompR is consistent with the low affinity of H-NS for this promoter ( Figure 4A ) . Surprisingly though , PompR was not induced by novobiocin in the Δhns mutant , which may be indirectly caused by the highly pleiotropic effects of the Δhns mutation . To determine how DNA supercoiling affects OmpR affinity for DNA , we used primer extension to resolve DNase footprints on supercoiled and linear DNA templates . This approach can also determine if OmpR binds to different target sites depending on DNA supercoiling state , thus PompR was used as the target DNA in this set of experiments because it has multiple , clearly delineated OmpR binding sites [23] . The grey filled boxes in Figure 5A highlight the regions of PompR protected from DNase I digestion by OmpR . Most protection in the absence of supercoiling ( linear DNA ) was observed at a 60 bp region , OmpR-2 , with lesser protection of regions on either side . Protected regions were assigned numbers to correspond with the OmpR sites identified previously by Bang et al . [23] ( horizontal , dark-grey lines ) . Unlike Bang et al . [23] , we analyzed OmpR binding to the full PompR intergenic region to resolve a more promoter-distal site , OmpR-4 . DNase I digestion of an end-labeled linear template confirmed that OmpR-2 is the primary site of OmpR binding to PompR ( Figure S2A ) . As supercoiling levels increased , protection by OmpR appeared to decrease ( Figure 5A ) , giving the impression that OmpR binds DNA better at lower supercoiling levels . However , this result was caused by a decrease in DNase I cutting of unprotected supercoiled DNA , perhaps due to DNA compaction and the loss of B-DNA conformation at higher superhelical densities . The amount of DNase I digestion in the presence of OmpR was consistent regardless of the superhelical density of PompR DNA . Thus , OmpR may reduce DNase I digestion across the entire promoter region by constraining a supercoiled-like state in DNA , as has been observed for FIS and H-NS [29] , [30] . DNA positions that are hypersensitive to endonuclease cutting offer additional insight into changes in DNA topology . For example , positions −207 and −182 became less sensitive whereas positions −154 and −9 became increasingly sensitive to DNase I digestion as supercoiling increased . Position −154 ( marked with an asterisk ) is particularly intriguing because it was ultra-hypersensitive to DNase I digestion when DNA was supercoiled . When DNA was fully relaxed , OmpR binding greatly enhanced DNase I cutting at position −154 , supporting a model in which OmpR binding creates DNA structures similar to those induced by negative DNA supercoiling . DNA supercoiling exerts torsional stress that weakens base pairing , and so reduces the amount of energy needed for DNA melting and transcription initiation . This is referred to as stress-induced duplex destabilization ( SIDD ) , and the energy required for strand separation at each base pair in a specific sequence , G ( x ) , can be predicted for different superhelical densities [31] . Stable base pairs have G ( x ) values around 10 , whereas lower values indicate positions prone to SIDD . We used WebSIDD [32] to predict the stability of PompR DNA at the approximate superhelical densities observed during exponential growth ( σ = −0 . 06 ) and after treatment with 15 µg/ml novobiocin ( σ = −0 . 045 ) . The G ( x ) profiles of PompR at both superhelical densities revealed a highly destabilized region ranging from positions −70 to −160 , with a weakly destabilized region ( −20 to −55 ) encompassing the ompR transcription start sites ( Figure 5B ) . This analysis makes the counterintuitive prediction that PompR becomes increasingly destabilized as DNA relaxes , which is nevertheless consistent with the observed gene activation in these conditions ( Figure 1B ) . DNA upstream of position −170 is highly stable , indicating that the destabilization effect is specific and concentrated at the main binding site used by OmpR . The primary OmpR binding sites , OmpR-1 and OmpR-2 , cover most of the highly destabilized regions , raising the possibility that OmpR binding transmits the destabilizing force to the adjacent RNA polymerase binding site where DNA strand separation can assist in transcription initiation . A similar SIDD-transmission function has been characterized for FIS and IHF [33] . Novobiocin treatment caused the same degree of DNA relaxation in ΔompR mutant cells as in wild type cells ( Figure S2B ) . This suggests that promoter DNA experiences the same stress-induced strand destabilization in both mutant and wild type , indicating that reduced promoter activation in the ΔompR mutant ( Figure 1B ) is due to the absence of OmpR binding , not an altered degree of DNA relaxation . In other words , DNA strand destabilization caused by DNA supercoiling is insufficient for PompR activation in the absence of OmpR binding .
S . enterica traverses various extracellular and intracellular environments during infection of host tissues , thus it requires genetic programs capable of balancing shifting requirements for the T3SS and effector proteins that mediate the invasion process . Two recent studies unexpectedly discovered that SPI-2 genes are expressed in the mouse intestinal lumen prior to cellular invasion , leading to the hypotheses that SPI-2 is either important for colonization of the intestine or requires priming before intracellular invasion [12] , [34] . SPI-2 expression during growth in rich medium , which roughly mimics conditions in the intestinal lumen , requires OmpR and FIS but is independent of SsrB , SlyA , and PhoP [12] , [18] . Here we describe a fundamental mechanism that activates both SPI-2 and SPI-1 promoters through changes in DNA topology , and this mechanism depends on OmpR and FIS but is independent of SsrB and HilD . It is intriguing that PssrA induction does not require FIS in culture conditions that mimic the vacuolar environment [12] , nor is FIS required for PssrA induction when DNA is highly relaxed [20] . These findings support a model in which fine-tuning of SPI gene expression by factors such as SsrB , HilD , SlyA , and PhoP may occur primarily in the vacuolar environment . Although OmpR-EnvZ is the archetypal two component signal transduction pathway , the environmental stimulus of EnvZ kinase activity remains unclear [35] , and this stimulus appears to differ between E . coli and S . enterica [36] . It is perhaps for this reason that a role for OmpR in the regulation of SPI-1 has been enigmatic . Previous studies have found no effect or only weak effects from deletion of ompR or envZ . It was initially proposed that OmpR-EnvZ directly controls hilC expression [37] , but others have favoured a model in which OmpR somehow acts post-transcriptionally through HilD protein function [2] . Here we provide evidence that under conditions of relaxed DNA supercoiling , OmpR binds directly to both the hilC and hilD promoters where it activates the former and represses the latter . There is a growing body of evidence that in addition to the classic role of OmpR as a site-specific transcription factor that activates gene expression through RNAP recruitment , it also exhibits NAP-like features and functions . Because OmpR makes few specific contacts with DNA , it demonstrates an affinity for non-specific DNA [38] , [39] . The preferred OmpR target sites , which are highly degenerate at the sequence level , may serve as nucleating points for cooperative recruitment of additional OmpR molecules . Nucleation and cooperative DNA binding can explain the broad regions of OmpR protection observed at PhilC ( Figure 3B ) , PompR ( Figure 5A ) , and PssrA [10] . Additionally , our ChIP data revealed OmpR binding to regions not predicted to be specific targets ( Figure S2C ) . This is similar to the ChIP survey of cAMP receptor protein ( CRP ) targets in E . coli which revealed a high background of CRP binding across the entire chromosome [40] . CRP binding to thousands of weak sites lead the authors to propose that this archetypal transcription factor should also be considered as a chromosome-structuring protein . An additional interesting parallel between OmpR and CRP is that both are calculated to have similar cellular concentrations: ∼3 , 000 CRP and ∼3 , 500 OmpR molecules/cell [41] , [42] . Thus , both OmpR and CRP may represent a class of NAP-like DNA-binding proteins whose high abundance allows them to have a broad influence on chromosome shape and function , yet whose modulation through allosteric effectors generates titratable DNA-binding modes that preferentially target specific promoters . The discovery that DNA relaxation results in increased OmpR binding to DNA in vivo presents an intriguing model in which this mechanism is complementary to phosphorylation of OmpR by EnvZ as a means to stimulate OmpR-DNA binding . Thus , phosphorylated OmpR may be recruited to promoters by DNA relaxation . Future global analysis of OmpR binding to the S . enterica chromosome will shed light on the relative contributions of phosphorylation and DNA relaxation to OmpR-DNA interactions . Whole-genome analysis of the transcriptional consequences of DNA relaxation in E . coli revealed that relaxation-induced promoters are significantly more A+T-rich than are uninduced promoters [43] . Because H-NS preferentially binds to regions of high A+T content , relaxation-induced promoters are very likely to be H-NS repressed . In E . coli , the OmpR targets PompR and PompC were induced whereas PompF was repressed by novobiocin [43] , and we found the same response in S . enterica ( Figure S2D ) . This shared response of OmpR-regulated genes to novobiocin in E . coli and S . enterica coupled with the proposed ability of DNA relaxation to weaken H-NS repression hints at an evolutionarily conserved gene regulatory mechanism that predates horizontal acquisition of SPI-1 and SPI-2 by Salmonella . Transcriptional activators of SPI genes ( HilC , HilD , SsrB , and SlyA ) function in large part through displacement of H-NS from SPI promoters [16] , [44] . Members of the AraC-like protein family , which includes HilC and HilD , have a well-documented ability to displace H-NS [45] . SsrB , a member of the NarL protein family , activates transcription by displacing H-NS but does not appear able to break H-NS bridges [44] . OmpR and SlyA are winged-helix DNA binding proteins . Like OmpR , SlyA relieves H-NS repression without displacing H-NS [46]; SlyA also generates regions of DNase I hypersensitivity , thus may have a topological restructuring mode that contributes to breaking H-NS bridges [46] . However , unlike OmpR , SlyA relies on activators such as PhoP to recruit RNAP . Variable modes of H-NS antagonism — from anti-polymerization by HilD , HilC , and SsrB to anti-bridging by OmpR and SlyA — may represent a gate-keeper mechanism that selects which of the numerous regulators known to act at SPI gene promoters are allowed access to their target DNA sites , thus fine-tuning transcriptional output .
Salmonella enterica serovar Typhimurium strain SL1344 was used for all experiments . Detailed descriptions of mutant strains used in this study are provided in Table S1 . E . coli XL-1 blue was used for all cloning steps . Strains and plasmids used in this study are listed in Table S1 . To generate S . enterica mutants , the kanamycin resistance cassette was PCR amplified from pKD4 [47] using primers listed in Table S2 , which were designed to replace only open reading frames . Because the ompR and envZ open reading frames overlap , special care was taken to preserve open reading frames when constructing deletion and epitope-fusion mutations . PCR amplicons were spin column purified then transformed into electrocompetent S . enterica SL1344 containing the Red helper plasmid pKD46 as previously described [47] , [48] . Mutations were transduced into a fresh SL1344 background by bacteriophage P22 generalized transduction [49] , then were confirmed by DNA sequencing . Site-directed mutagenesis of the ompB locus cloned in pUC18 was carried out using the QuikChange II kit ( Stratagene ) and primers listed in Table S2 , following the manufacturer's protocol . Transcriptional reporter fusions were constructed by cloning gene promoters in pZep and pZec vectors , which contain a promoterless gfp+ gene [20] , [50] . The cat gene was removed from pZep to generate pZec so that transcriptional reporter fusions could be cloned in cells containing pBAD . The removal of the cat gene from pZep to generate pZec had no effect on expression of cloned promoters ( not shown ) . Data presented in Figure 1 and Figure 2 is from cells carrying pZec reporter plasmids . pZep or and pZec plasmids were digested with SmaI and XbaI , spin column purified using the HiYield PCR DNA Fragment Extraction Kit ( RBC Bioscience ) , and dephosphorylated with Antarctic Phosphatase ( New England Biolabs ) . Gene promoter sequences were PCR amplified using the Phusion DNA polymerase ( NEB ) and primers listed in Table S2 . SmaI and XbaI digested amplicons were spin column purified , then ligated to pZep or pZec by T4 ligase ( Roche ) . The exception was the ompR promoter region which was digested with NotI and XbaI before cloning into similarly digested vector . The ompB locus along with its native ribosome binding site was PCR amplified using primers listed in Table S2 , followed by PstI and SacI digestion and spin column purification . PstI and SacI digested pBAD33 was gel purified , de-phosphorylated by Antarctic Phosphatase ( NEB ) , then ligated to the digested PCR amplicon . The effects of inducible ompB on promoter function were assessed by measuring GFP levels in cells carrying both pBADompB and pZec transcriptional reporter clones . Arabinose and novobiocin were added to cultures at the concentrations indicated in Figure 2 . Cells were cultured in a shaking waterbath at 37°C in LB ( 1% tryptone and 0 . 5% yeast extract ) without any NaCl added . Cells used for gfp reporter fusion experiments were cultured in 4 ml of LB in glass tubes ( interior diameter 14 mm ) shaking at 200 RPM whereas cells used for ChIP and quantitative PCR experiments were cultured in 55 ml of LB in 250 ml glass flasks shaking at 140 RPM . Previous studies testing the effects of novobiocin on gene expression in S . enterica have used high concentrations of novobiocin ( 25–150 µg/ml ) in cells transitioning from late exponential to stationary phase physiology [9] , [23] , thus introducing additional variables arising from growth phase transitions . To ensure a steady state of growth , we conducted all experiments using cells that had been growing exponentially for more than six doublings at low cell density ( OD600 less than 0 . 3 ) . In addition , we used low concentrations of novobiocin ( 15–25 µg/ml ) to minimize effects on growth rate . Cells were fixed after 3 hrs of continued growth at low density in the presence or absence of novobiocin ( final OD600 0 . 1–0 . 25 ) , as in [20] . Total RNA was isolated from cultures using the SV Total RNA Isolation System ( Promega ) and purity and quality was assessed by electrophoresis in 1% agarose ( 1×TAE ) . For each sample , 5 µg total RNA was DNase treated in a 50-µl reaction using the Turbo DNA-free kit ( AMBION ) , and cDNA templates were synthesized by random priming 0 . 5 µg RNA in a 20 µl reaction using the GoScrip Reverse Transcription System ( Promega ) . Quantitative PCR ( qPCR ) primers are listed in Table S2 . PCR reactions were carried out in duplicate with each primer set on an ABI 7500 Sequence Detection System ( Applied Biosystems ) using FastStart SYBR Green Master with ROX ( Roche ) . Standard curves were included in every qPCR run; standard curves were generated for each primer set using five serial tenfold dilutions of S . enterica chromosomal DNA . ChIP was conducted as previously described [27] . Two ChIP replicates were performed using a strain containing both the ompR:flag and fis:myc epitope fusions , allowing for simultaneous precipitation of OmpR:Flag , Fis:Myc , and H-NS from the same biological sample . One ChIP replicate was conducted for each strain carrying a single epitope tag ( ompR:flag or fis:myc ) . ChIP results overlapped between the double and single fusion strains , indicating that the epitope-tagged proteins did not negatively affect nucleoprotein interactions when combined . Precipitated DNA was quantified by quantitative PCR using primers listed in Table S2 . Cells were pelleted and resuspended in 1× Laemmli buffer ( 4% SDS , 20% glycerol , 10% 2-mercaptoethanol , 0 . 004% bromophenol blue , 0 . 125 M Tris HCl , pH 6 . 8 ) and denatured at 100°C for 5 min . Samples were electrophoresed on 15% polyacrylamide SDS gels . Gels and nitrocellulose membranes were equilibrated in transfer buffer ( 25 mM Tris HCl , 192 mM glycine , 0 . 02% SDS , 20% methanol ) and proteins were transferred to membranes at 150 V for 90 min using a Trans-blot ( BioRad ) apparatus packed in ice . Membranes were blocked overnight at 4°C in 5% non-fat powdered milk in PBS ( 137 mM NaCl , 12 mM Phosphate , 2 . 7 mM KCl , pH 7 . 4 ) , followed by incubation at room temperature for 2 hr with rocking in primary antibodies diluted as follows: 1/100 , 000 anti-DnaK mAb rabbit ( Enzo Life Sciences ) , 1/10 , 000 anti-FLAG mAb rabbit ( Sigma ) , 1/10 , 000 anti-Myc mAb rabbit ( Sigma ) , and 1/5 , 000 anti-H-NS polyclonal mouse [5] . Blots were washed thoroughly and probed with horseradish peroxidase-linked anti-rabbit and anti-mouse antibodies ( Millipore ) diluted 1/5 , 000 in PBS ( 1% blocking agent ) for 1 hr at room temperature with rocking , followed by thorough washing . Blots were incubated in ECL reagent ( Pierce ) for 1 min , and bands were visualized using an ImageQuant LAS 4000 scanner ( GE Healthcare ) then quantified using ImageJ v1 . 43 ( National Institutes of Health , U . S . A . ) . Probing for all proteins ( DnaK , FIS:Myc , H-NS , and OmpR:Flag ) simultaneously on the same blot allowed for protein quantities to be normalized to the internal standard ( DnaK ) and expressed relative to one another . Each cell sample was run on three independent western blots to improve the accuracy of quantification . Thus , the protein abundance value for each biological replicate is the average value from three replicate blots . The OmpR D55E mutation creates a constitutively active protein by mimicking phosphorylation [51] . OmpR ( D55E ) with a C-terminal His-tag was purified and used in bandshifts and DNase I footprinting . BL21 cells carrying the pET21-ompR ( D55E ) plasmid were grown in L broth ( 0 . 5% NaCl; 100 µg/ml carbenicillin ) and ompR ( D55E ) expression was induced at OD600 0 . 5 with 1 mM IPTG . Cells were harvested after 4 . 5 hr by centrifugation and the pellet were frozen overnight at −20° . Native OmpR ( D55E ) -His was purified as follows: the pellet was resuspended in lysis buffer ( 50 mM sodium phosphate , 300 mM sodium chloride , 10 mM imidazole ) , then treated with 1 mg/ml lysozyme for 30 min at 24°C followed by sonication on ice . Insoluble material was removed by centrifugation at 10 , 000 g for 25 min and the supernatant was then incubated with nickel-nitriloacetic acid agarose beads for 1 hr at 4°C with gentle rocking . The agarose beads were loaded in a column and washed twice with four column volumes of wash buffer ( 50 mM NaH2PO4 , 300 mM NaCl , 20 mM imidazole , pH 8 . 0 ) , and protein was collected in elution buffer ( 50 mM NaH2PO4 , 300 mM NaCl , 250 mM imidazole , pH 8 . 0 ) . Purified protein was desalted with Nanosep 3K Omega membranes ( Pall ) at 4°C , then resuspended in storage buffer ( 20% glycerol , 40 mM Tris , 200 mM KCl ) and stored at −80° . OmpR ( D55E ) purity was assessed on Coomassie stained SDS-PAGE gels , and concentration was quantified by both the Bradford assay and by comparison to protein standards on Coomassie stained SDS-PAGE gels . Bait DNA was PCR amplified using the primers pZec . 6FAM . R ( labeled with a 5′ 6-FAM fluorophore ) and pZec . confirm . F ( Table S2 ) , from pZec promoter clones . Amplicons were spin column purified then used as bait DNA in bandshifts . OmpR-DNA binding reactions ( 10 µl ) contained 0 . 2× TBE ( 89 mM Tris , 89 mM borate , 2 mM EDTA ( pH 8 . 3 ) , 40 µg/µl poly ( dI-dC ) DNA , and 40 nM bait DNA . Reactions were incubated at room temperature for 15 min before being loaded onto a running polyacrylamide gel ( 30∶1 acrylamide/bisacrylamide , 0 . 2× TBE , 2% glycerol ) with 0 . 2× TBE running buffer . After electrophoresis for 40 min at 100 V , 6-FAM-labeled DNA was visualized using a Typhoon scanner ( GE Healthcare ) . Bait DNA was prepared as for bandshifts using either the primer sets pZec . 6FAM . F and pZec . confirm . R ( top strand ) or pZec . 6FAM . R and pZec . confirm . F ( bottom strand ) . DNase I footprinting reactions were conducted in 15 µl reaction volumes containing 1× DNase I buffer ( Roche ) ( 40 mM Tris-HCl , 10 mM NaCl , 6 mM MgCl2 , 1 mM CaCl2; pH 7 . 9 ) , 0 . 01 mM dithiothreitol , 100 ng/µl BSA , 50 nM bait DNA , and 5 µM OmpR ( D55E ) -His . OmpR-DNA binding was allowed to equilibrate at 37°C for 15 minutes , then 1 µl ( 0 . 015 units ) of pre-warmed DNase I was added and mixed gently , then incubated at 37°C for 10 minutes . Reactions were stopped by addition of 2 µl EDTA ( 100 mM ) followed by vigorous vortex mixing and heat denaturation at 95°C for 10 min . Digestion products were desalted using MicroSpin G-50 columns ( GE Healthcare ) and were analyzed on an ABI 3130 Genetic Analyzer along with GeneScan 500-LIZ size standards ( Applied Biosystems ) . Plasmids with varying degrees of superhelical density were generated as follows: pZec-PompR was purified from E . coli CSH50 at different topological states by growing cells overnight in 25 ml L ( 0% NaCl ) in a well-aerated 250 ml glass flask ( low supercoiling ) or overnight in 6 ml L ( 0 . 5% NaCl ) in 14 mm diameter glass culture tubes ( high supercoiling ) . Topoisomers at the desired topological state were purified after separation on a 1% agarose gel containing 2 . 5 µg/ml chloroquine . The average superhelical density of each purified plasmid pool was determined by calculating the linking difference between the dominant topoisomer and fully relaxed DNA [52] . A 1% agarose gel containing 25 µg/ml chloroquine was used to improve the resolution of topoisomers in the low supercoiling sample; in these conditions , more relaxed DNA migrates faster through the gel . To generate a plasmid pool that lacked supercoiling , pZec-ompR was digested with XhoI , which cuts 500 bp away from the cloned promoter . Because nicking of plasmid DNA by DNase I will allow DNA supercoils to relax , footprinting reactions were treated with DNase I for no more than 1 minute to reduce the time in which nicked plasmids could lose their topology . In these reactions , 1 µl ( 0 . 15 units ) of DNase I was added to 15 µl OmpR-DNA binding reactions containing 10 µM OmpR and 1 . 5 nM bait DNA . OmpR-DNA binding was equilibrated as above . Primer extension was conducted using the primer pZec . 6FAM . F and Thermo Sequenase polymerase ( USB ) with the following thermocycle: 95°C for 30 sec , 53°C for 30 sec , and 72°C for 90 sec , repeated 50 times . Extension reactions contained 0 . 4 nM of nicked plasmid template . Amplification products were desalted using MicroSpin G-50 columns ( GE Healthcare ) and were analyzed on an ABI 3130 Genetic Analyzer along with GeneScan 500-LIZ size standards ( Applied Biosystems ) . To compare samples , each was normalized to have the same total fluorescence signal across the DNA region being analysed . | DNA is often considered to be a passive carrier of genetic information , but in fact DNA is an active participant in coordinating the expression of the genes it carries . This is because DNA is a dynamic molecule that can assume a wide range of topologies , and this has a direct impact on the formation of the protein–DNA complexes that drive gene expression . In a bacterium , the chromosome is supercoiled to variable levels according to environmental conditions , and supercoiling in turn governs the topology of gene promoters . Thus DNA supercoiling is able to transduce environmental signals to regulate promoter output . A previous study found that the intestinal pathogen Salmonella enterica may use changes in DNA supercoiling to detect when it has entered host immune cells , allowing the bacterium to induce the pathogenicity genes it requires to evade killing by macrophage . In dissecting the underlying molecular mechanisms , we have found that changes in DNA supercoiling also upregulate other key pathogenicity genes , and we have identified the proteins involved in this gene regulatory process . These findings indicate that a fundamental level of gene control arising from the interplay between protein transcription factors and DNA topology regulates Salmonella pathogenicity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"nucleic",
"acids",
"proteins",
"gene",
"expression",
"genetics",
"dna",
"biology",
"molecular",
"genetics",
"microbiology",
"genetics",
"and",
"genomics",
"bacterial",
"pathogens"
] | 2012 | A Fundamental Regulatory Mechanism Operating through OmpR and DNA Topology Controls Expression of Salmonella Pathogenicity Islands SPI-1 and SPI-2 |
In many settings , copying , learning from or assigning value to group behavior is rational because such behavior can often act as a proxy for valuable returns . However , such herd behavior can also be pathologically misleading by coaxing individuals into behaviors that are otherwise irrational and it may be one source of the irrational behaviors underlying market bubbles and crashes . Using a two-person tandem investment game , we sought to examine the neural and behavioral responses of herd instincts in situations stripped of the incentive to be influenced by the choices of one's partner . We show that the investments of the two subjects correlate over time if they are made aware of their partner's choices even though these choices have no impact on either player's earnings . We computed an “interpersonal prediction error” , the difference between the investment decisions of the two subjects after each choice . BOLD responses in the striatum , implicated in valuation and action selection , were highly correlated with this interpersonal prediction error . The revelation of the partner's investment occurred after all useful information about the market had already been revealed . This effect was confirmed in two separate experiments where the impact of the time of revelation of the partner's choice was tested at 2 seconds and 6 seconds after a subject's choice; however , the effect was absent in a control condition with a computer partner . These findings strongly support the existence of mechanisms that drive correlated behavior even in contexts where there is no explicit advantage to do so .
Humans learn a range of information from one another [1] and show a particular sensitivity to the influence of group behavior [2] . The ultimate evolutionary origins of these behaviors and their dependence on other relevant variables raise broad-ranging questions [3]–[6] however , they also invite important but narrower questions about the human propensity to assign value to the behavior of others even when there exists no external incentive to do so . Such assignments can reasonably be considered irrational because they explicitly violate external incentive structures . It has been suggested that this propensity to ‘follow-the-crowd’ – even in the face of information that suggests otherwise – is the basis of a range of herding behaviors displayed by humans interacting through markets including both bubbles and crashes [7]–[11] . One hypothesis for the origin of this class of ‘believe-the-group’ irrationalities is that while long ago group behavior tended to be a good proxy for value , the complexities of modern life , and especially modern markets , subvert this tendency , producing unpredictable behaviors in market settings . We used a tandem ( two-person ) sequential choice experiment , framed as a market investment task , to test the degree to which neural and/or behavioral responses change depending solely on the behavior of one's partner , and whether they do so in the absence of incentives . The task asks a subject to invest some fraction ( from 0 to 1 ) of their total holdings , shows the change in the market value which controls gains and losses , and later shows the fraction invested by their partner ( Figure 1 ) . The partner's investment has no bearing on the payoff of the subject or on the market's future movements . In addition to this tandem task we included a control condition in which subjects played in tandem with a computer that chose its investments randomly ( uniformly over [0 , 1] ) . In this control condition , subjects were informed that the other “investor” was a computer and that its choices were random . This experiment asks two empirical questions . ( 1 ) Does a subject change their behavior based on the difference with their partner's choice ( Jones ) ? ( 2 ) How does the brain respond to the difference between the subject's investment level and their partners ? We repeated the experiment twice and varied the time at which the partner's choice was revealed ( 2 seconds and 6 seconds after the subject's choice ) . In this task , there is no incentive for the answer to either question to be yes; however , a positive answer to either suggests that group behavior is deemed valuable by brain and behavior even in the absence of external economic incentives . The striatum is well known for encoding “prediction error” signals that aid humans and other animals learn the value of various stimuli and actions; therefore , we hypothesized that the “interpersonal prediction error” , i . e . the difference between the partner's and the subject's own bet ( henceforth referred to as Jones ) , would ( a ) correlate with activation in the striatum and ( b ) correlate with the bet in the next round of play . Hypothesis ( a ) is based on the idea that the subject's brain assumes that this difference with the partner's bet is an informative error signal . Hypothesis ( b ) – the idea that this difference would correlate with a tendency to adjust ones behavior toward that of the other investor – suggests one bias that would encourage irrational herding behavior . The setup for our tandem investment task and our framing of the behavior in terms that inform our notion of irrational herding behavior is also supported by economic ideas . Economists have laid out the theory of information cascades – situations where rational agents disregard their private signals and follow the choices of others [9] , [12] , [13] ‘as though’ the others have different or better information . This tendency to herd is also thought to play a role in more complicated situations , such as financial markets , where the phenomenon may lead to bubbles and crashes [14] . Recently neuroscientists have begun to explore the neural underpinnings of social learning [15]–[23] . We extend these results to consider the effects of others' past investment behavior on subsequent investment behavior when the risk parameters of the underlying market are fundamentally unknown . We hypothesized that modulation of the error signals in the ventral striatum would reflect the influence of social information on investment behavior .
To examine differences among the three versions of the experiment we performed a mixed-effects linear regression separating the three groups ( 2 sec human , N = 68; 6 sec human , N = 24; 6 sec computer control , N = 24; see Tables S1 , S2 , S3 for demographic information ) using indicator functions for the three groups ( interacting with all of the variables of interest ) . The dependent variable was the normalized investment . The independent variables in the regression were a constant , the normalized previous bet , the previous market return ( MKT ) , and a variable we call DJONES , equal to the difference between the other subject's investment and the subject's investment . Here we focus on the regression coefficient of DJONES ( Figure 2 ) . The coefficients from the 2 sec and 6 sec human experiments are both significantly greater than zero , and the coefficient in the computer control condition is not significantly different than zero . There is also a significant difference between the human 6 sec condition , and the computer control condition . See Text S1 for more regression details , and Table S4 and Table S5 for complete regression tables . To investigate the neural underpinnings of these signals we constructed a regression model for the imaging data using regressors suggested by behavioral model ( see Supporting Information for details ) . We limited our investigation of the neural data to the 6 sec human and computer control experiments . Specifically we included a parametric regressor for DJONES at the reveal of the other person's investment , and a parametric regressor for MKT at the time of the revelation of the market return to the subject . Figure 3A shows the activation corresponding to the DJONES regressor in the human condition while 3B shows the activation in the computer control ( both N = 24; both displayed with p< . 001 uncorrected , cluster size > = 5 ) . Note that there were no regions of significant negative correlation . See Figure S1 and Figure S2 for regression tables and glass brains . In the human condition , this activation survived a small volume correction for multiple comparisons over an ROI consisting of 5 mm radius balls centered on bilateral caudate/putamen voxels taken from peak activations in [24] . ( See Figure S3 for mask ) . Additionally , the comparison ( two-sample t-test ) of DJONES across the human and computer conditions survived a similar small volume correction yielding voxels in left caudate ( Figure S4 ) . Activation tables for both small volume corrections are in Figure S5 . While not our main focus , it is worth noting that the MKT regressor also produced , in both human and control conditions , robust activation in the striatum ( Figure S6 ) . Figure S7 shows a conjunction/disjunction analysis of the MKT and DJONES activation at the p< . 001 and p< . 05 levels in the human condition . We were also interested in the possible differences between the neural and behavioral effects of the variables obtained by splitting DJONES into its positive and negative parts ( e . g . POSDJONES = max ( DJONES , 0 ) ; see Text S1 for details ) . We find a significant difference in the behavioral regression coefficients , with the coefficient of the negative part of DJONES being larger in absolute value ( Table S3 ) . Neurally , however , we find no difference between the two conditions ( Figure S8 ) . Finally , we wanted to investigate the relationship between the neural correlates of DJONES and the individual behavioral regression coefficients of DJONES . Figure 4A shows the middle cingulate region for which the individual neural DJONES responses are significantly positively linearly related to the individual behavioral DJONES coefficients ( p< . 05 , FWE whole-brain corrected; behavioral coefficients from individual subject regressions . See Text S1 for details . ) . Figure 4B shows ( for illustrative purposes only ) a plot of the neural coefficients against the ( mean adjusted ) behavioral coefficients .
Using a tandem sequential investment task we show that when subjects play a human partner the inter-personal fictive error guides behavior ( subjects' next bet ) and correlates with a robust neural signature in the striatum . These findings are significant because the partner's choice is revealed after the subject's monetary outcome is revealed and the partner's choice has no bearing on the payoff to the subject . Despite these facts , the inter-personal fictive error still influences the subject's behavior on their next bet , correlates with a robust and parametric neural signature in an important reward processing structure , and depends on whether the partner is a human . Specifically , if humans play a computer partner expressing random investments on each trial this same inter-agent fictive error term has no behavioral impact on the next bet and has no significant neural correlate in the striatum . Our results are for the most part are consonant with the results of previous studies of social influence [15]–[23] that show neural responses to and behavioral influences of the choices of others . However , there are several key differences that allow us to expand on these results . First , the timing of private and social outcome revelations was significantly different in this design . Here , information about the market is revealed first , giving the subject all the information relevant to their payoff , and then the social signal from the partner is revealed . Second , our design is parametric in the choices and outcomes . Our design thus allows us to show that the striatal response and immediate subsequent behavior is fully parametrically influenced by both the market return signal and the interpersonal error signal . Additionally , we see a behavioral asymmetry in the effect of the partner's investment between the outcome where the partner invested more than the subject versus the case where the partner invested less . Subjects adjusted their subsequent investment more when their partner invested less than they did on the previous trial as though they were fleeing their own over-exuberance on that trial . Finally , Burke et al . [17] show that ventral striatum activation to social information covaries with behavioral sensitivity to herd information . We do not see this in our experiment . Rather , we see that neural activation to DJONES in middle cingulate cortex covaries positively with behavioral sensitivity to DJONES . One possible explanation for this correlation is suggested by two studies . Kishida et al . [26] found that athletes showed increased middle cingulate activity when imagining themselves playing their own sport as opposed to a different sport . Further , they saw the same result in subjects when they took a first , as opposed to third person perspective when imagining a sports scene . On the other hand , Chiu et al . [27] found decreased activity during the “self” phase of the trust game in the middle cingulate in autistic subjects . The effect covaried with symptom severity . These results suggest that this area is key for identifying with conspecifics , pointing to a hypothesis that neural sensitivity in middle cingulate to the DJONES signal is dependent on the tendency of a subject to identify with the other investor . This hypothesis is also supported behaviorally by the findings of Burke et al . [17] showing that herding behavior is more pronounced when investing alongside human conspecifics as opposed to non-human primates , as well as by the absence of a DJONES effect in our control condition . Our results suggest that the difference between the partner's investment and the subject's investment can be viewed as an error signal that guides behavior , rather than as simply an add-on affective response . The affective system has long been considered a necessary component effective decision-making [28] whose function can be seen as “ecologically rational” [29] . Neural signals correlated with affect may then be reinterpreted as error signals [30] . For example , much of the early work on anterior insula focused on emotions such as pain and disgust . [31] , [32] Recently , however , the function of the anterior insula has been recast in the language of error signals [29] , whereby activation in the insula is regarded as signaling a variance prediction error . Here our focus is on the striatum , but the idea is similar . Indeed multiple works [17] , [23] , [33]–[35] suggest that socially construed reward signals should appear in the striatum just as other control signals do . In this light , the results of this paper strongly suggest that we view the activation in the striatum not only as a hedonic signal , but also as a control signal . Correlation is a property that is vitally important in asset management: in order to maximize return with a minimum of risk an investment manager must know the correlation of the assets under management [36] . Our ancestors living in small groups were not “asset managers” , but it is likely the members of the group correlated their activities in an optimal way , an activity that would require the brain to track and control individual correlations . Finally these results provide biological evidence that standard theories of investment behavior that are variations on the Markowitz model [37] miss a fundamental driver of behavior by failing to account for the behavior of other investors . The response of the striatum to the Jones variable suggests that tendency to correlate actions is deeply rooted with potential evolutionary drivers . This lends weight to the “behavioral finance” approach espoused by Shiller and others [10] , [38] , [39] . In summary , previous work shows that the comparison of personal results to the results of another modulates neural activity . Our results further show that the comparison of the personal result to the outcome of the other person can be put in the context of an error signal , the interpersonal fictive error , which controls behavior and has a robust neural signature . Social comparison can thus be construed not merely as a possibly unseemly manifestation of envy , but rather as a potentially useful learning signal .
Informed consent was obtained for all research involving human participants , and all clinical investigation was conducted according to the principles expressed in the Declaration of Helsinki . All procedures were approved by the Institutional Review Board of the Baylor College of Medicine , or the Institutional Review Board of Virginia Tech . Experiment 1: 76 participants were recruited and 74 scanned in accordance with a protocol approved by the Baylor College of medicine IRB . In the two behavioral only subjects the log files of the experiment were incomplete , leaving unusable data; in two scanned subjects the experiment terminated prematurely; in 4 other scanned subjects the functional images were unusable , leaving 68 subjects with both behavioral and imaging data . Table S1 summarizes the demographic information of these 68 subjects . All data mentioned in the text and supplementary information referring to the first experiment refers to the behavioral data only of these 68 subjects . Experiments 2 and 3: 49 participants ( 24 for the human condition and 25 for the computer control condition ) were recruited and 49 scanned in accordance with a protocol approved by the Virginia Tech IRB . One subject's scanning session terminated prematurely in the control cohort leaving 24 subjects . All data mentioned in the text and SOM referring to the second experiment refers to these subjects . Table S2 summarizes the demographic information of these subjects . Participants arrived at the lab , were consented , and then read task instructions . In the versions with human partners the partners did not meet . After they were loaded in scanner , the task began . Each subject participated in 10 markets in a random order . There were two groups of markets , A and B ( originally described and used in Lohrenz et al . , 2007 ) . 30 subjects saw group A , and 41 subjects saw group B . After seeing initial market data , a participant selected an investment level ( 0% to 100% in increments of 10% ) using one button box ( shown on a slider bar on the screen ) and submitted the decision using the other button box . In the human partner versions the next market result appeared 750 ms after the later of the two partners' choice was submitted . In the computer partner version the result was displayed 6 seconds later . 2 or 6 seconds later ( depending on the experimental cohort , 1 or 2 , 3 ) the other partner's choice , was displayed by showing two red arrows on either side of the slider bar showing the level person's investment . This was repeated 20 times per market , for a grand total of 200 decisions . Subject's behavioral data were analyzed in R ( package nlme ) [40] , [41] ( see Text S1 for full details ) . | In this study we examine the neural substrates of inter-personal error signals on behavior in an investment task using real historical markets . We show that behaviorally , subjects correlate their investments , despite the fact that another trader has no extra information about how the market may move . These behavioral results are supported by neural data showing large , parametric responses in brain areas related to reward and learning when information about another trader's behavior is revealed , even though this occurs after all useful information about the market has already been shown . These results promise to elucidate some of the subconscious processes that guide people to correlate their behavior in markets and other group environments . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Keeping up with the Joneses: Interpersonal Prediction Errors and the Correlation of Behavior in a Tandem Sequential Choice Task |
Chromatin structure plays an important role in modulating the accessibility of genomic DNA to regulatory proteins in eukaryotic cells . We performed an integrative analysis on dozens of recent datasets generated by deep-sequencing and high-density tiling arrays , and we discovered an array of well-positioned nucleosomes flanking sites occupied by the insulator binding protein CTCF across the human genome . These nucleosomes are highly enriched for the histone variant H2A . Z and 11 histone modifications . The distances between the center positions of the neighboring nucleosomes are largely invariant , and we estimate them to be 185 bp on average . Surprisingly , subsets of nucleosomes that are enriched in different histone modifications vary greatly in the lengths of DNA protected from micrococcal nuclease cleavage ( 106–164 bp ) . The nucleosomes enriched in those histone modifications previously implicated to be correlated with active transcription tend to contain less protected DNA , indicating that these modifications are correlated with greater DNA accessibility . Another striking result obtained from our analysis is that nucleosomes flanking CTCF sites are much better positioned than those downstream of transcription start sites , the only genomic feature previously known to position nucleosomes genome-wide . This nucleosome-positioning phenomenon is not observed for other transcriptional factors for which we had genome-wide binding data . We suggest that binding of CTCF provides an anchor point for positioning nucleosomes , and chromatin remodeling is an important component of CTCF function .
The positioning of nucleosomes along eukaryotic chromatin affects accessibility of the genomic DNA in vivo . Nucleosomes may bind to some genomic regions tightly and prevent transcription factors from approaching their sites . Alternatively , strategically positioned nucleosomes can promote long-range DNA bending and allow distal enhancers to interact with the transcriptional machinery [1]–[3] . Crystal structures show that each nucleosome contains 147 base-pairs ( bp ) of DNA tightly wrapped around an octamer of H2A , H2B , H3 and H4 histone proteins [4] . The linker DNA between two neighboring nucleosomes is ∼20 bp in Saccharomyces cerevisiae [5] and estimated to be 70 bp in higher eukaryotes [6] . Defined lysine and arginine residues in histone tails are often methylated and/or acetylated , which can recruit chromatin remodeling factors and regulate transcription . Histone variants prefer selected genomic regions , e . g . H2A . Z tends to flank nucleosome-free regions [7]–[10] . High resolution maps of nucleosome and H2A . Z have been generated for S . cerevisiae by subjecting chromatin to micrococcal nuclease ( MNase ) and detecting the undigested DNA with tiling arrays [11]–[13] . These studies revealed that RNA polymerase II promoters contain a nucleosome-free region of ∼200 bp upstream of the transcription start site ( TSS ) , flanked by well-positioned nucleosomes on both sides . The same approach was used to map nucleosomes on a subset of human promoters [14] . Recently Zhao and colleagues generated a genome-wide nucleosome map using MNase digestion followed by deep sequencing ( MNase-Seq ) [15] . These studies confirmed the nucleosome-free region upstream of the TSS and several well-positioned nucleosomes downstream of the TSS in humans . In addition , the Zhao lab combined MNase digestion , chromatin immunoprecipitation , and deep sequencing to generate genome-wide maps of H2A . Z and 20 different types of histone methylation in humans [16] . Although the majority of occupied transcription factor binding sites are devoid of nucleosomes in yeast [11] , little is known about how transcription factors and nucleosomes compete for genomic DNA in human cells . We integrated several genome-wide maps of transcription factor binding [16]–[19] and susceptibility of chromatin to DNase I [20] with the aforementioned nucleosome , H2A . Z , and histone modification maps [15] , [16] to study this problem . We found that the insulator binding protein CTCF ( CCCTC-binding factor ) has an unusual ability to position multiple nucleosomes flanking its binding sites genome-wide . CTCF has been extensively studied for its impact on imprinting and X-inactivation [21] . It binds to insulator elements to prevent the spread of heterochromatin and to restrict transcriptional enhancers from activating unintended promoters . In addition , it may function as a transcriptional repressor as well as an activator [22]–[24] . The DNA-binding domain of CTCF contains 11 zinc fingers . One study indicated that only 4 fingers are essential [25] , while others showed that different combination of fingers are used to bind divergent sites [24] , [26] . CTCF is thought to form special chromatin structures or mediate long-range chromosomal interactions in mammalian cells [24] , [27]–[30]; however , the detailed mechanism remains unknown . Our analysis led to several major findings: 1 . CTCF binds in the center of a linker region , flanked by at least 20 well-positioned nucleosomes , symmetrically distributed around the CTCF binding side . We determined the extent that the TSS positions downstream nucleosomes with the same set of data , and were surprised to find that it is much less than that of CTCF . We also examined the genome-wide binding data of STAT1 , NRSF , and p53 , and found these factors to be incapable of positioning nucleosomes . 2 . The nucleosomes flanking a CTCF site are highly enriched in H2A . Z and enriched in 11 histone modifications to various extents . 3 . We determined that on average 150 bp of DNA in these nucleosomes is protected against MNase cleavage , and 35 bp of DNA is cleaved , although both quantities vary greatly among nucleosomes enriched in different histone variants or modifications . The two lengths for the same nucleosome are tightly anti-correlated , consistent with the nucleosome being well positioned . 4 . The nucleosomes enriched in those histone modifications previously associated with active transcription tend to be less protected against MNase , suggesting greater DNA accessibility to the factors that regulate transcription . 5 . CTCF protects roughly 60 bp of DNA and increases the linker between its two neighboring nucleosomes to 118 bp . 6 . Sequence conservation was only observed for the CTCF binding site and not for the other positions in the surrounding 2 kb region , indicating that there is no evolutionary pressure on the genomic DNA sequence that positions the nucleosomes . Furthermore , a previously published algorithm predicts CTCF binding sites to be occupied by nucleosomes . Finally , we performed in vitro nucleosome mapping experiments on two insulator DNA fragments that each contains three CTCF sites . We found these CTCF sites to be located between MNase cleavage sites and hence are likely to be occluded by nucleosomes in the absence of CTCF . Thus we suggest that the binding of CTCF provides an anchor for positioning neighboring nucleosomes and this may be important for CTCF function .
We developed a method to perform aggregation analysis of genome-wide mapping data , called Genomic Signal Aggregator ( GSA; see Methods and Figure S1 ) . GSA computes distribution of hybridization score obtained in a tiling array experiment or coverage of sequence tags obtained in a deep sequencing experiment , plotted as a function of the distance to a set of anchors . We applied GSA to the deep sequencing data on mononucleosome mapping [15] , with occupied CTCF binding sites ( defined in Methods ) as anchors . The average coverage of sequence tags for the DNA ends of mononucleosomes is shown in Figure 1B . We separately mapped the tags to the plus and minus strands ( defined according to the strand of the anchor CTCF site ) , resulting in the blue and orange curves in Figure 1B ( see Figure S2 for technical explanation on why there are two peaks per nucleosome ) . The CTCF site was observed in the center of a linker region , flanked on each side by up to 10 pairs of peaks with ∼185 bp intervals , indicating 20 well-positioned nucleosomes ( Figure 1A ) . As a negative control , we aggregated the same data [15] but using unoccupied CTCF sites ( defined in Methods ) as anchors and produced two curves which peaked at +69 and −25 bp respectively , suggesting that unoccupied CTCF sites are often occupied by a nucleosome . Because the distance between these two peaks is smaller than the size of a nucleosome , we suspect that the nucleosome is positioned in slightly different positions across the CTCF sites . In order to simulate the effect of sequencing depth , we generated two aggregation graphs with 20% and 5% of randomly sampled sequence reads from the original 154 . 6 M reads ( Figure S3 ) . Contrasting Figure S3 with Figure 1B indicates that greater sequencing depth leads to linearly taller aggregation graphs , such that the ratio of the sequence coverage around the occupied sites over the coverage around the unoccupied sites is largely independent of the sequencing depth . This ratio corresponds to how much more likely that a position , at a particular distance ( <2 kb ) away from an occupied CTCF site , is the end position of a mononucleosome , over the position anchored on an unoccupied CTCF site . In addition , greater sequencing depth leads to smoother graphs overall and in particular for the peaks that are far away from the occupied CTCF sites . As a result , more well-positioned nucleosomes are discernable with greater sequencing depth . In Figure 1B , the peaks that are more distal from the anchoring CTCF site are broader and lower . To determine whether this indicates lower nucleosome occupancy in the distal regions , we integrated the sequence coverage over non-overlapping 185-bp intervals and computed the ratio between the resulting values for each interval anchored on occupied CTCF sites over the values for the same interval anchored on unoccupied CTCF sites . This resulted in a largely flat distribution with an average ratio of 1 . 07 ( Figure S4 ) , indicating that the nucleosome occupancy does not decrease appreciably over a 2 kb distance from an occupied CTCF site . Figure S4 further indicates that the nucleosome occupancy in the 4-kb region around occupied CTCF sites is 7% higher than that of the 4-kb region around unoccupied CTCF sites . The diminished and widened peaks distal from the CTCF anchors are likely due to the more distal nucleosomes being less well positioned across the cell population , and/or are positioned at more varying locations from the CTCF site among different CTCF-bound loci . The graphs for individual loci do not exhibit the diminishing behavior ( not shown ) , suggesting that the distal nucleosomes are well-positioned across the cell population , but at more varying locations from the CTCF anchors than the proximal nucleosomes . All of the experimental datasets were generated on the entire cell population , while CTCF could occupy its sites in a subpopulation of the cells . We generated aggregation graphs anchored on two subsets of occupied CTCF sites: sites that are 2–5 kb away from the nearest occupied sites and sites that are more than 500 kb away from the nearest occupied sites , hypothesizing that sites in the former set are more clustered and hence likely to be occupied in a greater portion of cells than sites in the latter set . Indeed a stronger signal was observed for the graphs anchored on the more clustered CTCF sites ( Figure S5 ) . Nonetheless , the difference between the two sets of graphs is small , suggesting that our findings are unlikely affected by the subpopulation issue . We then applied GSA to all 20 ChIP-Seq datasets , each on mononucleosomes enriched in a type of histone modification [16] . With the exception of H3K9me3 , the other 19 datasets show similarly dramatic oscillation ( Figure 2 ) . The H3K9me3 data may be of poor quality because it is not enriched around any of the anchor sets tested in this study ( see the section after next for enrichment analysis ) . For 10 datasets ( e . g . , unmodified nucleosomes , or nucleosomes with modified H3K36 , H3K27 , or H3R2 ) , at least 10 blue and 10 orange peaks can be identified , supporting 10 well-positioned nucleosomes flanking the center CTCF site . The other datasets reveal 6–12 nucleosomes . The positions of these nucleosomes are in complete agreement with each other and with those seen for the mononucleosome mapping data ( Figure 1B ) . To compare the extent of nucleosome positioning by CTCF with that around the TSS , we applied GSA to the same mononucleosome mapping data [15] and the histone modification data [16] with the TSSs of actively transcribed genes as anchors . In agreement with previous findings [14] , [15] , there is a 200-bp-long nucleosome-free region around the TSS ( indicated by a pronounced dip in the curves ) and the +1 nucleosome is well-positioned , centered at ∼120 bp downstream of the TSS; in addition , two nucleosomes upstream of the TSS and four more nucleosomes downstream of the TSS are discernable ( Figure S6A ) . Among the histone modification datasets , the H3K4me3 dataset produced the strongest nucleosome-positioning signals , followed by H3K4me2 . By combining H3K4me2 and H3K4me3 data , one can make out 5 positioned nucleosomes downstream of the TSS , with the first two apparent in the H3K4me3 curves and the last four discernable in the H3K4me2 curves ( Figure S6B ) . Using TSS as anchors , the GSA curves of H3K27me1 , H3K4me1 and H3K9me1 also show oscillatory behavior; however , the peaks are poorly formed , similar to those of H3K4me2 ( Figure S6C ) . The GSA curves of other histone modification data do not show oscillatory behavior ( figures not shown ) . The distance between the centers of neighboring nucleosomes measured in the TSS-anchored graphs agrees with that in CTCF-centered graphs . Collectively , these results indicate that there are 2 and 5 positioned nucleosomes upstream and downstream of TSS , respectively; however , the sharp contrast between Figure S6 with Figures 1B and 2 indicates that the positions of the nucleosomes around the TSS vary among different loci to a much greater extent than the positions of the nucleosomes flanking occupied CTCF sites . We also investigated whether there were well-positioned nucleosomes flanking the binding sites of other transcription factors . The genome-wide maps of a number of transcription factors in living human cells have been produced using ChIP-chip or ChIP-Seq . Among them , the binding regions of STAT1 , NRSF and p53 are highly enriched in the cognate motifs of these factors , which allowed us to determine the occupied sites by scanning the ChIP-chip or ChIP-Seq target regions with the motif matrices . We used these three sets of sites as anchors to produce aggregation plots with the mononucleosome mapping dataset [15] ( Figure S7 ) and the histone modification datasets [16] ( figures not shown ) . None of the graphs in Figure S7A/B/C show oscillatory behavior as in Figure 1B ( the four graphs are drawn in the same scale ) , suggesting that these transcription factors do not possess the ability to position nucleosomes . The STAT1 and NRSF graphs indicate that the binding sites of these two factors have higher nucleosome occupancy than neighboring genomic positions , suggesting that their functions may be regulated by nucleosome positioning . We wanted to investigate whether some of the nucleosomes surrounding the occupied CTCF sites were enriched in H2A . Z or any of the histone modifications , in comparison with the nucleosome surrounding the unoccupied CTCF sites . We applied GSA to the ChIP-Seq dataset of H2A . Z with the occupied sites or the unoccupied CTCF sites as anchors , respectively , and obtained two sets of curves as shown in Figure 3 . The green and purple curves are completely flat , again suggesting that nucleosomes are not well-positioned around unoccupied CTCF sites . Moreover , Figure 3 indicates the histones that flank the occupied CTCF sites are highly enriched in H2A . Z , especially the −1 and +1 nucleosomes , although the enrichment can be seen across a +/−2 kb region in Figure 3 ( the blue and orange curves are cleanly above the green and purple curves ) . In order to account for the different sequencing depths among the datasets and the difference in nucleosome occupancy around occupied and unoccupied CTCF sites , we defined a histone variant or modification to be enriched if the ratio between the area under the curves anchored on occupied CTCF sites over the area under the curves anchored on unoccupied sites is higher than the ratio for mononucleosome mapping ( 1 . 07 as determined in the first section of Results ) , for the +/−2 kb region . By this criterion , subsets of nucleosomes flanking occupied CTCF sites are found to be enriched in H2A . Z and the following 11 histone modifications ( in descending order of enrichment ) : H3K4me3 , H3K4me2 , H3K4me1 , H3K9me1 , H4K20me1 , H3R2me1 , H3K27me1 , H3K36me1 , H2BK5me1 , H3R2me2 , and H3K79me1 . The other 9 histone modifications ( H3K27me3 , H3K79me2 , H4R3me2 , H3K36me3 , H3K79me3 , H3K27me2 , H4K20me3 , H3K9me2 , and H3K9me3 ) are not enriched; nonetheless , most of them exhibit strong oscillatory patterns , with the best examples being H3K27me3 and H3K36me3 ( Figure 2B ) . We applied the same criterion on consecutive 185 bp windows to determine whether individual nucleosomes are enriched in H2A . Z or the histone modifications . The resulting heatmap ( Figure 4 ) reveals a large variation in how far different levels of enrichment spread from the CTCF anchors . The aggregation graphs for more than half of the histone modifications in Figure 2 contain two extra prominent center peaks , which we suspect correspond to the 5′ and 3′ boundaries of the CTCF footprint . Take the H3K36me1 dataset as an example , because ChIP was performed with an antibody against H3K36me1 and not with an antibody against CTCF , we suggest that the blue peak resulted from the lack of digestion of the linker between the CTCF site and the +1 nucleosome; similarly , the orange peak resulted from the lack of digestion of the linker between the CTCF site and the −1 nucleosome . These two peaks coincide in position exactly with the only two peaks in the aggregation plot of the ChIP-Seq data of CTCF [16] with occupied CTCF sites as anchors ( Figure 1C ) , the distance between which was determined to be 64 bp ( see Methods ) . Note that sonication and not MNase digestion was used to generate the ChIP-Seq data of CTCF , thus there are no nucleosome peaks in Figure 1C . We do not observe the two center peaks for the mononucleosome mapping data ( Figure 1B ) , nor for the H2A . Z data ( Figure 3 ) . Also , the occurrence of these peaks does not correlate with whether the nucleosomes are enriched in the particular histone modification . Thus , it is unclear whether the occurrence of these peaks merely reflects the experimental condition of the MNase digestion , or has biological significance . Recently a genome-wide DNase I hypersensitivity map was produced on human CD4+ T cells with the DNase-Seq technology [20] . We applied GSA to this data with occupied CTCF sites as anchors , and the resulting pattern ( Figure 1D ) indicates that CTCF protects 30 bp of genomic DNA on the minus strand ( the distance between the two inner orange peaks ) and 42 bp on the plus strand ( the distance between the two inner blue peaks ) against DNase I digestion . The asymmetry between the protection lengths of the two strands is intriguing . We suggest that this is due to the closer contact of CTCF with the plus strand than with the minus strand . No crystal structure of a CTCF-DNA complex is available . Thus , we used the crystal structure of a six-zinc-finger protein with its cognate DNA [31] to model a CTCF-DNA complex . We computed the solvent accessible surface area of the two strands of the DNA in the crystal structure and determined the average area of one strand to be 23% higher than that of the other strand . Graphical display of the crystal structure indicates that the zinc-finger protein binds to the major groove of the DNA and that because the major groove has greater volume than the protein , the protein makes closer contact with one DNA strand than with the other stand . Lobanenkov and colleagues discovered that CTCF combined different subsets of its 11 zinc fingers when binding to divergent sites [24]–[26] . Such complexity can also result in asymmetric CTCF-DNA interaction . The fine structure of the footprint shown in Figure 1D suggests that the non-zinc-finger portion of CTCF also contacts DNA and protects it from DNase I cleavage in a characteristic way . The two small peaks centered at −204 bp and +260 bp correspond to the linker between the −1 and −2 nucleosomes and the linker between the +1 and +2 nucleosomes , respectively . In order to test whether there is any evolutionary pressure on the primary sequences surrounding occupied CTCF sites , we obtained the phastCons scores for the sequences from the UCSC genome browser ( http://genome . ucsc . edu ) and plotted the average score at each position of the 4-kb window centered on occupied CTCF sites . In comparison , we also obtained the conservation for positions surrounding unoccupied CTCF sites . The two curves are shown in Figure 1E . It is apparent that conservation is only restricted to the center 15 bp of the CTCF binding motif with the highest information content ( positions underlined in Figure S8 ) . The positions 22–24 bp away from either side of the CTCF binding motif are even less conserved than the background , suggesting that these positions are not recognized by the CTCF in a sequence-specific manner . Because the lack of sequence conservation at the mononucleotide level does not preclude these sequences from possessing intrinsic nucleosome-positioning ability , we applied a previously published computer algorithm that combines dinucleotide periodicity and a thermodynamic model to predict sterically allowed nucleosome placement [32] . We used the flavor of this algorithm trained with human data ( see Methods ) . The algorithm predicts a nucleosome to occupy the CTCF sites that are occupied by CTCF in vivo ( Figure S9 ) , which corresponds to the linker region according to experimental data ( orange and blue curves in Figure 1B and Figure 2 ) . Thus , the algorithm predicts a nucleosome-positioning pattern that disagrees with the pattern experimentally measured around occupied CTCF sites . To further test whether sequences that flank CTCF binding sites possess intrinsic nucleosome positioning signals , we reconstituted nucleosomes onto two different DNA fragments ( Insulators 23 and 44 ) that each harbor three CTCF binding sites and were shown to function as insulators by enhancer blocking assays [33] . As a positive control , nucleosomes were also assembled onto a DNA fragment that contains 10 head-to-tail repeats of a 5S rDNA nucleosome positioning sequence ( CP924 ) . Nucleosomal arrays were digested with increasing amounts of micrococcal nuclease , and nucleosome positions were mapped by indirect end-labeling and Southern blot analysis . As shown in Figure 5 , MNase analysis yielded a typical repeating pattern of cleavages and protections on the 5S repeat DNA , indicative of a positioned nucleosomal array . In contrast , nucleosomes assembled onto the DNA fragments that contain CTCF binding sites showed a much less regular pattern of MNase cleavages which is not consistent with a positioned nucleosomal array . Strikingly , each of the CTCF binding sites is located between MNase cleavages , indicating that these sites are bound by nucleosomes in vitro . Thus , these biochemical studies are in agreement with the predictions of the aforementioned computational algorithm ( Figure S9 ) , and they do not agree with the pattern of nucleosome positioning observed in vivo ( Figure 1B and Figure 2 ) . The well-positioned nucleosomes around CTCF sites provided an unprecedented opportunity to measure the distance between neighboring nucleosomes , the linker length , and the length of nucleosomal DNA protected against MNase cleavage . The distances between the +1 and −1 nucleosomes are increased due to the CTCF sites . We developed an algorithm to automatically determine the locations of the peaks in each GSA curve ( see Methods ) . These peaks mark the boundaries of the well-positioned nucleosomes ( Figures 1A/B and Figure S2 ) . In order to relate the peak positions to the three aforementioned quantities , we defined six distances: L-CTCF , L-Center , L-Digest , Unit+ , Unit− and L-Protect as illustrated in Figure 6A . Figure 6B plots these six distances for mononucleosome data , the 20 histone modifications and H2A . Z . For each dataset , the first two distances can be measured only once based on the two center nucleosomes while the last four distances can be measured multiple times for all but the H3K9me3 dataset and the standard deviations of these measurements are shown as error bars in Figure 6B . There is no statistically significant difference between distances measured on nucleosomes upstream and downstream of CTCF binding sites , or those between proximal and distal nucleosomes ( data not shown ) . It is apparent from Figure 6B that Unit+ and Unit− are largely invariant across the histone variants and histone modifications , consistent with the stable positions of the nucleosomes around the CTCF site . The length of nucleosomal DNA that are protected against MNase cleavage ( L-Protect ) ranges from 106 to 164 bp ( 139±15 bp ) among the sets of nucleosomes enriched in different histone modifications or H2A . Z . The L-Protect determined with the mononucleosome mapping data [34] is 150 bp . The length of genomic DNA between two neighboring nucleosomes that is digested away by MNase ( L-Digest ) also varies greatly , from 10 to 83 bp ( 43±18 bp ) , with 35 bp for mononucleosome mapping data . L-Protect and L-Digest are strongly anti-correlated ( R2 = 0 . 82 and P-value = 3e-8; Figure 6C ) , consistent with Unit+ and Unit− being largely invariant . These results may indicate that different histone modifications cause the ends of the nucleosomal DNA to wrap around the histone core with greatly varying extents of tightness . Tighter wrapping leads to lesser extent of MNase digestion and vice versa , without translational movement of the nucleosomes . Alternatively , different histone modifications may be associated with chromatin regions depleted in histone H2A/H2B dimers . Loss of dimers can be catalyzed by ATP-dependent remodeling enzymes that can be targeted to nucleosomes by histone modifications [35] . Loss of one H2A/H2B dimer will release ∼30 bp of nucleosomal DNA , and loss of both dimers yields an H3/H4 tetrasome particle that protects only ∼90 bp of DNA from MNase digestion [4] . Figure 6B indicates that H3K79me1 and H4R3me2 are the two histone modifications that lead to the best protection of nucleosomal DNA ends , while H3K9me2 and H4K20me1 are the two modifications that lead to the least protection . All 20 histone modifications are methylations , which do not change the net charges of the histones , thus the large variation among them is surprising . Moreover , different numbers of methyl groups on the same amino acid differ as much as the modifications on different amino acids . The histone modifications that correlate positively with gene expression level ( H3K27me1 , H3K9me1 , H3K4me2 , H3K4me1 , H4K20me1 , and to some extent H3K9me2 [16] , [36] ) correspond to the least protection against MNase , suggesting that increasing accessibility of the ends of nucleosomal DNA may be a mechanism for transcriptional activation . Nucleosomes enriched in H2A . Z have short L-Protect and long L-Digest , consistent with the effect of H2A . Z on transcriptional activation [10] , [29] . The footprint of CTCF ( L-CTCF ) ranges from 32 to 64 bp among the datasets of different histone modifications ( Figure 6B ) . With 11 fingers , CTCF can theoretically form direct atomic contact with 33 bp of DNA . The binding motif of CTCF has 15 positions with high information content ( Figure S8 ) , indicating that roughly five fingers contribute significantly to the binding specificity . Nonetheless , our results indicate that at least 32 bp of DNA is protected from MNase cleavage . L-CTCF strongly anti-correlates with the length of the MNase digested DNA between the −1 and +1 nucleosomes ( L-Center , 118 bp for mononucleosome data ) , with R2 = 0 . 53 ( P-value = 0 . 0002; Figure 6C ) . This suggests that when the DNA of these two nucleosomes is more accessible to MNase , the enzyme can cut closer to the CTCF site . Along this line of reasoning , we would also expect to see stronger MNase cleavage signal around the CTCF footprint for the histone modifications with greater L-Center . Because the nucleosomes flanking the CTCF site are enriched in different histone modifications to varying extents , we define the ratio between the average height of the CTCF footprint peaks and the average height of the +1 and −1 nucleosome peaks in Figure 2 as footprint-peak ratio ( FP-ratio ) . Indeed , FP-ratio anti-correlates significantly with L-Center ( R2 = 0 . 39 and P-value = 0 . 004; Figure 6E ) .
By integrating a large number of high-throughput sequencing and microarray datasets and performing aggregation analysis with transcription factor binding sites or the TSS as anchors , we discovered that there is an array of 20 nucleosomes flanking occupied CTCF sites genome-wide . These nucleosomes are so well positioned that remarkable oscillatory patterns were observed for 21 out of the 22 genome-wide datasets [16] , [34] . Two case studies reported CTCF binding in the IGF2/H19 and DM1 loci , both of which suggested that the CTCF binding sites occurred in linker regions between nucleosomes [37] , [38] . These are consistent with our findings in this study . We are unaware of other previous work on the genome-wide relationship between CTCF and nucleosomes . The TSS is the only genome-wide anchor for which well-positioned nucleosomes were reported , and only two nucleosomes upstream the TSS and five nucleosomes downstream of the TSS are well positioned . Here we show that 20 nucleosomes flanking CTCF sites exhibit much stronger oscillatory patterns , and hence are much better positioned than the nucleosomes around the TSS . No well-positioned nucleosomes have ever been reported to flank transcription factor binding sites . Among the human transcription factors for which genome-wide binding data are available , the ChIP target regions for only three factors are highly enriched in their binding motifs ( STAT1 , NRSF and p53 ) . We did not observe well-positioned nucleosomes around the occupied sites of any of these three factors . One complication is that the ChIP-Seq data of histone modifications were on CD4+ T cells while the binding site data were on other cell types . We must wait for future data to resolve this issue definitively , and to uncover whether well-positioned nucleosomes flank the binding sites of other transcriptional factors genome-wide . There are four possible mechanisms for the well-positioned nucleosomes around CTCF sites: 1 . CTCF binds to its sites first and then recruits chromatin remodeling factors to position neighboring nucleosomes; 2 . CTCF binds to its sites first , which provides a strong anchor for the neighboring nucleosomes to line up by themselves; 3 . Nucleosomes are well positioned in some regions of the genome due to DNA sequence features , and a CTCF site has co-evolved with the nucleosome positioning sequence features to exist in a lengthened linker region , which attracts the binding of CTCF; 4 . Some genomic regions contain nucleosome-positioning sequence features leading to an array of regularly positioned nucleosomes which occlude a CTCF site , and CTCF binds to its site and repositions the nucleosomes to create a lengthened linker region . We argue that our results mostly support the second scenario for reasons as follows . Three lines of evidence suggest that the well positioned nucleosomes are unlikely caused predominantly by the intrinsic sequence features of the genomic DNA surrounding occupied CTCF sites: 1 . There is a lack of conservation for the sequences that flank occupied CTCF sites , in sharp contrast with the strong conservation at the CTCF sites ( Figure 1E ) ; 2 . A computational algorithm predicts a nucleosome to occupy sites that are occupied by CTCF in vivo ( Figure S9 ) ; 3 . We performed in vitro nucleosome reconstitution and mapping experiments on two insulators that each contains three CTCF sites . The results showed an irregular pattern of MNase cleavages , indicating the lack of a positioned nucleosomal array . Moreover , all six CTCF sites in these two insulators are in nucleosomal regions , consistent with the computational prediction and in contrast with the in vivo data . The binding of CTCF lengthens the linker region to 118 bp . Thus if the 20 nucleosomes form with regular intervals before CTCF binds , all of them will need to slide apart to accommodate the binding of CTCF , which seems unlikely . CTCF has not been reported to recruit chromatin remodeling factors . Thus we propose that the second scenario is most likely to be biologically relevant in general . Our hypothesis is consistent with the statistical positioning mechanism , which states that nucleosomes prefer not to occupy some regions of the genome due to sequence features such as homo-poly A/T or the eviction by regulatory proteins , but are well-positioned in the remaining regions of the genome due to structural constraints imposed by DNA packaging [39] . We hypothesize that the binding of CTCF acts as a roadblock for translational nucleosome movements and as a result the nucleosomes are packaged between the CTCF binding sites and the nearest nucleosome-free regions . Nonetheless , our hypothesis does not preclude the possibility that in some loci other mechanisms cause well-positioned nucleosomes around CTCF sites . Indeed , Kanduri et al . reported that a subset of CTCF sites in the H19 locus was flanked by nucleosome positioning sequences and the authors argue that these sequences have evolved to ensure the constitutive availability of the CTCF binding sites [38] . Thus these results argue for the third scenario described above . The nucleosomes flanking CTCF sites are enriched in H2A . Z and 11 histone modifications . Among these , H2A . Z and 8 histone modifications are also enriched in promoters and are positively correlated with the transcriptional levels of downstream genes [16] . The remaining three , H3K79me1 , H3R2me1 and H3R2me2 , are enriched to much less extents among the 11 modifications ( Figure 4 ) . The large overlap between the epigenetic features of nucleosomes in promoters and the nucleosomes around CTCF sites is surprising , given that CTCF is mostly known to bind to insulators , suggesting that CTCF may play an important role in regulating promoters . The well-positioned nucleosomes around occupied CTCF sites allowed us to determine the length of the nucleosomal DNA protected against MNase digestion . Our results ( Figure 6 ) indicate that there is great variation in the accessibility of nucleosomal DNA that corresponds to various histone methylations . It would be interesting to quantify the amounts of variation for modifications that affect net charges of the histones , once the data becomes available . The histone modifications that correspond to greater DNA accessibilities and H2A . Z , which also corresponds to great DNA accessibility , are highly enriched in promoters of expressed genes . Collectively , these results suggest that one of the mechanisms by which histone modifications regulate gene expression can be by modulating accessibility to the genomic DNA . In light of the recent findings on histone turnover [40] , [41] , it is tempting to suggest that accessible DNA would facilitate rapid histone turnover and/or rapid turnover results in accessible DNA . In particular , rapid histone turnover was observed in chromatin boundaries and suggested to help delimit the spread of chromosome states [40] , [41] . Because the primary function of CTCF is to bind to insulators , which are the most well understood boundary elements , we suggest that those CTCF sites flanked by nucleosomes with highly accessible DNA can prevent the lateral spreading of chromosome states . Figure 6 also suggests that regions around occupied CTCF sites are of heterogeneous composition: subsets of them are enriched in different histone modifications , therefore producing different L-Digest measurements . Indeed , hierarchical clustering of the regions surrounding all occupied CTCF sites based on the ChIP-Seq signal levels of histone modification , H2A . Z and RNA polymerase II ( Figure S10 ) confirms that these genomic regions have diverse patterns of epigenetic marks . It would be interesting to investigate whether some of these patterns are correlated with the insulator function , and if so , which ones are . CTCF has also been reported to possess activating and repressing functions and it is possible that some epigenetic patterns correspond to these functions . Figure S10 further indicates that all the nucleosomes surrounding occupied CTCF sites are covered by H2A . Z and/or some of the histone modifications investigated in this study . Because well-positioned nucleosomes are observed for all but one histone modification datasets ( Figure 2 ) , we conclude that this is a universal feature of CTCF , regardless of the underlying biological function ( insulation , activation , repression or others ) of the particular locus . Because Unit+ and Unit− are on average 185 bp and largely invariant , we can deduce that the length of human linker DNA is 38 bp given that 147 bp of DNA is observed in the crystal structure of nucleosomes [4] . This linker length is somewhat shorter than the previous estimate of 70 bp in higher eukaryotes [6] . Because our analysis included data on all nucleosomes , nucleosomes with H2A . Z or one of 20 histone modifications , we believe that 38 bp is a robust estimate . Furthermore , this is unlikely to be specific to only the nucleosomes flanking occupied CTCF sites , because the well-positioned nucleosomes around the TSS have similar intervals as the nucleosomes flanking CTCF sites . In summary , we discovered that occupied CTCF binding sites in the human genome are flanked by 20 well-positioned nucleosomes . These nucleosomes are enriched in H2A . Z and 11 histone modifications , forming complex epigenetic patterns . Nucleosomes enriched in different histone modifications have diverse but compensating lengths of DNA that are protected from or digested by MNase . The binding of CTCF extends the linker to 118 bp and the CTCF footprint is smaller if the DNA of neighboring nucleosomes is more accessible . These results provide insights to the interplay between chromatin structure and CTCF function .
The genomic coordinates of mapped sequence tags for the mononucleosome mapping datasets [34] and for ChIP-Seq datasets [16] were kindly provided to us by Schones and Zhao . As in those two studies , only those sequenced reads that map uniquely to the human genome ( hg18 ) were used for all analyses in our study . Kim et al . annotated CTCF binding sites based on ChIP-chip data with an antibody against CTCF on IMR90 cells [17] . CTCF binding has been reported to be largely ubiquitous across multiple cell types [17] , [33] . Thus we took the subset of the CTCF sites annotated by Kim et al . within ChIP-chip target regions that are also overlap with CTCF ChIP-Seq data in CD4+ T cells [16] , and defined them as the occupied CTCF sites , used throughout this study . A subset of CTCF sites annotated by Kim et al . using the CTCF binding matrix to scan genomic regions [17] , which are outside the target regions of both the aforementioned ChIP-chip and ChIP-Seq datasets but within the scopes of these two experiments , were defined as unoccupied sites . A small number of these sites may be occupied by CTCF in cell types that have not been studied , but this does not affect our conclusions , because the vast majority of data on which we based our analysis was generated on CD4+ T cells . In total , we define 6432 occupied sites and equal number of unoccupied sites and the lists are available as Table S1 . Genome-wide DNase-Seq data was obtained from [20] . Transcription start sites for known genes were downloaded from the UCSC genome browser ( http://genome . ucsc . edu ) , and partitioned according to expression levels measured by Su et al . [42] . Occupied STAT1 binding sites were determined by scanning STAT1 ChIP-Seq target regions with TRANSFAC matrix M00223 [43] . Occupied NRSF binding sites were annotated by Johnson et al . [19] . Occupied p53 binding sites were annotated by applying p53-PET model in ChIP-PET ( Paired End di-Tag ) sequences with at least 3 tags [18] . Many types of genomic data including ChIP-Seq , ChIP-chip , MNase-Seq or DNase-Seq can be represented as a set of genomic positions , each associated with a score . For example , the ChIP-chip raw data is constituted of a set of short oligonucleotide probes each associated with a hybridization intensity score . These probes are mapped onto the genome , which assigns their scores to all the corresponding genomic positions . ChIP-Seq , MNase-Seq and DNase-Seq yield sequence reads . After all the reads are mapped to the genome , each genomic position can be assigned a score which corresponds to how many reads that cover the position . One type of highly informative analysis to perform on such genomic data is to aggregate over a set of genomic anchors at base-pair resolution . The analysis yields a continuous curve , with the average score for the genomic position at a particular distance away from the anchor plotted against the distance . In this study , we use two types of anchors , transcriptional start sites or the 5′-ends of transcription factor binding sites . We developed the genomic signal aggregation ( GSA ) algorithm for performing aggregation analysis ( Figure S1 ) . Specifically , each genomic position is assigned to the nearest anchor and classified as upstream , on-anchor or downstream , and the scores for all the positions at a specified distance from the anchor are averaged . In order to account for the scenario that a position at a certain distance to the anchor may correspond to more genomic locations than another position , due to the uneven distribution of the anchors in the genome , the sums of scores are divided by the numbers of contributing genomic locations instead of by the number of anchors . An earlier version of this algorithm was applied to a large number of ChIP-chip data generated by the ENCODE consortium [44] . We have built a freely accessible web server for GSA ( http://zlab . bu . edu/GSA ) . ChIP-Seq , MNase-Seq or DNase-Seq datasets are composed of sequence reads , which could map to either the same or the opposite strand of the nearest anchor . We separately performed GSA on the reads that map to the two strands , as indicated in Figure S2 , in order to reveal fine details of the data . This leads to the two sets of curves throughout this study ( e . g . , the orange and blue curves in Figure 1 ) . We assigned the count of a sequence tag to all the positions of the sequence reads ( 24 bp or longer ) instead of only to the 5′-end position , in order to smooth the curves . As a result , the peaks of the curves are not at the boundaries of the nucleosome , but are 12 bp inside the boundaries as illustrated in Figure S2 and all the GSA curves throughout this paper . Figure 6 defines distances between the exact nucleosome boundaries by factoring in the 12 bp . For the DNase-Seq data , we only used the 5′-end position for aggregation because there was sufficient number of tags ( 12 M ) , thus there is no such 12-bp shift . The y-axis of the GSA curve for a ChIP-Seq , MNase-Seq or DNase-Seq dataset indicates the average number of sequence reads that are mapped to a particular distance to an anchor , with the average taken over all anchors . Thus the height of a GSA curve depends on the sequencing depth . For example , 154 . 6 M reads were used to generate Figure 1B ( all mono-nucleosomes ) and 10 . 1 M reads were used to generate the H3K4me3 panel in Figure 2A , which directly accounts for the difference in the y-axis spans of these two figures . Figure S3 includes the remake of Figure 1B with 20% or 5% randomly sampled reads , and it is apparent that the heights of the curves in Figure S3 decrease proportionally when compared with those in Figure 1B . See the second paragraph of Results for discussion on how to compare GSA curves . The result of GSA analysis on nucleosome positioning is affected by the experimental method used to fragment the chromatin samples . For most of the results discussed in this article , we used the datasets prepared with MNase digestion: the all mononucleosome mapping dataset with 154 . 6 M sequence reads [15] , and the 20 histone modification ChIP-Seq datasets in [16] . To contrast MNase digestion with sonication , we also produced the GSA plots on another dataset in [15] , generated with ChIP of H3 followed by sonication to 200–300 bp long DNA fragments ( Figure S11 ) . Only 12 well-positioned nucleosomes are discernable around occupied CTCF sites ( Figure S11A ) , in contrast with 20 nucleosomes seen with the MNase digestion dataset ( Figure 1B ) . Consistent with the finding reported previously [15] , only 3 well-positioned nucleosomes are discernable around the TSSs of expressed genes ( Figure S11B ) , in contrast with 7 nucleosomes seen with the MNase digestion dataset ( Figure S6A ) . We wrote a PERL program for calling peaks in GSA curves . It searches for the local maxima compared with their flanking intervals , which were set to 70 bp for detecting nucleosome peaks and 15 bp for detecting CTCF footprints . Each inter-nucleosome distance was measured as the mean of the distance between neighboring plus-strand peaks and the distance between the corresponding minus-strand peaks . The counts of well-positioned nucleosomes start at the center of CTCF binding sites and continue in both directions until the variation between inter-nucleosome distances exceeds 40 bp . Positions of peaks and counts of well-positioned nucleosomes were visually inspected and minor defects due to imperfectly formed peaks were corrected . The 3 kb sequences flanking CTCF binding sites were downloaded from the UCSC genome browser ( hg18 ) . The sequences were fed to the program by Segal et al . [32] for predicting nucleosome occupation probability , with the human nucleosome model ( both downloaded from http://genie . weizmann . ac . il/pubs/nucleosomes06/ ) . The predicted per-base-pair probability values were sampled every 50 bp and then aggregated using GSA , with the occupied and unoccupied CTCF binding sites as anchors , separately . | The accessibility of genomic DNA to regulatory proteins and to the transcriptional machinery plays an important role in eukaryotic transcription regulation . Some regulatory proteins alter chromatin structures by evicting histones in selected loci . Nonetheless , no regulatory proteins have been reported to position nucleosomes genome-wide . The only genomic landmark that has been associated with well-positioned nucleosomes is the transcriptional start site ( TSS ) —several well-positioned nucleosomes are observed downstream of TSS genome-wide . Here we report that the CCCTC-binding factor ( CTCF ) , a protein that binds insulator elements to prevent the spreading of heterochromatin and restricting transcriptional enhancers from activating unrelated promoters , possesses greater ability to position nucleosomes across the human genome than does the TSS . These well-positioned nucleosomes are highly enriched in a histone variant H2A . Z and 11 histone modifications . The nucleosomes enriched in the histone modifications previously implicated to correlate with active transcription tend to have less protected DNA against digestion by micrococcal nuclease , or greater DNA accessibility . This nucleosome-positioning ability is likely unique to CTCF , because it was not found in the other transcriptional factors we investigated . Thus we suggest that the binding of CTCF provides an anchor for positioning nucleosomes , and chromatin remodeling is an important aspect of CTCF function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/chromosome",
"biology",
"computational",
"biology/transcriptional",
"regulation"
] | 2008 | The Insulator Binding Protein CTCF Positions 20 Nucleosomes around Its Binding Sites across the Human Genome |
An important goal of systems medicine is to study disease in the context of genetic and environmental perturbations to the human interactome network . For diseases with both genetic and infectious contributors , a key postulate is that similar perturbations of the human interactome by either disease mutations or pathogens can have similar disease consequences . This postulate has so far only been tested for a few viral species at the level of whole proteins . Here , we expand the scope of viral species examined , and test this postulate more rigorously at the higher resolution of protein domains . Focusing on diseases with both genetic and viral contributors , we found significant convergent perturbation of the human domain-resolved interactome by endogenous genetic mutations and exogenous viral proteins inducing similar disease phenotypes . Pan-cancer , pan-oncovirus analysis further revealed that domains of human oncoproteins either physically targeted or structurally mimicked by oncoviruses are enriched for cancer driver rather than passenger mutations , suggesting convergent targeting of cancer driver pathways by diverse oncoviruses . Our study provides a framework for high-resolution , network-based comparison of various disease factors , both genetic and environmental , in terms of their impacts on the human interactome .
Cellular function and behaviour are driven by highly coordinated biomolecular interaction networks . A prime example is the protein-protein interaction ( PPI ) network , also known as the protein “interactome” or interactome for short . A central focus of disease systems biology is to use interactome networks to study genotype-phenotype relationships in complex diseases [1] . The idea of using interactome networks to infer gene function and gene-disease association comes from the well-validated principle of “guilt by association” , which states that physically interacting proteins tend to share similar functions and , by extension , tend to be involved in similar disease processes [1–4] . Recent advances in systems biology have spawned the view of human disease as a manifestation of genetic and environmental perturbations to the human interactome , a key postulate being that similar perturbation patterns lead to similar disease phenotypes [5–8] . A corollary is that , for diseases with both genetic and infectious contributors , similar perturbations of the human interactome by either disease mutations or pathogens can have similar disease consequences . This corollary has been tested for several viral species at the level of whole proteins [9 , 10] . For example , Gulbahce et al . used yeast two-hybrid screens to map binary interactions between Epstein-Barr virus ( EBV ) and human papillomavirus ( HPV ) proteins and human proteins , and transcriptionally profiled human cell lines exogenously expressing HPV oncoproteins E6 and E7 [9] . They found that human genes associated with EBV- and HPV-implicated genetic diseases were often either directly targeted by the virus or transcriptionally regulated by viral targets . This finding led to the idea that oncoviral proteins may preferentially target host proto-oncogenes and tumour suppressors , which was experimentally validated in four families of DNA oncoviruses [10] . Despite insights from these studies on the etiology of virally-implicated genetic diseases , there has yet to be a systematic , structure-based comparison of mutation-induced and pathogen-induced perturbations of the human interactome . A high-resolution , structurally-resolved network biology approach is important for unravelling complex genotype-phenotype relationships , because mutations occurring in different PPI-mediating interfaces on the same protein often have distinct functional impacts and phenotypic consequences [5–8] . In this regard , structural systems biology has proved useful in uncovering evolutionary properties of single- and multi-interface PPI network hubs , systems-level principles governing human-virus interactions , and systems properties of disease variants [6 , 11 , 12] . For instance , by constructing atomic-resolution human-virus and within-human protein interactomes , Franzosa and Xia discovered that viral proteins tend to target existing endogenous PPI interfaces in the human interactome , rather than creating exogenous interfaces de novo , thereby efficiently perturbing multiple endogenous PPIs involved in cell regulation [12] . In a follow-up study , Garamszegi et al . expanded the coverage of the human-virus interactome using domain-resolved models of PPIs , and found that viral proteins tend to deploy short linear motifs to bind a variety of human protein domains [13] . The economical and pleiotropic nature of “host domain-viral motif” interactions reflects the efficiency with which viruses rewire the human interactome given limited genomic resources at their disposal . Meanwhile , Wang et al . constructed a domain-resolution within-human interactome where protein domains are annotated with disease variant information [6] . They found that mutations occurring in different PPI-mediating domains within the same protein tend to be associated with different disorders ( “gene pleiotropy” ) . By contrast , mutations occurring in the domains of two different but interacting proteins , where the interaction is mediated by said domains , tend to be associated with the same disorder ( “locus heterogeneity” ) . These studies attest to the utility of structural systems biology in the study of infectious and genetic diseases . Here , we apply structural systems biology to the study of virally-implicated genetic diseases ( VIDs ) , and rigorously test the postulate that endogenous genetic mutations and exogenous viral proteins give rise to similar disease phenotypes by inducing similar perturbations of the human interactome at the level of protein domains . Specifically , we constructed a domain-resolved human-virus protein interactome and characterized the distribution of genetic disease mutations with respect to human domains targeted by virus . Overall , we found that viral proteins and VID mutations induce similar perturbations of the human domain-resolved interactome , for individual viruses with clearly defined VIDs and sufficient numbers of host-virus PPIs ( including EBV , HPV and HIV ) , for oncoviruses , as well as for all viruses combined . We first analyzed the disease associations of host proteins targeted by viral proteins and confirmed that virus-targeted proteins tend to be causally associated with VIDs rather than non-VIDs . We then analyzed the domain-level distribution of disease mutations in virus-targeted proteins and found that virus-targeted domains are significantly enriched for mutations causing VIDs rather than non-VIDs . Using a pooled analysis of all oncoviruses and all oncomutations , we found oncovirus-targeted domains to be significantly enriched for mutations causing cancer rather than other diseases . Furthermore , domains of oncoproteins either physically targeted or structurally mimicked by oncoviruses are significantly enriched for cancer driver mutations rather than passenger mutations , which implies convergent perturbation of cancer driver pathways by diverse oncoviruses . Finally , we also assessed the extent to which viral proteins and VID mutations perturb the same domain-domain interactions ( DDIs ) in the human interactome . We found that viruses preferentially target DDI partners of domains harbouring VID mutations , regardless of whether the DDI partners themselves are susceptible to known disease mutations . By correlating the equivalent pathogenicity of viral proteins and VID mutations with their convergent perturbation of the human domain-resolved interactome , we provide a framework for high-resolution , network-based comparison of the functional impacts of both genetic and environmental disease factors . On a broader note , our finding implies that similar perturbations of the human interactome at the domain level can have similar phenotypic consequences , regardless of the source of perturbation .
We first acquired human-endogenous and human-virus binary PPI data from IntAct , HPIDB 3 . 0 , and the HIV-1 Human Interaction Database [14–18] . Only PPIs supported by at least one PubMed ID were included in the whole-protein resolution human-virus interactome , which consists of 173830 PPIs between 15995 human proteins , and 28531 PPIs between 7761 human proteins and 624 viral proteins . 7211 human proteins participate in both endogenous and exogenous PPIs . To build homology models of PPIs , we collected high-confidence domain-domain interaction ( DDI ) and domain-motif interaction ( DMI ) templates derived from 3D structures of protein complexes in the Protein Data Bank , and scanned protein sequences for the occurrence of Pfam domains and domain-binding linear motifs [19–23] . Structural models were assigned to each PPI by extracting all DDIs and DMIs possibly mediating the PPI . The resulting domain-resolved human-virus structural interaction network ( hvSIN ) consists of 61041 PPIs between 11596 human proteins , and 4654 PPIs between 1590 human proteins and 405 viral proteins . 1517 human proteins participate in both endogenous and exogenous portions of hvSIN . We then obtained manually-curated disease variant data from UniProtKB and ClinVar [24 , 25] , selecting missense variants located inside Pfam domains for our analyses . Overall , 19047 mutations associated with 5383 diseases were mapped to 3585 domains of 2622 proteins . 14720 mutations associated with 4185 diseases were mapped to 2642 domains of 1864 human proteins in hvSIN . Table 1 lists the number of mutations by the type of domain in which they occur . Incidentally , 1272 domains of 957 human proteins in hvSIN are susceptible to disease mutations , but lack interacting domains or motifs . 850 of these 1272 domains harbour a total of 4154 mutations associated with 1381 diseases that are not accounted for by mutations occurring in PPI-mediating domains in hvSIN . Because the completeness of a domain’s PPI profile depends largely on the interactome search space and availability of 3D structures of protein complexes , and domains often have important biological functions besides mediating PPIs ( e . g . enzymatic or nucleotide-binding activity ) , we included all domains of virus-targeted host proteins in a comprehensive analysis of the domain-level distribution of disease mutations . To relate the equivalent pathogenicity of viral proteins and VID mutations to their equivalent perturbation of the host interactome , we first characterized the mutational landscape of human proteins targeted by EBV , HPV and HIV , three viruses with clearly defined VIDs and sufficient numbers of host-virus PPIs . Since most oncoviruses are causally implicated in only a few site-specific malignancies ( e . g . HBV/HCV in hepatocellular carcinoma , KSHV in Kaposi’s sarcoma , and HTLV in adult T-cell lymphoma ) , and various types of cancer share common molecular hallmarks [26 , 27] , to increase the statistical power of our analysis and establish whether a general equivalence exists between endogenous and exogenous perturbagens of oncogenic pathways , we also performed a pooled analysis of host proteins targeted by diverse oncoviruses , by considering all types of cancer as interchangeable diseases , all oncomutations as interchangeable endogenous perturbagens , and all oncoviral proteins as interchangeable exogenous perturbagens . We found that for EBV , HIV , HPV and a broad spectrum of oncoviruses , virus-targeted host proteins tend to be causally associated with VIDs ( Fig 1 ) , and virus-targeted host domains tend to harbour mutations causally associated with VIDs ( Fig 2 ) . We discuss our findings for each type of virus below . A full list of VIDs and disease-associated proteins for EBV , HPV and HIV can be found in S1 Table . A main challenge in cancer research is to distinguish mutations which confer clonal growth advantage ( i . e . drivers ) , from mutations that do not cause clonal expansion ( i . e . passengers ) [75] . Large-scale cancer genome sequencing projects have enabled systematic identification of cancer driver proteins and mutations [76] . Rozenblatt-Rosen et al . previously constructed an oncovirus-human interactome and demonstrated , at the whole-protein level , comparability between oncoviral perturbation and conventional functional genomics approaches to cancer gene discovery [10] . However , by representing proteins and PPIs as generic nodes and edges , their approach is not sensitive enough to distinguish driver mutations from passenger mutations occurring in the same oncoprotein . As we demonstrated earlier in the case of pleiotropic oncoproteins , the oncogenicity or “driver-ness” of a mutation is often correlated with its occurrence in oncovirus-targeted domains ( OVTDs ) . To confirm that oncoviruses can help identify driver proteins , we first cross-classified human proteins in hvSIN by whether they are oncoviral targets , and whether they are curated by the Cancer Gene Census ( CGC ) as being causally implicated in cancer , i . e . driver proteins [76] . Out of 727 oncoviral targets , 93 ( 12 . 8% ) are in CGC , whereas out of 10897 remaining human proteins in hvSIN , 514 ( 4 . 7% ) are in CGC . In other words , there is a 3-fold enrichment of driver proteins among oncoviral targets ( Fisher’s exact test , two-tailed P = 3 × 10−16 ) ( Fig 5A ) . Next , to find out if oncoviruses can also help identify driver mutations , we cross-classified mutations in oncoproteins by whether they are drivers or passengers , and by whether they map to OVTDs . Oncogenic and resistance mutations with a ClinVar clinical significance value of “pathogenic” or “likely pathogenic” are considered drivers , while passengers include all other missense mutations in oncoproteins that are catalogued by ClinVar and COSMIC . Out of 194 oncoproteins with annotated driver mutations , we identified 30 oncoproteins as having at least one OVTD . Pooled analysis of all 30 oncoproteins mapped 340/398 ( 85 . 4% ) driver mutations and 3673/7177 ( 51 . 2% ) passenger mutations to OVTDs . In other words , the odds of finding a driver mutation in OVTDs is 5 times as high as that in non-OVTDs ( Fisher’s exact test , two-tailed P < 2 . 2 × 10−16 ) ( Fig 5B ) . Closer inspection identified 19 candidates for focused investigations into the common basis of viral and mutational oncogenesis ( Table 2 ) : ( I ) 7 oncoproteins where all domains are OVTDs , and the driver:passenger ratio is higher than the average ratio across all oncoproteins; ( II ) 8 oncoproteins where some domains are OVTDs , and driver mutations are exclusively found in OVTDs; and ( III ) 4 oncoproteins where some domains are OVTDs , and driver mutations are significantly enriched in OVTDs ( Fisher’s exact test , two-tailed P < 0 . 05 ) . An example of each type of candidate is given in Fig 6 . Viruses are known to encode structural homologues that mimic host domains in order to modulate the biological activities of host targets . Such viral homology domains ( VHDs ) play key roles in mediating immune response ( e . g . PF00048 in CMV and KSHV ) , apoptosis ( e . g . PF00452 in EBV and KSHV ) , cell differentiation ( e . g . PF07684 in feline leukemia virus ) , and protein phosphorylation ( e . g . PF06734 in CMV ) , among other cellular processes involved in virally-implicated diseases . VHDs often compete with cellular counterparts for interaction partners , thereby rewiring host signaling networks to the virus’s advantage . Table 3 lists instances of human proteins convergently targeted by human domains and oncoviral homology domains in hvSIN . The preceding section established that oncovirus-targeted host domains are enriched for cancer driver mutations . Here , we test the hypothesis that oncovirus-mimicked host domains are also enriched for cancer driver mutations , independent of whether they are physically targeted by the virus . To this end , we identified 21 oncoproteins having at least one oncovirus-targeted domain ( OVTD ) and at least one viral homology domain ( VHD ) . We further classified viral homology domains ( VHDs ) into those enriched in oncogenic viruses ( oncoviral homology domains , or OVHDs ) , versus those enriched in non-oncogenic , i . e . “generic” viruses ( generic viral homology domains , or GVHDs ) ( Methods , S2 Table ) . We found that domains structurally mimicked by oncoviruses ( OVHDs ) are more likely to harbour driver mutations , compared to domains structurally mimicked by generic viruses ( GVHDs ) , independent of whether the domain is physically targeted by oncoviruses ( OVTD ) ( CMH test , common odds ratio = 2 . 2 , P = 5 × 10−5 ) . We then analyzed the mutational landscape of 44 oncoproteins having at least one oncoviral homology domain ( OVHD ) but not physically targeted by the virus , i . e . having no OVTDs . Pooled analysis of all 44 oncoproteins mapped 245/298 ( 82 . 2% ) driver mutations and 5422/9554 ( 56 . 8% ) passenger mutations to OVHDs . In other words , the odds of finding a driver mutation in OVHDs is 3 times as high as that in non-OVHDs ( Fisher’s exact test , two-tailed P < 2 . 2 × 10−16 ) ( Fig 5B ) . Closer inspection identified 23 candidates for focused investigations into the common basis of viral and mutational oncogenesis ( Table 4 ) : ( I ) 4 oncoproteins where all domains are OVHDs , and the driver:passenger ratio is higher than the average ratio across all oncoproteins; ( II ) 16 oncoproteins where some domains are OVHDs , and driver mutations are exclusively found in OVHDs; and ( III ) 3 oncoproteins where some domains are OVHDs , and driver mutations are significantly enriched in OVHDs ( Fisher’s exact test , two-tailed P < 0 . 05 ) . An example of each type of candidate is given in Fig 7 . In summary , oncovirus-mimicked host domains are enriched for cancer driver mutations , regardless of whether these domains are physically targeted by the virus . Gulbahce et al . previously hypothesized , and established at the whole-protein level , that viruses and VID mutations induce similar perturbations of the human interactome [9] . Here , we test the same hypothesis at the higher resolution of protein domains , by examining whether viruses and VID mutations perturb the same domain-domain interactions ( DDIs ) in the human interactome . In other words , do viruses tend to target DDI partners of domains harbouring VID mutations ( viral disease domain-interacting domains , or VDDiDs ) , rather than DDI partners of domains harbouring non-VID mutations ( non-viral disease domain-interacting domains , or nVDDiDs ) ( Fig 8A ) ? As some domains can interact with both VID domains and non-VID domains , we define VDDiDs as domains that interact with at least one VID domain , and nVDDiDs as domains that exclusively interact with non-VID domains . We found that EBV and HPV exhibit a slight preference for targeting VDDiDs , although the effect sizes are not statistically significant ( 42/62 VDDiDs vs . 58/104 nVDDiDs for EBV , and 20/29 VDDiDs vs . 41/69 nVDDiDs for HPV ) . HIV targets 218/309 ( 70 . 6% ) VDDiDs and 193/346 ( 55 . 8% ) nVDDiDs , representing a 1 . 9-fold enrichment of VDDiDs among HIV-targeted domains ( Fisher’s exact test , two-tailed P = 1 × 10−4 ) . Similarly , oncoviruses target 204/285 ( 71 . 6% ) VDDiDs and 164/291 ( 56 . 4% ) nVDDiDs , i . e . a 1 . 9-fold enrichment of VDDiDs among oncovirus-targeted domains ( Fisher’s exact test , two-tailed P = 1 × 10−4 ) . Finally , a meta-analysis on the common effect of all viral proteins and all mutations causing proliferative and immunological diseases found that viruses target 424/599 ( 70 . 8% ) VDDiDs and 350/551 ( 63 . 5% ) nVDDiDs , i . e . a 1 . 4-fold enrichment of VDDiDs among virus-targeted domains ( Fisher’s exact test , two-tailed P = 0 . 01 ) ( Fig 8B ) . Virus’s preferential targeting of VDDiDs may be confounded by the tendency for viruses to target VID domains ( Fig 2 ) , and the tendency for VID domains to interact among themselves . We therefore excluded domains susceptible to known disease mutations and examined the extent to which virus targets “non-disease” domains that interact with VID domains . We found that HIV targets 179/250 ( 71 . 6% ) VDDiDs and 164/285 ( 57 . 5% ) nVDDiDs that do not harbour any known disease mutation ( Fisher’s exact test odds ratio = 1 . 9 , two-tailed P = 8 × 10−4 ) . Similarly , oncoviruses target 165/230 ( 71 . 7% ) VDDiDs and 137/237 ( 57 . 8% ) nVDDiDs that do not harbour any known disease mutation ( Fisher’s exact test odds ratio = 1 . 8 , two-tailed P = 2 × 10−3 ) . Pooled analysis of all viruses found that overall , viruses target 345/481 ( 71 . 7% ) VDDiDs and 295/456 ( 64 . 7% ) nVDDiDs that do not harbour any known disease mutation ( Fisher’s exact test odds ratio = 1 . 4 , two-tailed P = 0 . 02 ) . Virus’s preferential targeting of VDDiDs supports our hypothesis that viruses and VID mutations inducing similar disease phenotypes convergently perturb the host domain interactome , possibly unveiling core disease modules underlying clinically heterogeneous virally-implicated diseases ( Fig 9 ) .
Structural interaction networks serve as a valuable tool for understanding the molecular mechanisms of genetic diseases , as well as the fundamental differences between endogenous and exogenous PPI networks . As experimental determination of protein structure remains an arduous task , homology modelling offers an efficient alternative for the structural annotation of protein complexes . This is based on the observation that PPIs are often mediated by evolutionarily conserved structural modules , such as domains and short linear motifs [77] . Here , we reassess the role of viral proteins as surrogates for human disease variants in relating interactome network perturbation to disease phenotypes , using a domain-resolved human-virus protein interactome where human domains are annotated with disease variant information . Compared to previous work demonstrating general proximity between viral targets and VID proteins in the human interactome , our results provide a structural explanation for the equivalent pathogenicity of viral proteins and VID mutations . Whereas previous studies merely recognized the existence of viral homologues of cellular domains , we delve deeper into the functional implications of oncoviral domain homology . Our approach can readily identify domains convergently targeted or mimicked by diverse oncoviruses for focused screening of driver mutations across various types of cancer . Further characterization of cellular domains and motifs interacting with domains targeted or mimicked by viruses may uncover immune evasion strategies exploited in common by cancer cells and pathogens , and shed light on pathways dysregulated in other virally-implicated disorders . Although most of our findings are statistically significant , there are notable differences in the enrichment of VID mutations in virus-targeted domains , both among individual viruses ( EBV , HPV and HIV ) , as well as between single-virus analysis and pooled analysis on multiple viruses . For single-virus analysis , enrichment effect size and significance are impacted by the number of virus-host protein-protein interactions and virus-specific diseases , which ultimately determine the statistical power . Pooled analysis on all oncoviruses detected trends in the same direction as analysis on single oncoviruses ( EBV and HPV ) , but with higher statistical power . In addition to investigator bias resulting in some viruses having a higher number of mapped virus-host PPIs , it is also possible that certain viruses prefer to perturb host regulatory network , rather than host PPI network , which is beyond the scope of this work . Compared to direct targeting of VID domains ( a “first-degree” effect ) , viral targeting of the interaction partners of VID domains is expected to have a weaker , “second-degree” effect on the VID domains . This partly explains why results of the “first-degree” analysis on EBV and HPV ( Fig 2 ) are stronger than those of the “second-degree” analysis ( Fig 8B ) . Our pooled analysis of all oncoviral targets and all oncomutations is motivated by the assumption of convergent evolution and mimicry of endogenous oncogenic mechanisms by diverse oncoviruses . There is compelling evidence of different oncoviruses complementing each other’s replication and persistence strategies , thus eliciting multiple cellular responses associated with the hallmarks of cancer . One example is primary effusion lymphoma , a disease causally linked to KSHV but also having an EBV component . While expression of KSHV lytic genes such as vIL-6 and K1 promote VEGF secretion and angiogenesis , concomitant expression of EBV latent genes confers additional anti-apoptotic properties to infected cells in the initial phase of lymphomagenesis [78 , 79] . Given the paucity of context-dependent ( i . e . tissue- and disease-specific ) host-endogenous and host-pathogen PPI data , here we focus on establishing viral proteins and genetic mutations that induce similar disease phenotypes as generally equivalent perturbagens of the human interactome . Future work will also consider the diversity of host range and tissue tropism among different viruses , and the potentially distinct functional impacts of the same mutation in different cell types and diseases . One potential caveat of our interactome perturbation model is its incompleteness , due to the following reasons . Firstly , current mapping of the host-virus protein interactome is far from exhaustive . Secondly , some bona fide host-virus PPIs cannot be modelled by existing domain-based interaction templates . Thirdly , virus may not interact with a host protein via PPI , but rather regulate its expression via transcriptional or epigenetic mechanisms . Lastly , our study only considers missense mutations , because domain-based analysis of interactome perturbation requires precise positioning of mutations with respect to protein domains . Missense mutations can be unambiguously mapped to individual domains , whereas other types of mutations ( e . g . nonsense or frameshift ) may cause more drastic changes in the protein structure and are more difficult to map to individual domains . We are aware , however , of literature suggesting that nonsense and frameshift mutations tend to occur more frequently in tumour suppressor genes than in oncogenes [80] . Effects of these mutations on the integrity of the human interactome warrant further investigation . Still , despite the incompleteness of our model , we observed significant convergent perturbation of the human domain-resolved interactome by viruses and mutations inducing similar disease phenotypes . The advent of high-throughput biotechnology has made it possible to comprehensively characterize genomic variations in and interspecies interactions between human and microbes , which play important roles in health and disease . As more data on pathogen-implicated diseases and host-pathogen interactions emerge , our approach may be extended to the study of bacterial diseases and co-infections involving multiple pathogenic species , such as the co-pathogenesis of HIV and Mycobacterium tuberculosis . By combining these data within the framework of structural systems biology , our work sets the stage for multi-scale , integrative investigations into endogenous and exogenous perturbagens of the human interactome , thus helping to elucidate the molecular mechanisms of infection and its possible connections to genetic diseases such as cancer , autoimmunity , and neurodegeneration .
Human-endogenous and human-virus binary PPI data were obtained from IntAct [14] , HPIDB [15] , and the HIV-1 Human Interaction Database [16–18] . Structural templates for domain-domain and domain-motif interactions were obtained from 3did [19] , iPfam [21] and ELM [20] . Protein sequences were scanned for Pfam domains using InterProScan under default settings ( version 5 . 30–69 . 0 ) [23 , 81] , and for the occurrence of domain-binding motifs as defined by 3did and ELM . Domain-based interaction models were assigned to each PPI by extracting all DDIs and DMIs possibly mediating the PPI . Disease association and clinical significance of variants were obtained from UniProtKB , ClinVar , and COSMIC [24 , 25 , 76] . Ensembl Variant Effect Predictor ( VEP v93 . 0 ) was used for extracting variant genomic location , variation class , reference allele , HGVS notations , amino acid position , overlapping Pfam domains , among other features [82] . To facilitate counting of mutational events , variants are annotated with RefSNP IDs using VEP’s check_existing flag . Variants not co-located with any known variant are merged based on identical genomic location , variation class , and shared alleles , as per NCBI guidelines for merging submitted SNPs into RefSNP clusters ( https://www . ncbi . nlm . nih . gov/books/NBK44417/ ) . Only missense mutations located inside Pfam domains were retained for analyses . Assignment of each virally-implicated disease ( VID ) to EBV , HPV and HIV was based on at least two literature sources ( S1 Text ) . To minimize redundancy in disease annotation , UMLS and OMIM IDs given to subtypes of the same disease were merged into the more general Disease Ontology [83] , Orphanet [84] and MeSH IDs . Oncoviruses are as classified by CDC , IARC , and MeSH ( https://www . ncbi . nlm . nih . gov/mesh/68009858 ) . Cancer is defined as any disease whose parent terms include “DOID:162” , “ORPHA:250908” , or MeSH IDs beginning with “C04 . 557|C04 . 588|C04 . 619|C04 . 626|C04 . 651|C04 . 666|C04 . 682|C04 . 692|C04 . 697|C04 . 700|C04 . 730|C04 . 834|C04 . 850” . Diseases without Disease Ontology , Orphanet or MeSH IDs are manually labelled as “cancer” if their names match the following regular expression: “blastoma|cancer|carcino*|glioma|leukemia|leukaemia|lymphoma|melanoma|neoplas*|sarcoma|tumour|tumor” . Proliferative diseases have parent terms “DOID:14566” , “ORPHA:250908” , or MeSH IDs beginning with “C04” . Immunological diseases have parent terms “DOID:2914” , “ORPHA:98004” , or MeSH IDs beginning with “C20” . All statistical analyses were conducted in R [85] . Plots of domain-level distribution of disease mutations were created with Protter [86] . Pfam domain annotation for all human and viral proteins in UniProt was retrieved from InterPro ( Release 69 . 0 ) [87] . We define viral homology domains ( VHDs ) as Pfam domains conserved between human and viral proteins . For each VHD , the likelihood of it occurring in oncoviruses was calculated as the number of oncoviruses encoding the VHD , divided by the total number of unique oncoviral species in UniProt . Similarly , the likelihood of a VHD occurring in “generic” ( i . e . non-oncogenic ) viruses was calculated as the number of generic viruses encoding the VHD divided by the total number of unique generic viral species in UniProt . The observed likelihood ratio ( LR ) of an oncovirus vs . a generic virus encoding the VHD is then the ratio of the two likelihoods . We then permuted the label “oncovirus” and “generic virus” 10000 times among viruses encoding the VHD , thereby obtaining a null distribution for the LR . An empirical p-value for the enrichment or depletion of a VHD in oncoviral proteomes was calculated according to [88] . VHDs whose observed LR > 1 and Benjamini-Hochberg adjusted p-values ( q-values ) < 0 . 1 are considered enriched in oncoviral proteomes . These VHDs and other VHDs exclusively occurring in oncoviruses are called oncoviral homology domains ( OVHDs ) . Likewise , VHDs whose observed LR < 1 and q-values < 0 . 1 are considered enriched in generic viral proteomes . These VHDs and other VHDs exclusively occurring in generic viruses are called generic viral homology domains ( GVHDs ) . | Cellular function and behaviour are driven by highly coordinated biomolecular interaction networks . A prime example is the protein-protein interaction network , often simply referred to as the “interactome” . Recent advances in systems biology have spawned the view of human disease as a manifestation of genetic and environmental perturbations to the human interactome , a key postulate being that similar perturbation patterns lead to similar disease phenotypes . Here , we took a structural systems biology approach to compare mutation-induced and virus-induced perturbations of the human interactome in diseases with both genetic and viral contributors . Specifically , we constructed a domain-resolved human-virus protein interactome and characterized the distribution of genetic disease mutations with respect to human domains either physically targeted or structurally mimicked by virus . Overall , we found significant convergent perturbation of the human domain-resolved interactome by viruses and mutations inducing similar disease phenotypes . Structure-guided , integrated analysis of host genetic variation and host-pathogen protein interaction data may help elucidate the molecular mechanisms of infection and reveal its connections to genetic diseases such as cancer , autoimmunity , and neurodegeneration . On a broader note , our finding implies that similar perturbations of the human interactome at the domain level can have similar phenotypic consequences , regardless of the source of perturbation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"microbial",
"mutation",
"interaction",
"networks",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viruses",
"and",
"cancer",
"pathogens",
"cancer",
"risk",
"factors",
"microbiology",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"oncology",
"rna",
"viruses",
"medical",
"risk",
"factors",
"proteins",
"medical",
"microbiology",
"hiv",
"epidemiology",
"microbial",
"pathogens",
"pathogenesis",
"molecular",
"biology",
"biochemistry",
"host-pathogen",
"interactions",
"virology",
"viral",
"pathogens",
"protein",
"domains",
"genetics",
"biology",
"and",
"life",
"sciences",
"human",
"genetics",
"lentivirus",
"genetics",
"of",
"disease",
"organisms"
] | 2019 | Convergent perturbation of the human domain-resolved interactome by viruses and mutations inducing similar disease phenotypes |
Louse-borne relapsing fever ( LBRF ) borreliosis is caused by Borrelia recurrentis , and it is a deadly although treatable disease that is endemic in the Horn of Africa but has epidemic potential . Research on LBRF has been severely hampered because successful infection with B . recurrentis has been achieved only in primates ( i . e . , not in other laboratory or domestic animals ) . Here , we present the first non-primate animal model of LBRF , using SCID ( -B , -T cells ) and SCID BEIGE ( -B , -T , -NK cells ) immunocompromised mice . These animals were infected with B . recurrentis A11 or A17 , or with B . duttonii 1120K3 as controls . B . recurrentis caused a relatively mild but persistent infection in SCID and SCID BEIGE mice , but did not proliferate in NUDE ( -T ) and BALB/c ( wild-type ) mice . B . duttonii was infectious but not lethal in all animals . These findings demonstrate that the immune response can limit relapsing fever even in the absence of humoral defense mechanisms . To study the significance of phagocytic cells in this context , we induced systemic depletion of such cells in the experimental mice by injecting them with clodronate liposomes , which resulted in uncontrolled B . duttonii growth and a one-hundred-fold increase in B . recurrentis titers in blood . This observation highlights the role of macrophages and other phagocytes in controlling relapsing fever infection . B . recurrentis evolved from B . duttonii to become a primate-specific pathogen that has lost the ability to infect immunocompetent rodents , probably through genetic degeneration . Here , we describe a novel animal model of B . recurrentis based on B- and T-cell-deficient mice , which we believe will be very valuable in future research on LBRF . Our study also reveals the importance of B-cells and phagocytes in controlling relapsing fever infection .
Bacteria of the genus Borrelia are spirochetes that cause either Lyme disease or relapsing fever ( RF ) . Borrelia species are transferred from animals to humans by tick bites , with the single exception of B . recurrentis , which is transmitted between humans by the body louse Pediculus humanus humanus . The louse-borne disease is not transmitted by the bite per se , but rather through contamination of abraded skin by feces or coelomic fluid released from lice that are crushed by scratching . P . humanus humanus is strictly a human-specific parasite that lives on the body and in the clothing of its host , and B . recurrentis has been found only in lice and humans [1] , [2] . In most cases , louse-borne relapsing fever ( LBRF ) presents with a sudden onset of fever ( typically 38 . 7–41°C ) and chills . The first fever period lasts on average 5–7 days and is accompanied by malaise , nausea , general aches , and enlargement of the spleen and liver . Compared to tick-borne RF , LBRF usually involves fewer relapses , but it results in far greater mortality , which can be as high as 40% if left untreated but as low as 1%–5% when antibiotic therapy is given . The Jarish-Herxheimer reaction and delayed onset of antimicrobial therapy are associated with an elevated mortality risk [1] , [2] , [3] . LBRF previously occurred worldwide in massive epidemics , the latest of which were seen during the two world wars . Today , the only endemic area is in the highlands of Ethiopia , and sporadic outbreaks have been observed in Sudan , where the disease is associated with natural disasters , famine , and refugee camps [1] , [2] , [4] , [5] , [6] . Although B . recurrentis is currently found only in the Horn of Africa , it can become established wherever there are human body lice and thus it has a high potential to cause global epidemics [7] , especially today due to the massive political turmoil in the Horn of Africa . A population of lice can increase by 11% a day , which gives a clue as to just how rapidly an outbreak can spread in places like refugee camps [1] , [8] . Therefore , LBRF may be even more important now than it has been for decades . Despite increasing epidemic potential and important scientific progress such as the recent genome sequencing of B . recurrentis [9] , research is hampered by the lack of feasible animal models . In the early 20th century , scientists attempted to infect commonly used laboratory animals as well as several species of domestic and wild animals , but only experiments using primates seemed to have been successful [10] . Moreover , B . recurrentis was not isolated in vitro until 1994 , when Cutler and coworkers [11] managed to grow a few strains in BSK broth . That event paved the way for microbiological and biochemical investigations , but the lack of an animal model has continued to hinder performance of all types of studies focused on host-pathogen interactions , as well as other experiments that require an in vivo model system . We used immunodeficient mouse strains to develop an animal model of B . recurrentis infection that is more practical in all aspects compared to a primate-based model . SCID mice carry the Prkdcscid mutation , which results in severe combined immunodeficiency due to a defect in V ( D ) J recombination; this condition impairs the animals' ability to generate B- and T-cell antigen receptors , thus leading to very low numbers of functional lymphocytes [12] . NUDE mice lack a thymus , and they have impaired T-cell function due to the Foxn1nu mutation [13] , whereas the cells of their innate immune system ( monocytes , macrophages , natural killer cells and neutrophils ) and their complement system remain functional . Mice with the BEIGE mutation ( Lystbg ) have defective natural killer ( NK ) cells [14] . Hence , SCID BEIGE mice lack B- , T- , and NK-cells . Clodronate liposomes have been used extensively in infection models to study the pathobiological effect of systemic depletion of phagocytic cells ( e . g . , macrophages ) and the immunological importance of such cells in combating infection [15] , [16] , [17] , [18] . These liposomes are artificially prepared lipid vesicles that encapsulate clodronate , and they can be injected intravenously to attack phagocytic cells that are present in or in contact with the blood ( e . g . , macrophages in the spleen and liver ) . The clodronate is ingested by and accumulated within phagocytic cells , and after an intracellular threshold concentration of the drug is exceeded , the cells are irreversibly damaged and die by apoptosis , as described elsewhere [19] . Still , it is important to remember that clodronate does not eliminate all phagocytic cells , and that new phagocytes will continuously appear as soon as the liposomes are consumed; in other words , the phagocytic activity cannot be totally inhibited . Free clodronate ( i . e . released from dead macrophages ) has a very short half-life and is quickly removed from the circulation by the renal system , furthermore it can not easily pass through cell membranes and thus can not affect non-phagocytic cells [20] . By using animals deficient in various immune cells and inducing systemic depletion of phagocytes , much can be learned about the immune defense against RF borreliosis and host-pathogen interactions . In the current investigation , we found that SCID mice could support growth of B . recurrentis , which led to a low-grade , persistent disease . Moreover , employing clodronate liposomes to deplete phagocytic cells resulted in a one-hundred-fold rise in the spirochete titers . Thus , in this paper we present the first non-primate animal model for studies of LBRF . We also characterize B . recurrentis infection and compare it with the closely related species B . duttonii , which is virulent in wild-type mice and has been studied extensively in mouse models [21] , [22] .
The two B . recurrentis strains A11 and A17 were isolated in Ethiopia and were kindly provided by Sally Cutler , University of East London . Isolation and characterization of these strains has been described by other investigators [11] . B . duttonii 1120K3 was kindly provided by Guy Baranton , Institute Pasteur . All bacteria were cultured at 37°C in BSK-II medium supplemented with 10% ( v/v ) rabbit serum and 1 . 4% ( w/v ) gelatin , as described elsewhere [23] . The bacteria used in experiments had less than 10 passages through animals or in vitro in our lab . To create and validate a B . recurrentis non-primate animal model , we injected 1×106 B . duttonii , B . recurrentis A11 , or B . recurrentis A17 subcutaneously ( s . c . ) into four 6-week-old male mice of each of the following strains ( all from Taconic , Denmark ) : BALB/c ( BALB/cAnNTac ) , BALB/c NUDE ( C . Cg/AnNTac-Foxn1nu NE9 ) , SCID ( C . B-Igh-1b/IcrTac-Prkdcscid ) , and SCID BEIGE ( C . B-Igh-1b/GbmsTac-Prkdcscid-Lystbg N7 ) . The animals were kept in a filter cabinet and given food and water ad libitum , with all maintenance performed according to Swedish animal welfare guidelines . Tail blood was collected daily , and bacteria in the samples were counted by phase contrast microscopy during the first 20 days of infection . SCID and SCID BEIGE mice infected with B . recurrentis were kept until day 150 post infection ( p . i . ) , and spirochetemia was quantified weekly by microscopy . All animal experiments were approved in advance by the Laboratory Animal Ethics Committee of Umeå University . Starting one day before infection and subsequently every fifth day , an intravenous ( i . v . ) injection of 100 µl clodronate liposomes in phosphate-buffered saline ( PBS ) was administered in the tail of SCID BEIGE and BALB/c mice to deplete phagocytic cells present in the blood and organs ( e . g . , spleen and liver ) ; this was done as described elsewhere [19] . Since the phagocytes might also have ingested liposomes that contained PBS instead of clodronate , which would have reduced the phagocytic efficiency , we gave mice i . v . injections of 100 µl PBS as negative controls . The animals were infected with either B . recurrentis A17 or B . duttonii 1120K3 five days after the first clodronate injection . Spirochetemia ( non-Gaussian ) was analyzed by the Mann-Whitney U-test , and the results are presented as medians with 25th and 75th percentile bars to illustrate variance . Difference in spleen weight was assessed by Student's t-test .
We conducted tests to determine whether any of the commercially available immunodeficient mouse strains can support growth of B . recurrentis . Initially , the two B . recurrentis strains A11 and A17 , and B . duttonii strain 1120K3 were inoculated into immunocompetent wild-type BALB/c mice , NUDE mice lacking T-cells , SCID mice lacking B- and T-cells , and SCID BEIGE mice lacking B- , T- , and NK-cells . As expected , only B . duttonii established detectable infection in the BALB/c mice ( Figure 1A ) , which concurred with the results of similar previous experiments [22] , [24] . The total B . duttonii spirochetemia did not differ significantly between NUDE mice and BALB/c mice ( Figure 1A–B ) , and neither the A11 nor the A17 B . recurrentis strain caused detectable spirochetemia in BALB/c or NUDE mice , indicating that T-cells are of minor importance for RF immune defense . However , both the B . recurrentis strains did establish infection in SCID and SCID BEIGE mice , although the spirochetemia was about 200-fold lower than that caused by B . duttonii infection ( Figures 1 C–D and 2 ) . Due to the lack of antibodies , B . recurrentis spirochetemia did not display a relapsing pattern . Instead , spirochete titers remained fairly constant over time , and there were only minor fluctuations between the mice , which were probably caused by individual , biological variations in the host-pathogen interactions ( Figure 2 ) . Both the A11 and A17 bacteria were persistent and remained at fairly low levels in the blood ( about 2×105/ml ) at least until day 150 post infection . The A11 strain caused a significantly ( p<0 . 01 ) milder infection , with spirochetemia about half the magnitude of that caused by the A17 strain in both SCID and SCID BEIGE mice ( Figure 2 ) . The animals behaved normally and showed no outer sign of disease , and their weight pattern corresponded to what was seen in uninfected animals ( data not shown ) . To ascertain whether the passage through SCID BEIGE mice rendered the bacteria capable of causing spirochetemia in wild-type mice , we inoculated four BALB/c mice with spirochetes obtained from two A11-infected and two A17-infected animals . None of those four mice developed detectable spirochetemia . The B . recurrentis A17 infection induced significant ( p<0 . 01 ) splenomegaly in SCID BEIGE mice . However , the spleens were much smaller in B-cell- and T-cell-deficient mice than in wild-type mice ( Figure 3 ) , which implies activation and multiplication of splenic immune cells . Surprisingly , B . duttonii did not cause a lethal infection in the B-cell-deficient animals , despite the indispensable role of B-cells suggested by the antibody-mediated clearance of antigenic variants in immunocompetent models ( Figure 1C , D ) . Although both B . recurrentis and B . duttonii spirochetemia were higher in B-cell-deficient mice ( p = 0 . 02 ) , which verifies significance of the B-cells , the disease was kept under control by B-cell-independent mechanisms ( Figures 1C , D and 2 ) . Notably , we also found that clodronate-treated ( i . e . , phagocyte-depleted ) mice infected with B . duttonii were unable to restrict the bacterial infection . These animals developed very high spirochetemia , which killed all the SCID BEIGE mice before day 8 p . i . ( Figure 4B ) . The pattern was similar in BALB/c mice , although one single individ survived and managed to control the infection ( Figure 4A ) . B . recurrentis A17 also caused substantial spirochetemia in clodronate-treated SCID BEIGE mice ( Figure 4C ) , although the level that was reached was about 10 times higher than the spirochetemia induced by B . duttonii in untreated wild-type mice ( Figure 4A and 4C ) . However , B . recurrentis was unable to establish infection in phagocyte-depleted BALB/c mice , which underlines the significance of B-cells and possibly also T-cells .
Several attempts have been made to establish B . recurrentis infection in various animals , but , until now , only primate models have been successful [10] , which has severely hampered research on LBRF . In the present study , we used mice deficient in T- and B-cells and induced phagocyte depletion by administering clodronate liposomes to create the first non-primate animal model of LBRF infection . The results of our experiments show that B . recurrentis could infect both SCID and SCID BEIGE mice . The two B . recurrentis strains A11 and A17 caused moderate , persistent infection in B- and T-cell-deficient SCID mice and B- , T- and NK-cell-deficient SCID BEIGE mice ( Figure 2 ) , but did not induce detectable spirochetemia in either wild-type BALB/c animals or NUDE mice lacking mature T-cells . The A11 and A17 strains we used were isolated from different LBRF patients in Ethiopia [11] and had an estimated history of ∼20 BSK passages . Since RF spirochetes do not lose plasmids in vitro ( which is the case in Lyme Borrelia spirochetes ) , and they maintain infectivity even after several passages in vitro [9] , [25] , the two strains we chose to use were probably well representative of B . recurrentis . Despite the ability of RF spirochetes to evade the host humoral response through antigenic variation , antibodies are definitely an important part of immune defense against this disease [26] , [27] . This is also reflected by the present findings showing higher B . duttonii spirochetemia and establishment of B . recurrentis infection in the B-cell-deficient mice . Moreover , the interaction between the humoral response and B . recurrentis seems to be somewhat different than noted for other RF-inducing bacteria , since B . recurrentis generally causes 0–4 relapses , whereas other African species cause 3–9 , as reviewed by other researchers [1] , [2] . T-cells are apparently less important , since we observed that B . duttonii spirochetemia was equally high in wild-type BALB/c animals and the athymic , T-cell-deficient NUDE mice ( Figure 1A , B ) . Similar results have been reported in experiments on both the RF agent B . turicatae and Lyme borreliosis [28] , [29] . Furthermore , an investigation of RAG2−/− and RAG2/IL-10−/− mice infected with B . turicatae has suggested that NK-cells play an important role [30] . In contrast , we found that SCID mice with the BEIGE mutation , which causes NK-cell deficiency , showed the same levels of B . recurrentis or B . duttonii spirochetemia as seen in SCID mice ( Figures 1C , D and 2 ) . On the other hand , phagocyte depletion by use of clodronate liposomes had a dramatic effect on both spirochetemia and disease . B . duttonii infection was uncontrollable in all animals but one , and even B . recurrentis reached spirochetemia of over 5×108/ml . The apparent “relapses” in clodronate-treated animals were not due to antigenic variation , but instead to partial host recovery from the phagocytic depletion that was repeated every fifth day ( Figure 4A , C ) . Even between the two peaks in B . recurrentis-infected mice , median spirochetemia never receded below 8×106/ml , which is a fairly high level ( Figure 4C ) . These results clearly illustrate that it is important for phagocytic cells to be able to filter and to some extent also control spirochetes in the blood , even in the absence of B- and T-cells . Furthermore , experiments in vitro have indicated that B . recurrentis avoids complement opsonization and phagocytosis by binding complement regulators such as factor H and by degrading C3b from its surface in order to utilize “hijacked” host plasmin [31] , [32] . However , this remains to be convincingly demonstrated in vivo . Despite all of these isolated findings , it is essential to bear in mind that the immune system is a tightly connected web of cells and signal molecules , and that removal of one cell type might have other downstream effects on overall immunity . The recent genome analysis performed by Lescot et al . [9] revealed that B . recurrentis evolved from B . duttonii through extensive genetic decay , probably involving loss of the genes mutS and recA , which are important for DNA repair . Both of these species are considered to be more or less specific to humans , and it has been debated whether this is due to a restricted host infectivity range , or simply to the fact that both the B . duttonii tick vector Ornithodorus moubata moubata and the B . recurrentis louse vector P . humanus humanus strongly prefer humans as a source of food . B . duttonii establishes RF in wild-type laboratory mice and rats that is similar to the human disease , and this spirochete has recently been found in chickens and pigs raised in proximity to tick-infested human dwellings in Tanzania [33] , indicating a wider potential host range than was previously believed . In short , B . duttonii can infect different species of animals , but it is ecologically very restricted by the host range of its vector . It might be assumed that B . recurrentis should also be able to infect non-primate animal species , perhaps even more severely than does B . duttonii , since in humans it is basically a more pathogenic strain of B . duttonii . Interestingly , this greater virulence seems to originate from the loss of a gene or genes rather than the gain of novel genes [9] , as has also been described in other louse-borne infections , such as Rickettsia prowazekii [34] . B . recurrentis has lost its ability to survive in hosts other than humans , and the way this spirochete is transmitted ( i . e . , not by the louse bite but by fecal or hemolymph contamination of abraded skin ) is a somewhat unsophisticated strategy that suggests a short evolutionary adaptation . It is tempting to speculate that the lost capacity of B . recurrentis to infect non-primate animals may also be a result of genomic decay . B . duttonii probably possesses mechanisms for evading the immune defense of host animals , which have disappeared in B . recurrentis during its rapid evolution simply because they are not needed since humans are the only host . Inasmuch as SCID mice are defective in B- and T-cell production , they readily accept foreign cells and tissues without rejection . Future experiments by our group should be aimed at creating humanized SCID-hu mice that generate human immune cells or carry stem-cell-producing human xenotransplants to pinpoint the factors that restrict LBRF to humans . Such strategies have been successfully applied in other human-specific infectious diseases , as described by other investigators [35] . For instance , SCID-hu mice with human B-cells may constitute a better model of human B . recurrentis infection that can facilitate studies of aspects such as antigenic variation that seems to be responsible for a difference causing fewer relapses compared to tick-borne RF [2] . B . recurrentis was grown in vitro for the first time in 1994 [11] , which opened the door for microbiological and immunological studies conducted in vitro [32] , [36] . In addition , the complete genomes of B . recurrentis and B . duttonii were recently sequenced and published for the first time [9] , and the first site-specific genetic manipulation of an RF agent was performed last year when the variable tick protein was knocked out and reconstituted in B . hermsii [37] . All of these achievements will definitely encourage further development of genetic tools and facilitate molecular biological investigations of RF . Here , we have described the first non-primate animal model of LBRF , which we believe is the fourth cornerstone needed to bring B . recurrentis research into the 21st century . | Research on Borrelia recurrentis , the agent of louse-borne relapsing fever ( LBRF ) , has been hampered by the lack of a feasible non-primate animal model . By using immunocompromised SCID mice deficient in B- and T-cells , we were able to establish a stable , persistent B . recurrentis infection with low spirochetemia . Furthermore , systemic depletion of phagocytes by use of clodronate liposomes increased the numbers of bacteria in blood , which demonstrates the importance of both the humoral response and phagocytosis in controlling relapsing fever infection . Lice are favored by the conditions related to the unfortunate turmoil and refugee camps prevailing in the Horn of Africa , and hence LBRF is more important now than it has been for several decades . The newly published genome sequence of B . recurrentis and techniques to genetically manipulate RF borreliae will be instrumental in understanding its complex biology . We therefore believe that our novel animal model will be a great asset that can facilitate future studies of the infection biology of B . recurrentis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases/bacterial",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"infectious",
"diseases/tropical",
"and",
"travel-associated",
"diseases",
"microbiology/immunity",
"to",
"infections"
] | 2009 | A Novel Animal Model of Borrelia recurrentis Louse-Borne Relapsing Fever Borreliosis Using Immunodeficient Mice |
Angiogenesis plays a key role in tumor growth and cancer progression . TIE-2-expressing monocytes ( TEM ) have been reported to critically account for tumor vascularization and growth in mouse tumor experimental models , but the molecular basis of their pro-angiogenic activity are largely unknown . Moreover , differences in the pro-angiogenic activity between blood circulating and tumor infiltrated TEM in human patients has not been established to date , hindering the identification of specific targets for therapeutic intervention . In this work , we investigated these differences and the phenotypic reversal of breast tumor pro-angiogenic TEM to a weak pro-angiogenic phenotype by combining Boolean modelling and experimental approaches . Firstly , we show that in breast cancer patients the pro-angiogenic activity of TEM increased drastically from blood to tumor , suggesting that the tumor microenvironment shapes the highly pro-angiogenic phenotype of TEM . Secondly , we predicted in silico all minimal perturbations transitioning the highly pro-angiogenic phenotype of tumor TEM to the weak pro-angiogenic phenotype of blood TEM and vice versa . In silico predicted perturbations were validated experimentally using patient TEM . In addition , gene expression profiling of TEM transitioned to a weak pro-angiogenic phenotype confirmed that TEM are plastic cells and can be reverted to immunological potent monocytes . Finally , the relapse-free survival analysis showed a statistically significant difference between patients with tumors with high and low expression values for genes encoding transitioning proteins detected in silico and validated on patient TEM . In conclusion , the inferred TEM regulatory network accurately captured experimental TEM behavior and highlighted crosstalk between specific angiogenic and inflammatory signaling pathways of outstanding importance to control their pro-angiogenic activity . Results showed the successful in vitro reversion of such an activity by perturbation of in silico predicted target genes in tumor derived TEM , and indicated that targeting tumor TEM plasticity may constitute a novel valid therapeutic strategy in breast cancer .
Elucidating the various cell signaling cascades , pathway crosstalk , and how they influence final cell fate and behavior is crucial for defining therapeutic intervention points aimed at driving a cell towards a desired state . To this end , modeling approaches can be used to perturb a biological system in silico to test hypotheses on a scale that would be unfeasible to test experimentally . Boolean models have been extensively used in the past to simulate the behavior of cells based on their network activity [1] . In a Boolean modeling approach , the nodes in a regulatory network represent the state of activation of a gene ( protein , receptor or ligand ) using discrete variables ( On or Off ) . The state of the network at a given instant can change depending on the state of the other nodes and can ultimately stabilize into attractors of either a single state ( steady state ) or an oscillating set of states ( cycling attractors ) [2] . Introducing perturbations in a biological regulatory network can change the attractors and even transition the system from one attractor to another one . The Boolean steady state of the network has been shown to correspond to the cellular states for various regulatory networks in the past [3] . Boolean modeling of steady state transitions helps in understanding the influence of perturbations on system wide behavior and has been used to identify the key molecular mechanisms controlling gene expression [4 , 5 , 6] and regulation [7 , 8] , cell differentiation [9] and signal transduction [10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20] . Most of these models were developed in synergy by wet and dry laboratories . However , to date , only few of them have reported experimental validations ( in primary cells ) of the proposed in silico predictions [12 , 15 , 16] . In the present study we describe the application of a Boolean modeling based approach to investigate the molecular mechanisms underlying the angiogenic function of tumor monocytes from breast cancer patients and the experimental validation of in silico predictions derived from this modeling . The formation of tumor-associated vasculature , a process also referred to as tumor angiogenesis , is essential for tumor progression . Tumor vessels can form from local pre-existing capillaries . This process is promoted by the recruitment of bone marrow-derived angiogenic cells ( i . e . mainly monocytes , dendritic cells and neutrophils ) at tumor sites [21 , 22 , 23] . Clinical studies have demonstrated in a variety of human solid tumors a positive correlation between increased micro-vessel density , infiltration of tumor-associated macrophages ( TAM ) [24] and unfavorable prognosis in cancer patients [25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33] . Recently , monocytes expressing the TIE-2/tek receptor tyrosine kinase ( TEM: TIE-2 expressing monocytes ) have been identified in peripheral blood and tumors of humans and mouse [34 , 35] . In experimental mouse models , TEM recruited to tumors accounted for apparently all angiogenic activity of bone marrow-derived cells since their selective ablation fully suppressed angiogenesis and induced tumor regression [34] . Hence , TEM appear to be key players in tumor angiogenesis but the tumor micro-environmental signals and the related signaling pathways governing their functions remain to be elucidated . Of particular interest from a disease standpoint is how TEM can be directed away from potentiating tumor angiogenesis and progression to monocytes being immunologically potent cells . The VEGFR-1 ( Vascular Endothelial Growth Factor Receptor-1 ) , TGFBR-1 ( Tumor Growth Factor β Receptor-1 ) , TNF-R1 ( Tumor Necrosis Factor Receptor-1 ) pathways have been reported to regulate tumor angiogenesis [36 , 37] , but their activities have not been examined in human TEM . While we have previously reported that TIE-2 and VEGFR kinase activities drive immunosuppressive function of TEM in human breast Cancer [38] , in this study , we investigated the contribution of these pathways along with TGFBR-1 and TNF-R1 pathways to TEM pro-angiogenic activity . We observed that the pro-angiogenic activity of TEM increased drastically from blood to tumor in breast cancer patients . We constructed an integrative and predictive model of TEM behavior to predict in silico all minimal perturbations that can transition the highly pro-angiogenic phenotype of breast tumor TEM into a weak pro-angiogenic phenotype and vice versa . By experimentally validating our computational predictions , we demonstrate here that the inferred regulatory network captured accurately patient TEM behavior . Thus , the contribution of the computational approach was not only essential to predict and tune TEM pro-angiogenic activity but also to identify the key underlying components and pathways of their pro-angiogenic activity . Finally , gene expression profiling of TEM transitioned to a weak pro-angiogenic phenotype confirmed that TEM infiltrating carcinoma of the breast remain plastic cells that can be reverted from pro-angiogenic and protumoral cells to immunological potent monocytes .
The angigoenic profile of TEM was investigated in a group of 40 newly diagnosed breast cancer patients ( Table 1 ) . We characterized by flow cytometry the phenotype of TEM from patient peripheral blood and freshly dissociated tumor specimens obtained at time of surgery ( see Material and Methods ) . Based on our immunostaining and flow cytometry protocol we observed that TEM did not constitute a distinct subset of monocytes . In contrast , all monocytes showed expression of TIE-2 , which was particularly low in patient blood and substantially higher on monocytes isolated from tumor tissue ( S1 Fig . and Table 2 ) . Thus , CD11b+ , CD14+ monocytes from patient blood and tumor tissue were referred to as “TEM” and compared with respect to receptor and cytokine expression . However , tumor TEM co-expressed VEGFR-1 and TGFR-1 at significantly higher levels compared to peripheral blood TEM ( Table 2 ) . We next assess the pro-angiogenic activity of TEM using the in vivo corneal vascularization assay [39] . The cornea itself is avascular and was injected with TEM isolated from patient peripheral blood and tumor tissue . Thus , any growth of new vessels from the peripheral limbal vasculature must be due to injected TEM and reflect their pro-angiogenic activity . Tumor TEM showed a heterogeneous and consistently high pro-angiogenic activity inducing cornea and iris vascularization . By contrast , blood TEM were unable to induce de novo vascularization of the cornea but did increase the pre-existing vascular network of the iris ( Fig . 1A ) . Thus , tumor and blood TEM show distinct pro-angiogenic phenotypes with the expression levels of TIE-2 , VEGFR-1 and TGFR-1 mirroring their pro-angiogenic activity ( Fig . 1A and Table 2 ) . Finally , secretions were profiled in the conditioned medium of patient-isolated TEM and revealed that tumor TEM are paracrine inducers of tumor angiogenesis by releasing high levels of angiogenic factors ( i . e . VEGF , bFGF , and ANG-1 ) and MMP9 ( matrix metalloproteinase 9 ) ( Fig . 1B ) . Blood and tumor TEM display a mixed M1-like ( tumor-associated macrophages releasing inflammatory molecules ) and M2-like ( immunosuppressive macrophages polarized by anti-inflammatory molecules ) phenotype , with secretion of both the pro- and anti-inflammatory cytokines IL-12 and IL-10 , respectively ( Fig . 1B ) . Given that TEM circulating in the blood infiltrate tumor tissue where they further differentiate [34] , our data suggest that the tumor microenvironment shapes their highly pro-angiogenic phenotype . The identification of the ligands and the pathways controlling the highly pro-angiogenic activity of tumor TEM is of paramount significance because it represents the rationale for a treatment directing TEM away from being cells supporting tumor growth . The strategy we selected to reach this goal combined computational and experimental approaches to simulate and predict the behavior of patient TEM subjected to various ligand combinations . Given the limited amounts of patient specimens and the low frequency of TEM ( TEM represented 6 . 7% ±2 . 5% of peripheral blood mononuclear cells and 22% ±2 . 7% of the tumor hematopoietic infiltrate ) , only a limited number of ligand combinations could be investigated experimentally . The availability of limited amounts of patient TEM was partially overcome by taking advantage of our recently developed model system of TEM differentiated in vitro by exposing CD34+ cord blood hematopoietic progenitors to breast cancer cell conditioned culture medium [38 , 40] . In vitro differentiated TEM ( thereafter named ivdTEM ) are angiogenic [38 , 40] and display an intermediate phenotype relative to blood and tumor TEM ( Table 2 ) . Consistent with their phenotype ( Table 2 ) , ivdTEM released intermediate amounts of angiogenic factors relative to blood and tumor TEM ( Table 3 ) . Moreover , the in silico modeling and predictions helped us to focus on the most clinically relevant monocytic ligands and to spare precious patient specimen . The workflow of our approach consists of five steps ( Fig . 1C ) : 1 ) experimental measurement of the responses of TEM differentiated in vitro to a set of ligands , 2 ) construction of a dynamic regulatory network based on these experimental data , 3 ) in silico prediction of the treatments altering TEM behavior , 4 ) experimental validation of computationally predicted treatments using ivdTEM and 5 ) validation the best predicted treatments in patient TEM ( Fig . 1C ) . Finally , to help shed light on possible molecular mechanisms underlying TEM pro-angiogenic transformation , we selected several treatment combinations and measured genome wide expression profiles for the TEM differentiated in vitro , comparing the state of the cells before and after treatment . Our strategy was to expose TEM to several treatments to identify the ligands and pathways critically controlling their pro-angiogenic activity . TEM differentiated in vitro were exposed to angiogenic factors ( VEGF , PlGF and ANG-1 , ANG-2 which are the ligands of VEGFR-1 and TIE-2 respectively ) in combination with either TGF-β or TNF-α and the changes in their phenotype , angiogenic activity and paracrine secretion profile were examined . These experimental results were used as the foundations for a computational model that would allow predicting treatments increasing or dampening TEM proangiogenic activity . First , changes in TEM phenotype were evaluated by flow cytometry 36h post treatment . Globally , treatments combined with TGF-β or TNF-α displayed a stronger impact on TEM phenotype than single treatments with however , the exception of TGF-β . Overall , CD11b , CD14 , VEGFR-1 and TIE-2 expression displayed larger changes in response to treatment than CCR5 , TNF-R1 and TGFBR-1 ( Fig . 2A ) . A hallmark of TGF-β treatments was a strong decrease in VEGFR-1 and CD11b expression and an increase in TIE-2 expression ( Fig . 2A ) . By contrast , TNF-α treatments had no impact on VEGFR-1 expression and TNF-α increased TIE-2 expression when combined with PlGF or ANG-2 ( Fig . 2A ) . We assessed the impact of the treatments on the pro-angiogenic activity of TEM using in vitro HUVEC ( Human Umbilical Vascular Endothelial Cells ) sprouting assay ( see Methods ) . Treated TEM were applied to HUVEC grown on microcarrier beads and embedded in a fibrin gel to measure their aptitude to induce HUVEC sprouting i . e . the initial step of blood vessel formation . Single treatments show no significant impact on TEM proangiogenic activity relative to untreated cells with the exception of TGF-β which significantly reduced TEM pro-angiogenic activity ( Fig . 2B ) . Interestingly , combining TGF-β with PlGF further decreased VEGFR-1 expression ( Fig . 2A ) and TEM proangiogenic activity ( Fig . 2B ) suggesting that TGF-β synergized with PlGF to reduce TEM proangiogenic activity . We examined the impact of combined treatments on TEM using in vivo corneal vascularization assay . Indeed , in vitro sprouting assay was preferred for quantification but is however less reliable because it does not recapitulate the intricate balance of signals from growth factors , mural cells and extracellular matrix of in vivo angiogenesis . TNF-α in combination with ANG-2 ( or PlGF ) significantly increased TIE-2 expression whilst leaving VEGFR-1 expression unchanged ( Fig . 2A ) , and raised TEM pro-angiogenic activity ( Fig . 2C . Cornea and iris vascularization in AU: control: 1; untreated: 1 . 81; TNF-α+Ang-2: 4 . 58 ) . Conversely , TGF-β in combination with VEGF resulted in a comparable induction of TIE-2 but decreased VEGFR-1 expression ( Fig . 2A ) , and reduced TEM pro-angiogenic activity ( Fig . 2C . Cornea and iris vascularization in AU: TGF-β+VEGF: 1 . 36 ) . Taken together these results show , for the first time , that both Tie2 and VEGFR1 pathways control TEM pro-angiogenic activity . Furthermore , TIE-2 and VEGFR1 pathways synergized with the TNF and TGF pathway to induce and reduce TEM pro-angiogenic activity respectively . Finally , we examined the impact of the different ligand treatments on TEM secretions . Thus , cumulated TEM secretions from ivdTEM were measured experimentally and the secretions for TEM were mathematically inferred ( ivdTEM correspond to double positive DP cell population , see Materials and Methods and S2 Fig . ) and display in Fig . 2D . Of note , none of the single or double treatments we have examined experimentally ( Fig . 2D ) shifted completely the paracrine secretion profile of TEM differentiated in vitro toward that of blood or tumor TEM ( compare Fig . 2D and 1B ) . These results suggest that transitioning ivdTEM into blood or tumor TEM requires a model to simulate computationally the impact of a larger number of ligand combinations on TEM behavior . The limited amounts of patient TEM and the combinatorial nature of the ligands precluded experimental testing of all the ligand combinations and was the rationale for building an integrative and predictive model of TEM behavior . We used TEM differentiated in vitro to derive a dynamical regulatory network from experimental data obtained with a selected number of ligands ( Fig . 2 ) and used then as a proxy to assess the clinically most relevant ligand combinations . To create the models , data sets of receptor expression ( Fig . 2 and S2 Table ) and paracrine secretion profiles ( Fig . 2 and S3 Table ) were combined to infer relevant relationships ( or links ) between ligands and receptors . Briefly , relevant links were identified based on the amplitude of their expression or secretion changes , their reproducibility , and their coherent variations across the treatments ( see Methods ) . Based on these criteria , amongst 924 possible links ( 7 receptors × 11 secreted factors × 12 treatments ) we retained 74 relevant links ( S4 Table ) . Globally , TNF-α , TGF-β and PlGF appeared as key regulators of TEM network . However , TNF-α in contrast to TGF-β , was strongly regulated by other factors ( Fig . 3 ) . Dynamical Boolean modeling was then performed by integrating the retained links into an algorithm for computing Minimal Intervention Set ( MIS ) of TEM regulatory network . Given a regulatory network , MIS patterns represent a set of simultaneous perturbations ( or treatments ) to force the network into a desired steady state , where a subset of nodes remain at a fixed expression level of either low or high [41 , 42] . The term minimal implies that no other sub-set of an MIS pattern can lead to the desired steady state behavior . However , for a given network , there can be more than one MIS patterns to generate the same steady state . The MIS algorithm proposed by Garg et al [43 , 44] was used for assessing TEM regulatory network by computationally predicting all possible set of up to three simultaneous treatments that can force the TEM network into a weakly ( i . e . blood TEM ) or highly ( i . e . tumor TEM ) pro-angiogenic phenotype . Relative to their blood counterparts , tumor TEM display a higher pro-angiogenic activity ( Fig . 1A ) , a paracrine profile shifted toward angiogenesis ( Fig . 1B ) and higher levels of TIE-2 and VEGFR-1 ( Table 2 ) . Therefore blood and tumor TEM can be viewed as two distinct cell steady state behaviors and ivdTEM as an intermediate state ( Tables 2 and 3 ) . Using the regulatory network model of TEM differentiated in vitro we predicted the minimal treatments required for transitioning tumor TEM to blood TEM and vice versa . Because the expression levels of TIE-2 and VEGFR-1 controlled ( Fig . 2 ) and mirrored ( Fig . 1B ) TEM pro-angiogenic activity , we assigned to TIE-2 and VEGFR-1 nodes a fixed polarity of either both over-expressed or down-modulated for highly pro-angiogenic ( i . e . tumor TEM ) or weakly pro-angiogenic ( i . e . blood TEM ) steady states respectively . Computationally predicted minimal perturbations sets ( MIS ) are reported in Table 4 . It is interesting to note that all the predicted treatments were composed of at least two , and mostly three simultaneous perturbations . Only one treatment , combining three perturbations , was predicted by the model to promote TEM pro-angiogenic activity ( TNF-α , ANG-2 and PlGF; Table 4 ) . Conversely , eleven distinct treatments were predicted to dampen TEM proangiogenic activity and resulted in three main groups ( Table 4 ) . The first group of treatments combined TIE-2 tyrosine kinase inhibitor with TGF-β and a ligand of VEGFR-1 or TIE-2 . Treatments from the second group involved VEGFR-1 kinase inhibitor , and the third group of treatments associated TGF-β with TNF-α and a ligand of TIE-2 or VEGFR-1 ( Table 4 ) . It is worth noting here that we assumed that possible compensatory mechanisms resulting from the blocking of the receptor signaling ( rather than knocking down the receptor ) do not significantly affect the angiogenic activity . Results showed that this assumption was valid for the particular case of the receptors under study . With currently available tools , VEGFR-1 kinase activity is almost impossible to manipulate . Indeed , to date , all available VEGFR-1 kinase inhibitors also inhibit VEGFR2 and VEGFR3 to a lesser extent , thus preventing experimental validation of any treatment of the second group . For experimental validations , we therefore selected the combined TNF-α/ANG-2/PlGF treatment , predicted to promote angiogenesis , two treatments of the first group ( TIE-2inhibitor/TGF-β/PlGF , and TIE-2 inhibitor/TGF-β/ANG-2 ) , one treatment of the third group ( PlGF/TGF-β/TNF-α ) and a TIE-2 kinase inhibitor alone . These experimental validations were first conducted in TEM differentiated in vitro . As predicted , the TNF-α/ANG-2/PlGF combined treatment induced TIE-2 and VEGFR-1 expression ( Fig . 4A ) and increased their proangiogenic activity ( Fig . 4B ) . Importantly , this combined treatment induced TIE-2 and VEGFR-1 expression and TEM pro-angiogenic activity more efficiently than PlGF/TNF-α ( Fig . 4A ) and PlGF or ANG-2 single treatments ( Fig . 2A and B ) . These results validate our in silico prediction and reveal the synergistic effect of TIE-2 , VEGFR-1 and TNF-α pathways in controlling TEM pro-angiogenic activity . The predicted inhibitory effect of the other treatments was assessed on TEM pre-treated with TNF-α/ANG-2/PlGF , which display an increased pro-angiogenic phenotype compared to untreated cells ( Fig . 4A and B ) . This pre-treatment increased the dynamic range and therefore the sensitivity of detecting inhibitory effects . TIE-2 kinase inhibitor/TGF-β combined with ANG-2 or PlGF significantly decreased TIE-2 and VEGFR-1 receptor expression ( Fig . 4A ) consistently reduced their pro-angiogenic activity ( Fig . 4B ) . The PlGF/TGF-β/TNF-α treatment was not as effective , but still reduced their pro-angiogenic activity . These combined treatments were synergistic and minimal since TIE-2 kinase inhibitor ( Fig . 4A and B ) or single treatments alone ( Fig . 2A and 2B ) or double treatments ( PLGF/TGF-β or TIE-2inhibitor/TGF-β display on Fig . 2B and Fig . 4B , respectively ) were not sufficient to decrease VEGFR-1 expression and TEM pro-angiogenic activity . By contrast , PlGF/TGF-β/TNF-α heterogeneously decreased TIE-2 and VEGFR-1 expression ( Fig . 4A ) and TEM pro-angiogenic activity ( Fig . 4B ) . In summary , from these validation experiments we found that the best computationally predicted treatment promoting TEM pro-angiogenic activity was TNF-α/ANG-2/PlGF and the best dampening activity was found using TIE-2 kinase inhibitor/TGF-β associated with a ligand of TIE-2 or VEGFR-1 . Having identified the critical ligands and pathways controlling TEM plasticity , we next examined in TEM differentiated in vitro whether differential gene expression might also contribute to the molecular basis of TEM plastic behavior . This analysis may shed light on the molecular mechanisms underlying the observed TEM responses . To this end , we selected VEGF/TNF-α , ANG-2/TGF-β and PlGF/TGF-β treatments for gene expression profiling using Affimetrix whole genome microarrays , because these treatments were present in 17 , 16 and 14 , respectively of the 74 links ( treatment/receptor/cytokine ) retained in TEM regulatory network ( S4 Table and Fig . 3 ) . All the other treatments occurred less frequently . Hierarchical clustering demonstrated that TGF-β-based treatments ( ANG-2/TGF-β and PlGF/TGF-β ) clustered separately from VEGF/TNF-α and control treatments . A total of 398 genes were significantly ( p<0 . 05 ) and differentially expressed between the two clusters among which 369 and 72 genes were altered by TGF-β/ANG-2 and TGF-β/PlGF treatments respectively ( S5 Table , NT unique lists ) while 43 were regulated in common ( S5 Table , NT intersect list ) . Enrichment analyses of the gene expression data against known pathways and functional gene categories were conducted as described in Materials and Methods . No enrichment of specific pathways of interest was observed due to the fact that the gene annotations were too general and did not correspond to specific functions of monocytes . Therefore , the 398 differentially expressed genes were annotated and classified in categories manually ( S5 Table ) . Similar expression profiles were obtained for untreated and TNF-α/VEGF treated cells consistent with their weak impact on TEM functional angiogenic phenotype ( Fig . 2 ) . By contrast , ANG-2/TGF-β and PlGF/TGF-β treatments inhibited TEM pro-angiogenic activity ( Fig . 2 ) and down-modulated the expression of pro-angiogenic genes ( Fig . 5 and S5 Table ) . Furthermore and interestingly , the expression of VASH1 ( vasohibin 1 ) and UCN ( urocortin ) genes coding for anti-angiogenic proteins was simultaneously up-regulated ( Fig . 5 and S5 Table ) . In response to both ANG-2/TGF-β and PlGF/TGF-β treatments , 95% of the genes functionally related to the cell cycle displayed a down-modulated expression indicating that TEM stopped proliferating with profound changes in their metabolism but without , however undergoing apoptosis ( the expression of metabolism and apoptosis related genes was down-modulated for 76% and 88% of them respectively ) . TEM treated with TGF-β/ANG-2 or TGF-β/PlGF show the expression of some genes ( P2RY12 , TMCC3 , NPDC1 , IFFO1 , UBASH3B , C11orf52 , SLC4A7 , TMEM87A , NPL , EMB , PCNA , DNA2 , TMEM86A , MMP12 , CTSD , AXL , RASGRP3 , TUBB , FCGR1A , CR1 , MX2 ) previously ascribed to mouse TAM [45 , 46 , 47] . However , in response to ANG-2/TGF-β and PlGF/TGF-β treatments , TEM down-modulated the expression of genes involved in macrophage differentiation ( Fig . 5 ) and started to acquire the profile ( RGS1 , CXCL11 , CXCL9 , STAT1 , IFIH1 , ISG20 , NT5C3 , ADC , PDGFRL , TNF-ASF12 , IFIT5 , RGS10 , TRAF3IP3 , CIDEB , APOBEC3A , PYGL , RRM1 , MAF , NLRC4 , IL10 , MYC , DUT , POLE4 , CXCL17 ) of dendritic cells matured in vitro by exposure to lipopolysaccharide and interferon-gamma [48 , 49] . Along these lines , genes encoding for dendritic cell markers , antigen processing and adaptive immune response were upregulated while genes involved in immune suppression show markedly decreased expression ( Fig . 5 and [38] ) . Finally the expression of genes related to adhesion and migration were up- and down-regulated respectively indicating that TEM mobility was strongly reduced; an observation consistent with the arrest of their cell cycle and the alteration of their differentiation program ( Fig . 5 and S5 Table ) . Along these lines , we observed experimentally that ivdTEM treated with PlGF/TGF-β/TIE-2i display reduced mobility towards the human epithelial tumor cell line MDA-231 ( S3A Fig . ) and slowed down the growth of MDA-231 cells ( S3B Fig . ) Taken together , our results suggest that ANG-2/TGF-β and PlGF/TGF-β treatments are not only anti-angiogenic but also shift the gene expression profile of monocytes toward the one of cells promoting immune surveillance , thereby limiting tumor growth . We next sought to validate the computationally predicted treatments in TEM isolated from patient breast carcinoma . Tumor TEM were exposed to TIE-2 kinase inhibitor combined with TGF-β and simultaneously engaged their VEGFR-1 using VEGF ( alternatively PlGF , Table 4 and Fig . 4 ) . This combined treatment strongly reduced the pro-angiogenic activity of tumor TEM in the mouse cornea vascularization assay ( Fig . 6A and B ) and decreased the expression of TIE-2 and VEGFR-1 ( Fig . 6D ) . Furthermore , this treatment reduced the secretion of IL-6 , IL-8 , MMP9 , bFGF and VEGF , consistent with a paracrine profile shifted toward a M1-like phenotype and closer to the one of blood TEM ( Fig . 6C and 1B ) . Conversely , TEM from patient blood exposed to the combined treatment of TNF-α/PlGF/ANG-2 increased their pro-angiogenic activity in the mouse cornea vascularization assay ( Fig . 6B ) and was associated with significantly higher secretion of IL-1β , IL-6 , IL-10 , MMP9 and VEGF ( Fig . 6C ) and increased expression of TIE-2 and VEGFR1 ( Fig . 6D ) . These results highlighted the validity of our combined experimental and computational approach to revert the pro-angiogenic phenotype of TEM and revealed , for the first time , that tumor TEM remain plastic cells representing attractive targets for anti-angiogenic therapies . We addressed the question whether or not the expression levels of ANG-2 and PIGF when considering the overall breast tumor have impact on the survival ( considered here as relapse free survival ) . To this end we analyzed a dataset including tumor expression profiles and clinical data of 1809 breast cancer patients [50] and compared two subsets of patients: those with lowest and highest expression values for ANG-2 , PIGF and CD14 ( as TEM marker ) , using as threshold the first and fourth quartile respectively . These quartiles were computed independently for each gene , and the two groups of selected patients resulted from the intersection of them all ( Fig . 7B-F ) . The Kaplan-Meier plot showed a clear separation between patients with low ( n = 40 ) and high ( n = 62 ) expression for these three genes , with a p-value of 0 . 0257 derived from log-rank analysis ( Fig . 7B and F ) . Interestingly , we observed that the same analysis repeated for patients with high and low levels of ANG-2 and CD14 or PIGF and CD14 ( and not for the remaining gene ) resulted on p-values not statistically significant ( 0 . 0587 and 0 . 521 respectively , Fig . 7C-E ) , suggesting that the synergistic effect of the corresponding pathways is required to have a significant impact on the survival . These results suggest that TEM infiltrating a tumor microenvironment enriched in Ang-2 and PlGF , which synergistically trigger TEM angiogenic activity through Tie-2 and VEGFR-1 ( Fig . 4 and 5 ) , may contribute to a worse patient survival . Further , tumor size correlated positively with the amounts of PlGF and Ang-2 content in the tumor microenvironment ( Fig . 7A , P<0 . 01 ) while no significant correlation was observed with VEGF , Ang-1 , MCP-1 , SDF-1 , TGF-α and TNF-β . Moreover , we measured by reverse phase protein arrays that in tumors the extent of TEM infiltration was significantly and linearly correlated with PLGF content ( Fig . 7A ) thus highlighting that Tie-2 and VEGFR-1 axes , as well as their cognate angiogenic TEM ligands Ang-2 and PlGF represent attractive therapeutic targets in breast cancer .
The key relevance of this study is a comprehensive understanding of the behavior of TEM in breast tumor vascularization . This goal was achieved by constructing an integrative and predictive model of TEM behavior based on experimental data . This model was interrogated to identify combined treatments that would alter TEM pro-angiogenic activity . Quite remarkably , four of the five predicted combined treatments that we validated experimentally proved to be extremely efficient at inhibiting or promoting tumor TEM proangiogenic activity , demonstrating the robustness of our model . Furthermore , this study demonstrates that the synergistic effect of these treatments relies on crosstalk between TNF-R1 , VEGFR-1 , TGF-β and TIE-2 pathways resulting in altered angiogenic activity ( Figs . 2 , 4 and 5 ) , modulated expression of angiogenic receptors ( Fig . 4A ) and shifted paracrine profile ( Fig . 6C ) . Taken together , our results highlight crosstalks between TIE-2 , VEGFR-1 , TGF-β and TNF-α pathways of outstanding importance to promote ( TNF-α/ANG-2/PlGF ) or abrogate ( TGF-β/TIE-2 inhibitor/VGFR1 or TIE-2 ligand ) patient TEM pro-angiogenic activity . Another contribution of this study is an effective approach to model relatively sparse data from distinct individuals ( newborns and patients ) , who are inherently heterogeneous in nature . This challenge was overcome by a sustained and tight collaboration between the experts in the fields of computational and experimental sciences throughout all steps of the workflow ( Fig . 1C ) . By setting up a rigorous experimental design we identified coherent variations and links across biological replicates and data sets , which provided a robust basis to reconstruct the TEM signaling network ( Fig . 3 ) . Furthermore , the modeling framework was an integral part of our experimental strategy , enabling the model predictions to address the biological questions , an issue that is of particular importance in systems biology [1 , 51 , 52] . In a traditional approach , it would have been unfeasible to experimentally test the complete set of up to three simultaneous perturbations using 12 distinct ligands , which would have led to 596 ligand combinations . The physiologically relevant combinations of ligands were discovered by applying the recently proposed MIS algorithm [43 , 44] to predict all minimal perturbations in the inferred regulatory network that can transition TEM into desired steady states ( Table 4 ) . The in silico minimal perturbations predicted by applying the MIS algorithm on the inferred ivdTEM regulatory network comprised not only a handful of the set of perturbations ( or ligand combinations ) and they were all shown to be experimentally valid when tested on ivdTEM and patient TEM ( Figs . 4 and 5 ) . The in silico prediction algorithm helped us to focus on the most clinically relevant monocytic ligands and to unravel treatments abrogating TEM pro-angiogenic activity at breast tumor sites . These results highlight the importance of mutual relationship between experimental and computational sciences . Furthermore , the combined computational and experimental approach followed in this study may provide a general strategy to study the behavior of limited cell subsets from patients in cancer and other diseases . The main outcome of this modeling strategy for experimental and clinical oncology is the validation of treatments abrogating tumor TEM pro-angiogenic activity and thus simultaneously revealing their functional plasticity . Our study shows that treatments targeting TEM plasticity may constitute a valid therapeutic strategy to shift TEM to acquire a more anti-tumor M1-like phenotype . Moreover , the relapse free survival analysis showed a clear separation between patients with low and high expression for pro-angiogenic genes ( ANG-2 and PIGF ) , and suggested that the synergistic effect of the corresponding pathways is required to have a significant impact on the survival . Overall , our results obtained by employing the combined modeling and experimental approach suggest novel treatments for abrogating tumor TEM pro-angiogenic activity and reveals the functional plasticity of TEM .
This study was approved by the ethics committee of the University Hospital of Lausanne ( reference number 170/07 ) . Patient or subject tissue specimens were obtained according to the declaration of Helsinki and upon written informed consent . A series of 40 primary invasive breast carcinoma specimens ( Table 1 ) were resected from patients with breast cancer and tumors enzymatically dissociated as described [53] . All patients underwent surgery and sentinel node biopsy before treatment . The presence of nodal metastases and tumor pathological features were confirmed histologically and are detailed in Table 1 . All patients were untreated before surgery . Peripheral blood was collected before surgery and processed as described [54] . CD34+ hematopoietic progenitors from cord blood were isolated by immunomagnetic selection ( StemCell Technologies Inc . ) and cultured as previously described [40] . At day 6 , cells were activated for 2h at 37°C with 100 ng/ ml of recombinant ligands ( TNF-α 20 ng/ ml ) , washed and cultured for 36h in RPMI containing 10% FCS . When two treatments were successively applied , TEM were first exposed for 2h to a combined treatment of PlGF/TNF-α/ANG-2 , washed and kept in culture for 30h in RPMI containing 10% FCS . They were then exposed to inhibitory treatments for 2h , washed and kept 24 hours longer in culture . TEM phenotype , cytokine secretion and pro-angiogenic activity were assessed by flow cytometry and in vivo or in vitro vascularization assay , respectively . Alternatively TEM were isolated from patient peripheral blood or tumor by CD14 immunomagnetic selection , exposed to treatments for 36h ( 30h exposure to angiogenic and inflammatory factors followed by 6h exposure to TIE-2 kinase inhibitor at 8 μM ) and extensively washed . TEM viability was not affected under the different conditions of stimulation used and was > 95% We undertook a rigorous experimental design consisting in profiling changes in phenotype , cytokine secretion , gene expression and angiogenic activity from the same cell sample in response to treatments . Changes induced by the treatments were normalized to untreated cell in each biological replicate and cumulated across the treatments . In this study , a biological replicate was a population of ivdTEM from a distinct cord blood and exposed to the same treatment . Hence , biological replicates originated from distinct individuals ( newborn or patients ) . This experimental design was kept rigorously for each biological replicate to allow analysis across biological replicates and data sets . Following blocking of Fc receptors with antibodies , cells were labeled with CD14 ( PerCP-Cy5 . 5 ) , CD11b ( FITC ) , TIE-2 ( Alexa 647 ) , VEGFR-1 ( PE ) , TGFBR-1 ( Pacific Blue ) , TNF-R1 ( Pacific Orange ) , CXCR4- , CCR5- , α5β1-biotinylated specific antibodies ( followed by streptavidin-Marina Blue ) and analysed by flow cytometry using a Facs LSRII ( BD Biosciences ) equipped with a 610/20 nm filter on the violet detector . Pacific Blue and Pacific Orange NHSE ( Invitrogen ) were used to couple TGFBR-1 and TNF-R1-specific antibodies respectively as well as the corresponding isotype controls . The cell populations were manually examined based on their CD14 and CD11b intensities to identify DN , SP and DP cell populations and the frequency count and a mean intensity value for each channel were calculated . Secreted cytokines and angiogenic factors were quantified in cell conditioned medium using FlowCytomix technology ( Bender MedSystems and RnD ) . Importantly , VEGF and PlGF were used as treatments ( see in vitro TEM differentiation and treatments above ) and also measured in conditioned medium as angiogenic factors secreted by TEM in response to the treatments ( this section ) . Mouse experiments were approved by the veterinary service of Vaud Canton . The bacterial lipopolysaccharide membrane receptor CD14 is a component of the innate immune system mainly expressed by monocytes and macrophages and commonly used as a marker of these cell populations . Monocytes were isolated by CD14 immunomagnetic selection from patient tissue . For in vivo corneal vascularization assay , 20 , 000 CD14+ cells isolated by positive immunomagnetic selection ( Stemcell Technologies ) from peripheral blood ( purity>95% ) or dissociated tumors ( purity>85% with no detectable CD45- contamination ) were injected ( 5μl ) into the stromal part of the corneas of anesthesized NOD-scid IL2Rγnull mice [39] using a 35 gauge nanofil injection kit ( WPI , Stevenage , UK ) . Cornea vascularization was monitored with a digital stereomicroscope ( Leica ) . Mice were euthanized 25 days post-injection and isolated eyes were fixed in 4% PFA , cryoprotected in a 30% sucrose solution and embedded in Yazulla media ( 30% egg albumin , 3% gelatin ) . Vascularization was assessed by immunostaining of the sagittal sections ( 10μm ) with CD31-specific antibodies ( Platelet Endothelial Cell Adhesion Molecule-1 , PECAM-1 ) using a Zeiss motorized Axio Imager M1 fluorescent microscope . Retina was used as a positive control in all CD31 stainings . Quantification was performed with Image J software by measuring the fraction of the iris and cornea surface area containing vessels ( CD31 positive surface areas of the iris and cornea/ surface area of the iris and cornea ) . This ratio was set up at 1 for the control eyes ( no cell injected ) and the data were normalized to the control ( AU ) . In vitro angiogenesis sprouting assay was performed with HUVEC spheroids as previously described [40] . The corneal angiogenesis assay is still considered one of the best in vivo assays [55] . However , the surgical procedure is technically difficult and the assay time consuming . Therefore , we use in vitro angiogenesis sprouting assay [56] to assess the impact of multiple treatments and we validated the most relevant one in vivo . Chemicals unless indicated otherwise were from Sigma-Aldrich . Common stocks of cytokines , inhibitors and assay reagents were used to minimize experimental variability . Human recombinant cytokines were purchased from PeproTech ( London , UK ) and R&D Systems . All the antibodies used are listed in S1 Table . TIE-2 kinase inhibitor compound 7 was from Alexis Biochemicals ( San Diego , CA ) . Statistical analysis was performed using GraphPad Prism version 4 . 00 for Windows , GraphPad Software , San Diego , California , USA . Unless indicated differently , T test was used to determine p values . A p value < 0 . 05 was considered statistically significant . All data shown are means ± standard deviation . The effect of each treatment on the phenotype and secretions of TEM was calculated as the log2 of the mean fluorescence intensity ( MFI ) percent change compared to untreated cells and a heatmap produced with R ( http://www . R-project . org ) . Cumulated secretions from DN , SP and DP ( i . e . ivdTEM ) were measured experimentally and the secretions of each population mathematically inferred . For a given treatment , let Na , Nb and Nc be the relative number of cells present in each population ( a = DN , b = SP , c = DP ) and K be the amount of cytokine experimentally measured and expressed as a percentage of change of cytokine secretion to untreated cells . Then the relative contribution C of each population was expressed as: Ca × Na + Cb × Nb + Cc × Nc = K Using three independent biological replicates ( 1 , 2 , 3 ) and their associated cytokine measurements ( K1 , K2 , K3 ) we could write the following three equations: Ca × Na1 + Cb × Nb1 + Cc × Nc1 = K1 Ca × Na2 + Cb × Nb2 + Cc × Nc2 = K2 Ca × Na3 + Cb × Nb3 + Cc × Nc3 = K3 The unknowns were Ca , Cb , and Cc while Na , Nb and Nb have been obtained from cytofluorimetry data . These equations were solved for all possible combinations of 3 biological replicates ( 4 to 84 ) and the median of the obtained C coefficients calculated . The coefficients were then used to infer the amount of cytokine by DN , SP and DP cell populations . In the few instances , the inferred cytokine amount was lower than 0% of untreated cells and the predicted value was set to a minimal positive value ( 1% ) . Within the TEM regulatory network , a link represents an effect ( increase or decrease ) on either receptor expression or cytokine secretion in response to single or combined ligands ( Fig . 3 ) . Of note , some treatments and secreted cytokines are identical . We applied the three following criteria to identify a link between treatment/receptor/cytokine . We retained only the links that are reproducible , of sufficient amplitude and coherent . Only 8% of the possible links matched these three criteria and were retained to construct the TEM regulatory network . A link was considered as reproducible when the treatment induced a reproducible effect across at least 3/4 of the biological replicates . Second , a treatment was retained if it induced a change of sufficient amplitude , i . e . when the effect of the treatment on receptor and cytokine was among the upper ( the treatment increases the expression of a receptor or a cytokine ) or lower ( the treatment decreases the expression of a cytokine or a receptor ) quartile of variation . Third , only coherent links were retained meaning that they should be always correlated or anti-correlated across the treatments , and not sometimes correlated and other times anti-correlated . As an example , a receptor A and a cytokine B are linked when they show similar or/and opposite variations across the treatments e . g . receptor A up-regulated ) / cytokine B up , and/or receptor A down/cytokine B down . The list of the retained links is provided in S4 Table . The interactions between treatment/receptor/cytokine as predicted by our experimental and computational approach ( see previous section ) were used to generate the dynamical regulatory network of ivdTEM ( Boolean equations are provided in S6 Table ) . The MIS algorithm proposed in [43 , 44] was then applied to compute all possible minimal perturbation sets to force the network into desired steady state or phenotype . The MIS algorithm starts by unrolling the inferred regulatory network of ivdTEM into a tree-like structure starting from the nodes which have a fixed polarity ( i . e . either high or low ) in the desired final steady state . In the angiogenesis model of ivdTEM , the nodes corresponding to TIE-2 and VEGFR-1 are assigned a fixed polarity of either both high or both low for highly pro-angiogenic and weakly pro-angiogenic steady states respectively . The nodes with the fixed polarity are referred to as the root nodes of the network . The network is unrolled along a path in the regulatory network until a duplicate node is found . At that instance the unrolling process is terminated along this path . Once the network is unrolled along all paths originating from the root node , the MIS patterns are generated by scanning this unrolled network in two iterations . In the first iteration , the required polarity ( i . e . over-expression or knock-down ) of each node is propagated from the root node to the leaf nodes . In the second iteration , possible perturbations that can lead to the required polarity at each node are listed by scanning the unrolled network in the reverse order from the leaf nodes towards the root node in the breadth first manner . When two paths merge at one node , then only those MIS vectors that are compatible ( i . e . the same node is not over-expressed on one path and knocked-down on the other path ) along both the paths are taken into consideration while scanning the rest of the network . This process when terminates at the root node , only MIS vectors that have been compatible throughout the unrolled network shall remain in the list . The MIS algorithm does not involve explicit enumeration of perturbation patterns but rather generates these patterns by traversing the topology of the network ensuring that only the patterns leading to a desired cellular behavior are generated . Detailed methodology for generation of the MIS vectors and the network unrolling process are further described in [43 , 44] . Total RNAs from 100 000 monocytes were isolated and purified with the Qiagen RNeasy micro plus kit . RNA samples were hybridized to Affymetrix Human Gene 1 . 0 ST Arrays and images were processed to obtain probe intensities using standard procedures at the GTF ( Gene Technology Facility , CIG , University of Lausanne ) . Background subtraction , RNA normalization and probeset summarization were performed using the Affymetrix Power Tools software package ( Affymetrix CEL files ) . Sample correlation was performed on the top 1000 expressed probesets using Bioconductor affy and affyPLM packages in R [57] . This analysis indicated separate clustering of VEGF/TNF-α and ANG-2 or PIGF/TGF-β samples . Differentially expressed genes between different treatments were detected by fitting linear models and computing empirical Bayes moderated t statistics , comparing two groups at a time , using the limma package in R [58] . P values were adjusted for multiple comparisons using the Benjamini Hochberg procedure [59] and genes with an adjusted p value of < = 0 . 05 were selected as differentially expressed . For pathway analysis , differentially expressed genes were ranked according to fold change ( high to low ) comparing two treatments and Gene Set Enrichment Analysis ( GSEA ) was performed on the ranked lists against MSigDB gene sets using the NCBI gene id as a unique identifier [60] . Enrichment p values were adjusted for multiple comparisons using the Benjamini Hochberg procedure [59] . The microarray data from this publication have been submitted to the GEO database http://www . ncbi . nlm . nih . gov/geo/info/linking . html and assigned the identifier GSE34559 . Publicly available normalized expression data from 1809 breast cancer patients was downloaded from http://kmplot . com [50] . For the relapse free survival analysis we selected lowest and highest expression values of 205572_at , 209652_s_at and 201743_at probes , corresponding with ANG-2 , PIGF and CD14 genes respectively , using as threshold the first and third quartile respectively . These quartiles were computed independently for each gene , and the two groups of selected patients resulted from the intersection of them all ( see Fig . 7 ) . To generate the Kaplan-Meier plots and to evaluate the separation between groups ( log-rank statistic ) we used the survival package in R . Tables with the resulting data and R scripts are included in supplementary table 7 . | Tumor vascularization is essential for tumor growth and cancer progression . In breast cancer , monocytes are angiogenic , i . e . able to induce tumor vascularization . In patients , blood circulating monocytes drastically increase their angiogenic activity when reaching the tumor , suggesting that the tumor microenvironment shapes their angiogenic activity . The identification of the tumor signals inducing the angiogenic activity of monocyte is of paramount significance because it represents the rationale for anti-angiogenic therapies in breast cancer . This goal was achieved by constructing an integrative model of monocyte behavior based on experimental data . The model predicted treatments abrogating the angiogenic activity of monocytes , which were experimentally validated in monocytes isolated from patient breast carcinoma . Importantly , these treatments reverted angiogenic monocytes into immunological potent cells . The main outcome of this modeling strategy for experimental and clinical oncology is the identification of effective treatments abrogating the angiogenic activity of monocytes and thus simultaneously revealing their functional plasticity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Angiogenic Activity of Breast Cancer Patients’ Monocytes Reverted by Combined Use of Systems Modeling and Experimental Approaches |
Drosophila leg morphogenesis occurs under the control of a relatively well-known genetic cascade , which mobilizes both cell signaling pathways and tissue-specific transcription factors . However , their cross-regulatory interactions , deployed to refine leg patterning , remain poorly characterized at the gene expression level . Within the genetically interacting landscape that governs limb development , the bric-à-brac2 ( bab2 ) gene is required for distal leg segmentation . We have previously shown that the Distal-less ( Dll ) homeodomain and Rotund ( Rn ) zinc-finger activating transcription factors control limb-specific bab2 expression by binding directly a single critical leg/antennal enhancer ( LAE ) within the bric-à-brac locus . By genetic and molecular analyses , we show here that the EGFR-responsive C15 homeodomain and the Notch-regulated Bowl zinc-finger transcription factors also interact directly with the LAE enhancer as a repressive duo . The appendage patterning gene bab2 is the first identified direct target of the Bowl repressor , an Odd-skipped/Osr family member . Moreover , we show that C15 acts on LAE activity independently of its regular partner , the Aristaless homeoprotein . Instead , we find that C15 interacts physically with the Dll activator through contacts between their homeodomain and binds competitively with Dll to adjacent cognate sites on LAE , adding potential new layers of regulation by C15 . Lastly , we show that C15 and Bowl activities regulate also rn expression . Our findings shed light on how the concerted action of two transcriptional repressors , in response to cell signaling inputs , shapes and refines gene expression along the limb proximo-distal axis in a timely manner .
In developing arthropod and vertebrate appendages , morphogen gradients play critical roles in instructing the spatially-restricted expression of patterning genes that mostly encode transcription factors ( TFs ) [1 , 2] . However , how their expression domains are set up precisely and how their cross-regulation refines pattern layouts , remain to be deciphered , particularly by characterizing the cis-regulatory modules within the gene regulatory networks governing limb development . The Drosophila leg provides a paradigm with which to tackle the issue of the molecular mechanisms underlying proximo-distal limb development [2–4] . The primordia of the adult appendages , the leg imaginal discs , originate from clusters of roughly 20–30 embryonic ectodermal cells which proliferate during the three larval stages . During the first and second instar larval stages , in response to antagonistic wingless ( wg ) and decapentaplegic ( dpp ) cell signaling pathways , the TF-encoding Distal-less ( Dll ) , dachshund ( dac ) and homothorax ( hth ) genes are activated in broad concentric domains within the leg disc , that prefigure the distal , medial and proximal subdivisions of the adult appendage , respectively [3] . By the end of the third instar larval ( L3 ) stage , the leg disc is highly folded presaging movements towards a tri-dimensional structure . The distal portion of the adult leg comprises the tarsus ( ts ) , which is divided into five segments ( ts1-5 , from proximal to distal ) , and the pretarsus ( pt ) , which is characterized by a pair of terminal claws . Around the early L3 stage , ~72–80 hours ( h ) after egg laying ( AEL ) , the combinatorial activities of the leg “gap” genes Dll , dac and hth induce the expression of the tarsal-specific rotund ( rn ) , bric-à-brac ( bab ) complex ( bab1 and bab2 ) and Bar complex ( BarH1 and BarH2 ) genes in distal circular domains [5–8] . At 80-82h AEL , secreted Wg and Dpp molecules jointly activate the epidermal growth factor receptor ( EGFR ) signaling pathway in distalmost pretarsal cells [5 , 7] . In turn , EGFR cell signaling activates the expression of pretarsal patterning genes , including the homeobox genes aristaless ( al ) , dlim1 and C15 ( also known as clawless ) , and spatially extinguishes expression of tarsal patterning genes , including bab and Bar complex loci [5 , 7–10] . As for leg gap genes , most pretarsal and tarsal patterning genes encode DNA-binding transcriptional regulators , expressed either in the very center of the leg disc ( e . g . , C15/clawless and al ) or in concentric rings ( e . g . , bab2 , BarH1 and rn ) [2 , 4] . Lastly , at mid-late L3 stage , activation of the Notch ( N ) cell signaling pathway through the restricted expression of the Delta ( Dl ) and Serrate ( Ser ) ligands ( in response to leg gap combinatorial activities ) is required for formation of the flexible joint connecting each leg segment , through induction of a set of target genes , including the related genes odd-skipped ( odd ) , drumstick ( drm ) , bowl and sister-of-bowl ( sob ) [4 , 11 , 12] . Dl and Ser expression boundaries are maintained by a negative feedback loop between N signaling and the Lines/Bowl pathway [13] . In Dl expressing cells , the Lines protein is nuclear , leading to specific degradation of the Odd-family zinc-finger ( ZF ) protein Bowl [13–15] . In the adjacent joint cells , N signaling induces the expression of redundant Odd-family ZF proteins ( Odd , Sob and Drm ) , which specifically re-localize Lines to the cytoplasm , thus preventing Bowl destabilization; consequently , Bowl TF accumulates in the nuclear compartment where it down-regulates Dl expression [13] . This feedback mechanism reinforces a Dl+/Dl− boundary , ensuring maintenance of N signaling through joint morphogenesis . Several genes required for distal leg segmentation have altered expression in bowl mutant cells , such as bab2 [16] , but none have been shown so far to be under the direct control of the Bowl transcription factor . The bric-à-brac complex is formed by the duplicated paralogous genes bab1 and bab2 , encoding BTB/POZ domain TFs with overlapping functional roles in several developmental processes [17 , 18] . While bab1 is expressed similarly to bab2 , only the latter is critically required for distal leg and antennal segmentation [18] . The bab2 gene displays dynamic expression in a restricted proximo-distal ( P-D ) expression pattern in the distal leg and antennal primordia [17–20] . Initially expressed as a broad circle within the Dll-expressing distal domain at early-mid L3 stage , the bab2 pattern in late-L3-stage leg and antennal discs resolves to four or two concentric rings , respectively [20] . Later on , at the pupal and adult stages , a P-D graded pattern is observed for each bab2-expressing ring , which is essential for ts2-5 and antennal ( a ) a3-5 segment joint formation [18] . While many DNA-binding factors have been identified , little is known about their molecular relationships within the gene regulation landscape governing limb formation along the P-D axis . To this aim , we previously identified within the bric-à-brac locus a single cis-regulatory module that reliably governs bab2 expression in developing appendages , termed LAE for leg/antennal enhancer , and showed that the Distal-less homeodomain and Rotund zinc-finger TFs interact directly with discrete critical sites within the LAE enhancer to activate or up-regulate bab2 in all or specifically in the proximalmost expressing cells , respectively , within the developing distal leg and antenna [19] . Here , we show that the C15 homeodomain and Bowl zinc-finger proteins act jointly by binding directly conserved LAE sequences , to restrict bab2 expression within the developing leg , in response to cell signaling . We show that C15 competes with the Dll activator for LAE binding on neighboring cognate sites . Furthermore , instead of engaging direct partnership with the Aristaless homeoprotein , we find that C15 interacts physically with Dll through contacts between their homeodomain . Taken together with our previous data , this study provides a deeper understanding of how a transcriptional enhancer integrates diverse repressive cell signaling inputs produced along the proximo-distal axis of the developing limb .
Limb-specific bab2 expression is governed by a single 567-bp cis-regulatory module ( CRM ) , termed LAE [19] . Among the three evolutionarily-conserved regions ( CR1-3 ) comprised in this CRM , CR1 is critical for enhancer activity , notably through Dll activator binding [19] . Here , we have investigated the functional role of the 41-bp CR2 sequence . To this end , expression of an internally-deleted LAE construct ( LAEΔ2-GFP ) ( Fig 1A ) was compared to expression of a wild-type ( wt ) LAE reporter ( LAEwt-RFP ) , faithfully reflecting the endogenous bab2 gene [19] . Prominent ectopic expression from the LAEΔ2-GFP construct was detected at the proximal ( ts1 ) and distal ( ts5 ) edges of the RFP ( bab2 ) -expressing tarsal domain ( compare Fig 1B to 1C and 1D , white brackets ) , indicating a requirement for CR2 in defining bab2 expression limits . However , GFP expression was absent from developing pretarsal cells , indicating that CR2 is dispensable for EGFR-signaling mediated bab2 extinction there . We conclude that within the LAE enhancer , the conserved CR2 sequence is absolutely required for repressing LAE reporter ( bab2 ) expression along the proximo-distal axis of the leg , to delimit precisely the tarsal expression borders . The lines/bowl gene cassette has been proposed to regulate bab2 expression along the leg P-D axis [13 , 15 , 16] . Indeed , tarsal LAE-GFP expression is flanked both proximally and distally by Bowl-expressing domains ( Fig 1E ) . In contrast , LAEΔ2-GFP is up-regulated in Bowl-expressing cells at the ts5-pretarsal boundary ( S1 Fig , compare panels A to B ) . Further , CR2 contains a consensus Bowl binding site ( see Fig 2A ) . We thus tested the functional relationships between bowl and the LAE enhancer in loss- ( lof ) and gain-of-function ( gof ) clonal analyses in the leg disc . As for endogenous bab2 [13] , LAEwt-RFP expression was cell-autonomously de-repressed in bowl-/- mitotic clones situated both proximally and distally within the developing distal leg ( Fig 1F; clones are circled with white dashed lines ) . Ectopically-expressed Bowl protein accumulates poorly within nuclei of developing tarsal cells due to instability induced by nuclear Lines protein , a dedicated Bowl antagonist specifically expressed there [13 , 16] . bowl gain-of-function was therefore obtained through RNA interference ( RNAi ) coupled to “flip out” ( FO ) Gal4 expression ( see Materials and methods ) to achieve clonal down-regulation of lines [13] . Endogenous bab2 [13 , 15] and LAEwt-RFP expression were both autonomously repressed in most lines-deficient FO cells ( GFP+ ) generated within the developing leg ( Fig 1G; a large clone is circled ) . Our data indicate that LAE enhancer activity is regulated by the Lines/Bowl pathway , raising the possibility that repressive Bowl transcription factor might bind directly to LAE sequences . Bowl belongs to the Odd family of zinc-finger TFs , and closely-related consensus DNA binding motifs have been recently defined for Odd [5’- ( A/C ) CAGTAGC] and Bowl [5’- ( C/A ) C ( A/G ) G ( T/A ) AG ( C/T ) ] ZF domains by a bacterial one-hybrid approach [21] . Two sequences matching consensus binding sites ( BS ) for Odd/Bowl are present within the LAE enhancer ( Fig 2A ) , and both are conserved among 24 available Drosophilidae bric-à-brac locus sequences ( S2A and S2B Fig , respectively ) , one being precisely a part of the repressive CR2 sequence acting along the leg P-D axis ( see Fig 1C and 1D ) . In addition to Drosophilidae species , the CR2 consensus site is conserved also in the Mediterranean fruit fly Ceratitis capitata , the tsetse fly Glossina morsitans and the domestic house fly Musca domestica ( S2A Fig ) . Moreover , the putative Bowl BS in CR2 ( 5’-ACAGAAGC ) is embedded within an imperfect palindromic sequence ( 5’-ACAGAAGCCGTCTGG; underlined nucleotides may constitute a second Bowl BS in an inverted orientation ) , which appears , however , to be strictly conserved only among Drosophilidae species ( S2A Fig ) . Full-length Bowl GST fusion is toxic when expressed in bacteria . To test for direct interaction with LAE sequences in vitro , we therefore expressed a 163 amino-acid long segment of the Bowl protein encompassing its zinc-finger domain ( termed BowlZF ) as a glutathione S-transferase ( GST ) fusion protein in E . coli . Purified GST-BowlZF was examined for specific DNA binding in an electrophoretic gel mobility shift assay ( EMSA ) ( Fig 2B ) . As a positive control , we used a DNA probe ( OddZF BS ) previously shown to interact with a recombinant GST-OddZF fusion protein [22] . DNA probes encompassing each consensus site within CR2 ( 1S+ ) and in the 3’-end neighboring conserved sequence ( 2S+ ) ( see Fig 2A ) interacted with GST-BowlZF , albeit weakly compared to the Odd ZF BS ( Fig 2B , compare lanes 5 and 13 with lane 3 ) , but not with unfused GST used as a negative control ( lanes 2 , 4 and 12 ) . By contrast , GST-BowlZF interacted strongly with a larger DNA probe ( 1L+ ) encompassing the entire CR2 near-palindromic sequence ( compare panels 5 and 9 ) , and binding was abolished when the common CAG nucleotides ( underlined in Fig 2A ) present in each probe were mutated to TTT ( 1Sm , 1Lm and 2Sm; lanes 7 , 11 and 15 , respectively ) . The shifted 1S+ and 1L+ DNA-protein complexes ( black arrowheads ) migrated similarly , suggesting that Bowl ZF fusion protein might interact as a monomer in both cases . Taken together , these data indicate that the Bowl ZF DNA-binding domain interacts specifically with dedicated LAE sequences in vitro , particularly with the near-palindromic site within CR2 . To evaluate whether the two conserved Bowl ZF-binding sites within the LAE enhancer are critical for bab2 repression in the developing distal leg , we sequentially mutated the common CAG motif of the double site within CR2 and of the single one outside CR2 to TTT , to yield the LAEm2-GFP and then LAEm3-GFP derivatives ( Fig 2C ) . When compared to the normal bab2-expressing tarsal domain ( monitored with LAEwt-RFP ) , LAEm2-GFP displayed strong GFP de-repression in proximal ts1 and distal ts5 ( Fig 2D ) , as previously seen for the CR2-deleted LAEΔ2-GFP reporter ( above , Fig 1C and 1D ) . LAEm3-GFP ( mutated for both Odd/Bowl ZF BS ) was likewise de-repressed proximally and distally ( Fig 2E ) . However , proximal LAEm3-GFP up-regulation is significantly broader , compared to LAEm2-GFP ( compare the extents of the green brackets in Fig 2D and 2E ) , indicating that both conserved Bowl binding sites are required for full repression of the LAE enhancer in proximal ts1 , while only the CR2 site is apparently needed in distal ts5 . To confirm that the Lines/Bowl pathway acts in vivo through the biochemically-defined binding sites within the LAE enhancer , we ectopically-stabilized nuclear Bowl protein in flip-out mitotic clones , by down-regulating lines activity within the developing leg through mosaic RNAi knockdown ( above , see Fig 1G ) and examining fluorescence patterns driven by thrice CAG-mutated LAEm3-GFP . Whereas the wild-type LAE reporter was repressed in lines-deficient tarsal cells ( RFP+ ) , LAEm3-GFP was impervious to repression by stabilized Bowl within the same cells ( compare Fig 2F , 2F’ , 2G and 2G’; lines-deficient FO clones are circled with white dashed lines ) . Therefore , we conclude that the Bowl transcription factor acts directly through its evolutionarily-conserved cognate binding sites within the LAE enhancer , to limit bab2 expression in the developing tarsus . Previous work has shown that EGFR signaling extinguishes bab2 expression within the distalmost subdivision of the developing leg , the pretarsus [5 , 8] . The mutant LAE reporter LAEm3-GFP , while resistant to Bowl-mediated P-D repression ( Fig 2G ) is still repressed in the pretarsal primordium ( Fig 2E ) , as visualized upon co-staining with the C15 specific marker ( S1C Fig ) . Moreover , Bowl protein is expressed in a row of C15-expressing pretarsal cells at the ts5/pretarsal boundary , where LAEm3-GFP expression remains undetected ( S1D Fig ) . Since LAE reporter expression is absent from C15-expressing cells , we asked whether C15 TF activity mediates EGFR signaling-dependent down-regulation of LAE enhancer activity in the developing pretarsus . C15 homozygous mutants for a null allele ( C152 ) survive to adulthood with distal leg defects , exemplified by the absence of claws [8] . In early-mid L3 , C15/clawless expression is activated in the pretarsal primordium , in response to activation of EGFR signaling at the center of the leg disc , then subsequently maintained until the adult stage [8 , 9] . To ask whether C15 regulates LAE enhancer activity , we first compared expression of the bab2 ( LAE-RFP reporter ) and C15 genes in the leg disc . At ~84-90h AEL , C15 and LAE-RFP are expressed in adjacent territories within the emerging tarsus/pretarsus ( Fig 3A ) . Soon afterwards at mild-late L3 stage C15 and LAE-RFP expression domains become well separated from each other , through gradual restriction of the LAE reporter by repressive Bowl activity in developing distal ts5 cells ( Fig 3B; white bracket ) [16] . Note that in the mid-L3 leg disc , LAE-RFP reporter expression is never detected in C15+ cells split away from the emerging central C15-expressing domain ( Fig 3A , circled areas; n = 20 ) ( see also below ) , suggesting that C15 TF might be sufficient to cell-autonomously extinguish LAE activity in the developing pretarsus , as suggested by Campbell [8] . Consistent with this possibility , bab2 ( LAE-RFP ) de-repression was observed in all pretarsal cells in C152 homozygous late-L3 leg discs ( Fig 3C and 3D ) . Surprisingly , in C15 homozygous mutant larvae bab2 ( LAE-RFP ) expression was likewise de-repressed in developing distal ts5 cells , where repressive Bowl activity should be acting given our previous data ( see Fig 2D and 2E ) . In fact , distal bowl expression is fully lost in C15-/- leg discs ( compare S3A and S3B Fig ) [8] . Altogether these data support the notion that the C15 transcription factor might mediate directly EGFR-induced bab2 repression in developing pretarsal cells , in addition to allowing indirectly gradual repression in ts5 cells by stabilizing Bowl there through Notch signaling activation . To examine whether C15 activity is sufficient for bab2 ( LAE-RFP ) down-regulation when ectopically expressed in developing tarsal cells , in accordance with a putative direct repression , we used the FO technique to misexpress C15 protein in the leg disc . Ectopic C15 expression led to cell-autonomous bab2 ( LAE-RFP ) repression in tarsal FO clones ( Fig 3E ) . Both the loss- and gain-of-function genetic experiments indicate that the C15 transcription factor down-regulates bab2 and LAE enhancer activity within the developing pretarsus , and this negative regulation might occur possibly through direct interaction with LAE sequences . The first indication for direct C15 binding to the LAE enhancer came from its identification as an LAE–interacting protein through a one-hybrid ( Y1H ) assay in yeast cells . Taking advantage of a nearly-complete D . melanogaster TF prey library ( i . e . , 692 upon 755 predicted DNA-binding regulators ) [23] , we performed a Y1H screen using as DNA bait the entire LAE or a 230 base pair ( bp ) 3’-truncated derivative ( minimal CRM or miniLAE; see Fig 4A , left part ) . Note that miniLAE appeared sufficient ( albeit with lower enhancer efficiency ) for tissue-specific expression in a reporter assay , notably repression in pretarsal cells [19] . Significantly , from two independent Y1H screens , each done in duplicate , a strong TF-LAE interaction , corresponding to the C15 homeodomain protein ( homeoprotein ) , reproducibly stood out for both constructs ( Fig 4A , right part ) . Given that Y1H assays reveal direct interactions , we sought to identify C15 binding site ( s ) [24 , 25] within the miniLAE sequence ( Fig 4B ) , that includes both the conserved CR1 region critical for activation , and the Bowl-interacting CR2 [19] . Neither the deletion of CR2 ( above , Fig 1C and S1B Fig ) nor systematic mutagenesis of CR1 sub-motifs [19] resulted in LAE-driven reporter gene de-repression in developing pretarsal cells . Thus , repressive C15 homeoprotein TF might act through redundant homeodomain ( HD ) binding sites or , alternatively , compete with activating Distal-less homeoprotein for shared HD cognate sites within CR1 [19] . The 69-bp CR1 sequence is highly enriched in A/T nucleotides ( 67%; Fig 4B ) , including the previously-identified Dll-binding sequence ( DBS ) within its 3’-moiety ( a 22-bp region encompassing three canonical TAAT homeodomain binding sites ) [19] . Using a purified recombinant GST fusion protein , we asked whether a 101 amino-acid-long C15 fragment encompassing the homeodomain ( C15HD ) can bind stably the DBS in a gel retardation assay . Contrary to a purified GST-Dll fusion used as a positive control , GST-C15HD did not bind detectably the DBS probe ( Fig 4C , lanes 1–2 ) . In striking contrast , GST-C15HD bound strongly a DNA probe encompassing the whole CR1 sequence , forming up to three protein-DNA complexes depending on EMSA conditions ( Fig 4C , lanes 3–4 ) . The simplest interpretation is that several GST-C15HD molecules can interact simultaneously with several cognate binding sites distinct from the Dll DBS within CR1 . To test this possibility , we mutated the Dll DBS TAAT sites , as reported in [19] . The GST-C15HD fusion readily interacted with mutated CR1 ( CR1DBSm ) , leading to equivalent retarded protein-DNA complexes ( Fig 4C , lane 6; quantification from three experiments is shown in Fig 4B , on the right side ) . This result suggested that C15 homeodomain interacts mainly with sequences outside the DBS . Next , we sought to identify C15HD binding sequences among the numerous A/T-rich motifs ( prone to interact with C15 HD ) present within CR1 ( termed S1 to S6; underlined in Fig 4B ) . In addition to the TAAT sequences within the DBS , we tested a probe ( CR1S1+4+DBSm ) also mutated for the non-consensus TAA sequences within the S1 and S4 motifs . The GST-C15HD fusion still interacted with CR1S1+4+DBSm ( Fig 4C , lane 8 ) , albeit with slightly lower efficiency ( see Fig 4B , right side ) and with only one detectable retarded complex ( Fig 4C , compare lanes 4 and 8 ) . Second , we tested a series of probes mutated for at least three A/T motifs ( formed by at least two consecutive A or T ) , substituting them by purely G/C sequences ( Fig 4B; depicted in blue ) , leaving intact the DBS . Binding was faint for all probes mutated for S6 , or absent when S3 or S4 were also mutated ( Fig 4C and S4A Fig ) , suggesting that S6 and S3-4 are respectively high and low affinity cognate sites . We then tested a second series of probes specifically mutated for S3-4 ( CR1S3+4m ) , S6 ( CR1S6m ) or both ( CR1S3+4+6m ) ( see Fig 4B for sequences ) . As expected , GST-C15HD interaction was slightly reduced ( CR1S3+4m ) , strongly reduced ( CR1S6m ) or abolished ( CR1S3+4+6m ) ( Fig 4C , lanes 14 , 16 and 18 , respectively; see Fig 4B for quantification ) . Altogether , these biochemical data indicate that C15 homeodomain interacts strongly with S6 and much less efficiently with S3-4 binding sites within CR1 . We then asked whether the C15 HD binding sites within the critical 69-bp CR1 region contributes to LAE repression in vivo . Given that systematic linker scanning mutagenesis of CR1 never resulted in de-repression [19] , transgenic lines expressing LAE-RFP derivatives mutated for both the major ( S6 ) and minor ( S3-4 ) sites were generated . In contrast to a wild-type LAE-RFP construct inserted at a same genomic site , none of the two tested mutated derivatives ( miniLAE and F2LAE , a larger form without the TAA-rich CR3 prone to interact with homeoproteins; Fig 4A and [19] ) was detectably expressed within the leg disc ( S4B Fig ) . These results suggest that in addition to a role in C15-mediated repression S3-4 and/or S6 CR1 sequences are also required for enhancer activation in the developing leg , which thus preclude from evaluating their implication in distal repression . Consistent with this hypothesis , our previous linker scanning mutagenesis revealed partially-redundant activating regulatory information within CR1 , notably for subsequences encompassing the C15 binding sites [19] . Although C15 binding remains to be formally established in vivo , our data suggest that the EGFR-responsive C15 transcriptional repressor extinguishes bab2 in the developing pretarsus by interacting directly with the LAE enhancer , at least through two cognate binding sites within CR1 . C15 has been shown to function together with Aristaless ( Al ) , another homeoprotein [9] . Al and C15 proteins are co-expressed in the developing pretarsus and both HD proteins interact cooperatively with a composite DNA binding site [ ( T/C ) TAATTAA ( T/A ) ( T/A ) G][26] . However , the assumption of a systematic partnership between C15 and Al is based on the single identified common target locus Bar . Given that C15 HD binds directly LAE sequences distinct from the Bar enhancer ( i . e . without TAATTAA core sequence; see Fig 4B ) , we next asked whether C15 acts independently of Al in repressing LAE activity . First , we tested whether C15 binds cooperatively with Al to LAE sequences in vitro . Given that C15 and Al interact physically through contacts between their homeodomain [26] , we performed EMSA experiments with protein fragments sufficient for intermolecular interaction . Purified GST-C15HD ( see above ) was combined with in vitro translated Al homeodomain ( AlHD; Fig 5A ) . As a positive control , we used the Bar enhancer probe ( BarEnh ) . In accordance with previous data [9] , while GST-C15HD bound poorly to the BarEnh probe , a new retarded complex was clearly observed upon addition of AlHD extracts ( Fig 5A , compare lanes 2 and 3 ) . In striking contrast , GST-C15HD bound strongly to the CR1 probe alone and adding AlHD-containing extracts did no yield heterodimeric complex ( lanes 7–8 ) . In fact , C15 HD binding was not enhanced , but instead diminished presumably due to the presence of competitors in the reticulocyte extracts ( compare lanes 8–9 ) . While Al homeodomain bound strongly to the BarEnh probe , it interacted poorly with the CR1 probe ( compare lanes 5 and 10 ) . Altogether these data indicate that C15 HD interacts strongly in vitro with LAE sequences independently of Al . Second , we asked whether C15 homeoprotein is able to down-regulate LAE activity independently of Al in vivo . A first indication came from the observation in early-mid L3 that LAE-RFP expression is never detected in C15+ cells split away from the central domain ( above , see Fig 3A ) , suggesting a cell autonomous C15-mediated repression . We found that those peripheral C15+ cells never co-expressed Al ( n = 15 ) ( Fig 5B , see white arrows ) , indicating that this repression does not require Al activity . A second indication came from C15 mis-expression within the bab2-expressing tarsal field . We previously showed that ectopic C15 expression in FO clones is sufficient to down-regulate LAE activity within the developing tarsus ( above , see Fig 3E ) . C15 ectopic expression in tarsal FO clones has been shown in fact to be sufficient to induce al expression in a cell autonomous manner [8] , although not in all the C15-misexpressing cells ( Fig 5C ) . To separate the roles of C15 and Al in LAE-RFP repression , we used the FO technique to couple C15 gain-of-function with RNAi-induced Al knock-down . Although dsRNA treatment was not sufficient for al extinction everywhere , an autonomous LAE-RFP down-regulation was still observed in all C15+ cells , even those without detectable Al protein ( Fig 5D ) . Altogether these data provide evidence that C15 transcription factor represses directly LAE activity in vivo independently of Al . Repressive C15 and activating Dll homeoproteins bind strongly adjacent cognate sites within CR1 ( above , Fig 4B ) . We next asked whether they can do so simultaneously . To this end , we tested their binding with a probe encompassing the high-affinity C15 as well as the neighboring Dll DBS sites ( corresponding to the 3’ half of CR1; Fig 6A , see upper part ) . In addition to the GST-C15HD protein described above , we purified a Dll fusion protein ( GST-DllHD ) comprising only its homeodomain . As expected each homeodomain fusion bound the 3’ CR1 probe ( Fig 6A , left lower part ) , yielding either a strictly monomeric GST-DllHD or a monomeric as well as some detectable dimeric GST-C15HD complexes . To ask whether C15 and Dll homeodomains can interact jointly with their respective CR1 binding sequence , we simultaneously added the same amount of proteins in the DNA retardation assay . Significantly , no heterodimeric complexes could be detected , and respective monomeric complexes were even diminished . Moreover , adding increasing amounts of GST-C15HD resulted in disappearance of GST-DllHD bound probe in favor of bound GST-C15HD ( Fig 6A , right lower part ) . Thus C15 and Dll compete for CR1 binding in vitro . C15 and Al interact physically through contacts between their homeodomain [26] . We therefore asked whether C15 also establishes intermolecular interactions with the Dll HD , by performing pull-down experiments with purified GST-C15HD and GST-DllHD fusions . In vitro translated 35S-labelled Dll homeodomain interacted with GST-C15HD as efficiently as Al HD ( Fig 6B , compare middle to left panels ) . Likewise , in vitro translated 35S-C15 homeodomain bound to GST-DllHD in the reciprocal test ( right panel ) . Significantly , GST-DllHD did not interact with radiolabeled Al homeodomain . Thus , Dll and C15 engage specific protein-protein interaction via their respective DNA-binding domain , providing a possible rationale for their competitive interaction with LAE sequences . Lastly , we examined the functional antagonism of C15 and Dll in vivo . We postulated that the down-regulation of LAE expression observed on ectopically expressing C15 in developing tarsal cells ( above , Fig 3E ) , involves direct competition with activating Dll . Accordingly , we asked whether Dll up-regulation in pretarsal cells could impede repression by C15 . On overexpressing Dll activator in C15+ FO clones ( that are likewise Dll+ ) , clear de novo LAE-RFP expression could be detected in all pretarsal clones ( n = 12 ) ( Fig 6C , see white arrow ) , while C15 expression remained apparently unchanged ( Fig 6C’ ) . These data indicate that increased Dll protein is able to specifically counteract C15 repressive activity in pretarsal cells . Importantly , LAE-RFP up-regulation was never observed on ectopically expressing Rn protein ( Fig 6D and see also below ) , indicating that the Dll-induced de-repressive effect is not a general effect of bab2 activators . Given that C15 and Rn interact directly with well separated LAE sequences [19] , the functional specificity of up-regulated Dll on C15 repressive activity is consistent with their binding site proximity . Taken together , these data provide evidence for a new layer of bab2 regulation by C15 , through direct physical interaction with the Dll activator and competitive binding to neighboring cognate sequences within the LAE . In fact , three non-exclusive distinct mechanisms can be envisioned for C15 repressive activity: ( 1 ) active repression due to direct C15 binding to CR1 sequences , ( 2 ) steric hindrance toward Dll activator binding to adjacent cognate sites and ( 3 ) heterodimerization with Dll changing its binding specificity . Surprisingly , while in C152 homozygous leg discs we observed a clear ectopic bab2 expression in all cells of the developing mutant pretarsus ( above , Fig 3D ) , only a faint de-repression of LAE-RFP was detected within mutant clones ( Fig 7A ) . Moreover , LAE-RFP de-repression only occurred in mutant cell subsets . This localized slight de-repression ( red arrow ) corresponded to those mutant cells situated farthest away from remaining C15+ cells ( as detected by the presence of C15 antibody staining as well as absence of GFP fluorescence ) . This observation suggested that in addition to its cell-autonomous repressive activity , C15 could also activate non-autonomously another bab2 repressor . As mentioned above , C15 induces bowl expression in a non-cell autonomous manner through N signaling activation [8] . We therefore asked whether Bowl TF could be responsible for this non-autonomous repressive effect . As shown in Fig 7B , Bowl protein was indeed expressed ( as detected by the presence of Bowl antibody staining ) in some C15 mutant clones ( expressing GFP ) surrounding remaining C15+ cells ( GFP- ) . Furthermore , LAE-RFP expression was detected only in C15-/- cells ( n>50 ) that corresponded precisely to those that lacked detectable Bowl protein accumulation ( Fig 7B , red arrows ) . This observation suggested that in C15 mutant clones even low amounts of Bowl protein ( magenta arrows ) are sufficient to fully extinguish LAE reporter activity . In support of this , in mosaic leg discs harboring bowl loss-of-function clones a faint LAE-RFP de-repression was observed in all bowl-/- clones situated within the C15-expressing pretarsal field ( Fig 7C , see yellow arrow ) . Moreover , the developing pretarsus does not detectably accumulate nuclear Bowl TF except for a single row of cells at the ts5-pt boundary ( see S1B Fig ) . To explore the hypothesis of Bowl-mediated C15 dependent repression , we then performed C15 clonal loss-of-function analysis with LAEm3-GFP , specifically refractory to Bowl repressive activity ( above ) . Note that in this experiment , wild-type cells ( as detected with C15 immunostaining ) are marked with RFP . In this effectively Bowl-inoperative context , LAEm3-GFP expression was fully de-repressed ( green arrow ) in most if not all C15-/- pretarsal cells ( visualized here by the absence of RFP fluorescence ) ( Fig 7D ) , indicating that the previously-observed non-autonomous repressive effect induced by C15+ cells is no longer at work . Taken together , we conclude that C15 homeodomain and Bowl zinc-finger proteins act as a repressive TF duo by repressing directly LAE enhancer activity in the developing leg , in response to a cell signaling cascade , to refine the distal bab2 expression border . Together with the Dll and Rn activators , Bowl and C15 repressors constitute a TF quartet binding directly discrete sequences within the LAE enhancer , to ensure dynamic resolution of bab2 expression in the developing distal leg . We next asked whether they engage cross-regulatory interactions . Contrary to Dll , transcriptional enhancers have not been yet characterized for rn , bowl and C15 . First , we sought to examine rn expression . To this end , we generated a polyclonal antibody recognizing Rn ( see Materials and methods ) . While in situ hybridization indicated transient expression limited from early-mid to mid-late L3 stages [10 , 27] , nuclear Rn could be detected later on . Indeed , at the late L3 stage Rn and LAE-RFP are co-expressed in most tarsal cells ( Fig 8A ) , with Rn extending more proximally other several cell rows , while conversely LAE-RFP expression extends more distally ( Fig 8A , see brackets ) . Second , we asked whether the Bowl/Lines cassette regulates rn in leg discs expressing LAE-RFP as an internal control . As previously observed for bab2 ( Figs 1F and 7C ) , Rn was cell-autonomously up-regulated in some bowl mutant clones ( Fig 8B , yellow arrows ) . Indeed , distalmost and proximalmost mutant clones only up-regulated LAE-RFP ( white arrows ) . Conversely , rn was cell-autonomously repressed in Bowl gain-of-function clones ( Fig 8C , yellow arrow ) , obtained through FO expression of dsRNA directed against the dedicated Bowl antagonist lines . Altogether these data indicate that rn expression in the leg disc is regulated negatively by the Lines/Bowl pathway , as previously shown in the antennal disc [13] . Third , we wondered whether C15 regulates rn expression . Contrary to bab2 which is first expressed as a distal circular domain [20] , onset of rn expression occurs at the early-mid L3 stage in the emerging tarsal field in the form of a ring between Dac- and Bar-expressing ring-shaped domains [10] . Rn and C15 proteins are thus never abutting . As previously shown for bab2 ( Fig 3E ) , clonal analyses revealed that rn was autonomously repressed in all C15-misexpressing cells ( Fig 8D ) . As previously described [8] , tarsal C15 misexpression stabilized nuclear Bowl in some but not all FO cells ( Fig 8E , white arrows ) , thus ruling out that ectopically-stabilized Bowl mediates Rn extinction in C15+ cells . Fourth , we asked whether rn regulates Dll , bowl and C15 . In mosaic leg discs expressing the LAE-GFP or LAE-RFP construct , no effect on Dll , Bowl and C15 expression could be observed either in rn loss- or gain-of-function ( S5 Fig and Fig 6D ) . As seen in our previous work [19] , we did observe cell-autonomous LAE-GFP down-regulation in rn-/- clones within ts1-2 , but not ts3-5 ( S5 Fig , panels A-C ) . Conversely , clonal Rn overexpression led to cell-autonomous LAE-RFP up-regulation ( S5D Fig , see white arrow ) , particularly in bab2-expressing ts4-5 cells ( n = 12 ) . Fifth , as previously reported [8] , misexpressed C15 protein down-regulated bowl expression distally ( S6A Fig ) , while conversely ectopically-stabilized Bowl protein did not affect C15 expression ( S6B Fig ) . Importantly , bowl gain-of-function clones ( through lines down-regulation ) leading to LAE-RFP repression ( Fig 1G ) , never induced C15 up-regulation ( S6C Fig ) , confirming that LAE-RFP repression by Bowl stabilization within the tarsal cells occurs independently of C15 TF activity . Lastly , though the distal selector gene Dll is required directly or indirectly for expression of bowl , C15 and rn [5 , 12 , 28–30] , none was upregulated in mosaic leg discs misexpressing Dll activator ( S6D and S6E Fig and Fig 6C ) and none affected in turn Dll expression in clonal loss- or gain-of-function ( see S5A Fig for rn as well as S7A and S7C Fig for bowl and C15 , respectively ) . While as expected ( see Fig 1F ) LAE-RFP was up-regulated in bowl mutant clones ( S7A Fig ) , its re-activation occurred only in Dll-expressing cells , either distally or proximally ( compare white and yellow arrows , respectively ) . The observation that LAE-RFP could only be de-repressed in Dll+ cells is consistent with a critical activating role of the Dll TF in bab2 expression throughout limb development . In conclusion , these results indicate that rn is also a repressed target of Bowl and C15 , drawing new connections within the gene regulatory network governing distal leg development .
Repressive bowl activity has been previously proposed to mediate graded bab2 extinction in response to N signaling , in distal ts5 and in proximal ts1 cells [13 , 16] . As for other tissues , bowl activity in the leg disc is down-regulated by lines activity [13 , 15 , 31] , presumably through a direct physical association between the Bowl and Lines proteins [15] . In the present study , we provided several lines of evidence that the Lines/Bowl pathway regulates bab2 expression in developing distal leg cells through direct interaction of the Bowl ZF domain with conserved Odd family binding motifs within CR2 [21 , 22 , 32] . First , both bab2 expression and LAE-RFP reporter activity are cell-autonomously de-repressed in bowl mutant clones . Second , ectopic nuclear stabilization of Bowl TF ( through lines down-regulation ) is sufficient for cell-autonomous bab2 ( LAE-RFP ) down-regulation in developing tarsal cells . Third , recombinant Bowl ZF interacts directly with conserved LAE motifs , in a sequence-specific manner . Lastly , these binding sites are relevant in vivo , since ectopically stabilized Bowl protein is unable to extinguish a bab2 reporter gene mutant for the two biochemically-defined binding sites , confirming their critical role in mediating LAE enhancer repression . As predicted from a bacterial one-hybrid assay [21] , Bowl and Odd proteins interact in vitro with highly similar DNA-binding motifs [22] ( this study ) . It is worth noting that bab2 is the first direct target gene identified so far for the Bowl TF . Although the near-palindromic CR2 cognate site is critical , a second one situated approximately 100 bp away is required specifically for full repression of the LAE enhancer in proximal leg tissues ( Fig 2C ) . How the two Bowl ZF-binding sites cooperate in proximal ts1 cells remains to be deciphered . Interestingly , cooperation between distant cognate sites has been suggested for transcriptional targets of the related Odd TF during embryogenesis [22] . Initially expressed within a broad , distal circular domain at the early L3 stage , bab2 is extinguished in the developing pretarsus in response to EGFR signaling activation at early-mid L3 stage [5] . Among known EGFR signaling targets [3] , we provided several lines of evidence that the C15 transcription factor is directly repressing bab2 and do so independently of its regular partner Al . First , C15 protein behaves as a strong LAE interactor in yeast cells . Second , C15 homeodomain interacts strongly in vitro with dedicated binding sites within the critical CR1 LAE sequence , without synergistic effect of Al ( contrary to Bar enhancer sequences ) . Third , both bab2 expression and LAE-RFP reporter activity are de-repressed distally in leg discs deficient for C15 activity . Fourth , ectopic expression of the C15 homeoprotein is sufficient to extinguish cell-autonomously bab2 ( and LAE reporter activity ) in tarsal cells , even in Al deficient cells . Altogether our data suggest that Bar regulation by the C15-Al heterodimer does not reflect a rule . Characterization of other direct target genes , among candidates such as rn ( this work ) and Dl [8] , will indicate to what extent C15 TF is acting independently of Al . Within the essential CR1 sequence , C15 interacts strongly with an A/T-rich motif ( S6 ) adjacent to the critical TAAT-rich Dll homeoprotein interacting DBS ( Fig 4 ) , which is not required for efficient C15 binding to its neighboring site S6 . This proximity has prompted us to examine whether C15 and Dll are able to bind simultaneously to CR1 sequences encompassing both sites . Surprisingly , in EMSA experiments , simultaneous addition of both homeoproteins never revealed C15-Dll heterodimeric binding . We showed that C15 is able to compete with Dll upon binding to CR1 sequences , in a dose-sensitive manner ( Fig 6A ) . Moreover , we found that C15 , but not Al , interacts physically with Dll through their homeodomain ( Fig 6B ) , providing a molecular mechanism for their antagonistic DNA binding to LAE sequences . Thus , C15 engages intermolecular interactions with both Al and Dll , through their related DNA-binding domain but with opposite outcomes: cooperative binding to specific DNA sequence ( C15-Al ) versus competition for LAE binding ( C15-Dll ) . Further studies will determine relevant mechanism ( s ) of bab2 repression by C15 among at least three non-exclusive possibilities: ( 1 ) Dll-independent C15 repressive activity through direct binding on its own cognate sites within LAE ( presumably through interaction with co-repressors , see below ) ; ( 2 ) direct C15 binding to the S6 CR1 sequence sterically hindering Dll binding to its own adjacent cognate site ( DBS ) ; and ( 3 ) heterodimerization with Dll , inhibiting its direct interaction with LAE or changing its global DNA binding specificity , independently of C15 DNA binding per se . Whatever the mechanism ( s ) involved , our data suggest that C15 and Dll compete functionally in vivo in differentially regulating LAE activity , i . e . repression versus activation ( Fig 6C ) . Given that both C15 and Dll ( as well as Al ) are co-expressed in pretarsal cells , we speculate that the relative amounts of C15 and Dll homeoproteins are critical for full repression in the distal leg . Interestingly , mammalian homologs of Dll form heterodimers with repressive Msx-type homeoproteins , through their respective HDs , with mutually antagonistic interactions in vitro and in-vivo [33] , indicating an evolutionarily-conserved propensity of Dll-type HDs in assembling heterodimeric complexes with repressive homeoproteins . We thus propose that at early-mid L3 stage , before C15 activation , Dll expressed throughout the distal territory allows bab2 induction as a distal circular domain , by binding the DBS within LAE . At the onset of C15 expression in response to EGFR signaling , increasing amount of C15 protein competes with Dll activator in pretarsal cells , thus resolving bab2 expression pattern as a single ring of cells at the mid L3 stage . As stated above , in addition to interacting and/or competing with activating Dll on binding CR1 sequences , C15 might repress LAE activity on its own . Consistently , C15 harbors an eh1-type repressor domain that can interact with the Groucho ( Gro ) co-repressor [34] . Additionally , we showed that Bowl must function together with C15 to fully repress LAE-activity ( Fig 7C ) . It is noteworthy that Bowl also harbors an eh1 repressive motif [35] . As for C15 ( above ) , Bowl ZF protein is therefore predicted to interact with the Gro co-repressor and both repressive transcription factors may extinguish bab2 through the same molecular mechanisms , delimiting the tarsal expression domain in a timely manner . Future work should determine whether the C15 transcription factor functions together with the Gro protein to extinguish bab2 in the developing pretarsal primordium . Taken together with our previous work [19] , we propose an updated model for LAE regulatory activity in the developing distal leg ( Fig 9 ) . The tissue-specificity of the LAE enhancer is ensured by its CR1-2 region , while its 3’-end moiety , containing CR3 , is mainly required for transcriptional enhancer strength . Situated downstream of the Wg and Dpp signaling inputs , the Distal-less homeodomain TF is required to induce and maintain overall bab2 expression within the developing tarsus , from L3 to adult stages , by acting through CR1 [19] . Given that Dll is expressed early on during larval development , well before bab2 expression onset as a circular distal domain , another transcriptional activator functioning at early-mid L3 stage remains to be identified . This factor may operate through activating CR1 sequences overlapping the C15 binding sites ( S4B Fig ) . From early-mid to mid-late L3 stages , in response to Tal signaling induced by EGFR together with Dll inputs and via spineless-mediated activation [10 , 29] , Rn is activated in the ts1-3 primordia and up-regulates LAE activity in the proximalmost bab2-expressing cells . Given the dynamic activity of Bowl protein during tarsal growth , de Celis and Bray [16] anticipated that a transiently-expressed bab2 activator must counteract repressive Bowl TF in developing tarsal cells . Our previous work indicates that the Rn ZF protein could be this factor [19] . The Rn transcriptional activator is indeed expressed at the right time and in the proper cells , counteracting directly ongoing repressive Bowl on LAE enhancer activity . Interestingly , dynamic bowl activity is under the negative control of the transiently-accumulated Spineless ( Ss ) bHLH-PAS TF [16] , which itself is positively required for rn activation [10] . Lastly , Bowl activity represses rn expression ( this work ) , suggesting that Ss may in fact induce rn by counteracting Bowl repression . Concomitant to the onset of Rn activity in proximal tarsal cells , and in response to EGFR signaling at early-mid L3 stage , C15 is activated and alone ( i . e . without Al ) extinguishes bab2 in the pretarsal primordium , presumably in a dose-dependent manner ( see above ) , through direct competitive interactions with Dll ( this work ) . At the same time , C15 cell-autonomously down-regulates Dl in the developing pretarsal cells , allowing bowl up-regulation in response to activated N signaling at the tarsus/pretarsus boundary [8] . Thus , C15 is required non-cell-autonomously for bowl expression in developing ts5 [8] . There , the Bowl TF in turn represses LAE enhancer activity ( through its CR2 cognate site ) to reshape the distalmost bab2-expressing ring during mid to late L3 stages . Lastly , at the late L3 stage and beyond up to the adult stage , graded bab2 expression is observed in each tarsal segment [17] . To complete our model regarding the molecular inputs controlling LAE enhancer activity , the previously-identified CR1-binding inter-ring repressive activity [19] ( Factor Y in Fig 9 ) and the bab2 activator acting during the early L3 stage ( X ) remain to be characterized . Finally , this work also shed light to new epistatic relationships between bab2 regulators within the leg genetic cascade . In addition to bab2 , we showed that C15 and bowl regulate also rotund . Nevertheless , direct interactions with rn enhancer sequences remain to be established . It would be interesting to determine whether C15 ( i ) functions jointly with Bowl and/or ( ii ) antagonizes Dll activity in repressing other target genes .
Drosophila lines were grown on standard yeast extract-sucrose medium . The vasa-PhiC31 ZH2A attP stock ( kindly provided by F . Karch ) was used to generate the mutant LAE-GFP reporter lines , as previously described [19] . LAE-RFP constructs were generated by insertions on both the ZH2A ( X chromosome ) and ZH86Fb ( third chromosome ) attP landing platforms , and they display identical expression patterns [19] . UAS-C15 , UAS-Dll and UAS-Rn stocks were obtained from T . Kojima , S . Cohen and the Bloomington stock Center ( #7403 ) , respectively . C152/TM6B , Tb1 stock was kindly provided by G . Campbell . Mutant mitotic clones for null alleles of bowl , C15 and rotund were generated with the following genotypes: y w LAEwt-RFP; DllGal4EM2012 UAS-Flp / +; Ub-GFP FRT40A/bowl1 FRT40A ( i . e . , bowl mutant clones are GFP negative; Figs 1F , 7C and 8B and S7A Fig ) , y w LAEwt-RFP; DllGal4EM2012 UAS-Flp / +; FRT82B/FRT82B Ub-GFP C152 ( i . e . , C15 mutant clones harbor two Ub-GFP copies; Fig 7A and 7B ) or y w LAEwt-GFP ( or LAEm3-GFP ) ; DllGal4EM2012 UAS-Flp / +; Ub-RFP FRT82B / FRT82B C152 and y w LAEwt-GFP; DllGal4EM2012 UAS-Flp / +; Ub-RFP FRT82B/FRT82B rn16 ( i . e . , C15 or rn mutant clones are marked by the absence of RFP fluorescence; Fig 7D and S5A–S5C and S7C Figs ) . UAS-dsRNA stocks used to obtain RNAi against lines ( #40939 ) or al ( #26747 ) were obtained from the Bloomington stock Center . “Flip-out” mitotic clones over-expressing dsRNA against lines were generated by 40 mn heat shocks at 38°C , in early second- to early-mid L3 larvae of genotypes: y w LAEwt-RFP hsFlp; UAS-dsRNAlines / Pact>y+>Gal4 UAS-GFP ( i . e . , FO clones express GFP in Fig 1G ) and y w LAEwt-GFP ( or LAEm3-GFP ) hsFlp; UAS-dsRNAlines / Pact>y+>Gal4 UAS-RFP ( i . e . , Lines-deficient FO clones express RFP in Figs 2F , 2G and 8C and S6B , S6C and S7B Figs ) . FO clones overexpressing C15 , Rn or Dll were generated by 40 mn heat shocks at 38°C , in mid second to early-mid L3 larvae of genotypes: y w LAEwt-RFP hsFlp; UAS-C15 / Pact>y+>Gal4 UAS-GFP; ( +/-UAS-dsRNAal / + ) ( i . e . , FO clones express GFP; Figs 3E , 5C , 5D , 8D and 8E and S6A and S7D Figs ) , y w LAEwt-RFP hsFlp; UAS-Rn / Pact>y+>Gal4 UAS-GFP ( i . e . , FO clones express GFP; Fig 6D and S5D and S5E Fig ) , and y w LAEwt-RFP hsFlp; Pact>y+>Gal4 UAS-GFP / +; UAS Dll / + ( i . e . , FO clones express GFP: Fig 6C and S6D and S6E Fig ) , respectively . LAE fragments were cloned into pBP-S3aG vector as previously described [19] . pLAE-GFP or -RFP site-directed mutagenesis was performed by PCR , using the overlap extension method [36] . All plasmid constructs used in this work were sequence verified . Mutated LAE reporters were all inserted on the same ZH2A attP landing platform . The Y1H screen was performed as described in [23] . LAE and mini-LAE DNA baits were amplified by PCR and cloned into pENTRY-5’ using standard restriction-ligation techniques . The sequence-verified LAE baits were sub-cloned into the Y1H-compatible pMW2 ( HIS3 ) vector by Gateway LR cloning . LAE destination clones were integrated in Y1H-YM4271 yeast strain using lithium acetate-polyethylene glycol transformation followed by selection on a synthetic complete ( SC ) ( -His ) medium plate . Leg discs were dissected from wandering L3 larvae . Indirect immuno-fluorescence was carried out as previously described [19] using a LEICA TCS SP5 or SP8 confocal microscope . Rat anti-Bab2 [18] , rabbit anti-Bowl [16] , rabbit anti-C15 [9] and rat anti-Al [8] antibodies , kindly provided by F . Laski , S . Bray , T . Kojima and G . Campbell , respectively , were used at 1/2000 , 1/1000 , 1/200 and 1/1000 , respectively . Guinea pig antibody raised against Roe ( an isoform encoded by the rotund/roe locus , see below ) was prepared by Eurogentec , from a full-length GST-Roe fusion ( a kind gift from D . del Alamo ) [37] extracted from E . coli and enriched by affinity chromatography . Note that ( i ) anti-Roe cross-reacts with Rn , because both proteins share a large common C-terminal region ( with the ZF domain ) and ( ii ) Roe protein isoform is not expressed in the leg [27] . Anti-Roe/Rn sera from terminal bleeds was used for immunocytology without purification at a 1/500 dilution . The Open Reading Frame ( ORF ) encompassing the C15 HD-encoding sequence ( codons 192 to 292 ) was amplified by PCR , using as a template a full length ( FL ) cDNA insert ( IP08859 ) , obtained from the Drosophila Genomic Resource Center ( DGRC ) , and inserted as an EcoRI/XhoI fragment into pGex4T , to yield the pGexC15HD construct . Similarly , Dll-encoding sequences ( full-length ORF or codons 120–188 ) were amplified by PCR , using as a template a full ORF cDNA insert [23] , and inserted as EcoRI/XhoI fragments into pGex4T , to yield the pGexDllFL and pGexDllHD constructs . Non-fused GST ( produced from empty pGex4T ) , GST-C15HD and both GST-Dll proteins were all expressed in BL21 [DE3] cells ( Novagen ) , from which soluble forms were purified on glutathione-Sepharose beads ( GE Healthcare ) and then isolated through elution with free glutathione . Purified proteins were used in EMSA , as described in [26] . Given that FL Bowl protein appeared toxic when expressed in E . coli cells , ZF domain encoding bowl ORF ( codons 220 to 382 ) was amplified by PCR , using as template a FL cDNA clone ( a kind gift from Sarah Bray ) , and inserted as an EcoRI/XhoI fragment into pGex4T , to yield the pGexBowlZF construct . The GST-BowlZF fusion was expressed in BL21 [DE3] cells , from which soluble forms were purified by glutathione-Sepharose affinity . For control EMSA and pull-down experiments , transformation with empty pGex4T was also performed , to produce and then purify soluble non-fused GST . EMSA experiments with Bowl and HD proteins were performed as described in [22] and [26] , respectively , using purified GST fusion proteins ( above ) and/or in vitro translated Al HD ( see below ) . Protein-DNA complexes were separated by electrophoresis on 6% native polyacrylamide gels and revealed by PhosphoImager detection . DNA Probes ( see Figs 2A and 4B for sequences ) were made from annealed synthetic oligonucleotides and 5’end-labeled with [32P]dCTP using Klenow DNA polymerase , as previously described [19] . However , a single labeled dCTP ( instead of 8 for the LAE-derived probes ) was incorporated into the Odd-binding probe , as described in [22] . The BarEnh probe is described in [26] . Pull down was down as previously described [38] , in duplicate . 35S protein probes ( C15HD , DllHD and AlHD ) were obtained from coupled in vitro transcription/translation ( TNT with rabbit reticulocyte extracts; Promega Inc . ) from DNA matrices generated by PCR , using 5’ oligonucleotides comprising each a T7 RNA polymerase promoter and a translation initiating ATG within an optimal Kozak context ( sequences of oligonucleotides are available upon request ) . Radiolabeled C15 and Dll HDs ( amino acids 192–292 and 120–188 , respectively ) were produced from full-length cDNA clones ( described above ) . In vitro translated Al HD ( amino acids 86–156 ) was produced from a full-length cDNA clone ( RE68460 ) , obtained from the Drosophila Genomic Resource Center ( DGRC ) . bric-à-brac locus sequences were recovered from dipteran genomic sequences and aligned as previously described [19] . | Limb morphogenesis is controlled by a well-known genetic cascade , mobilizing both cell signaling and tissue-specific transcription factors ( TFs ) . However , how their concerted action refines gene expression remains to be deciphered . It is thus crucial to understand how cell signaling inputs are integrated by transcriptional “enhancers” . The Drosophila leg provides a good paradigm to dissect the molecular mechanisms underlying gene regulation . Here , we used the bric-a-brac2 ( bab2 ) gene as a model to study the integrated regulation of patterning genes implicated in tarsal segmentation . bab2 expression in the leg primordium is dynamic and complex , going from initial broad distal expression to precisely positioned tarsal rings . By genetic and molecular analyses , we show here that the cell signaling-responding TFs C15 and Bowl interact directly with the limb-specific bab2 enhancer as a repressive duo . Moreover , C15 acts independently of its partner Aristaless through physical interaction with the Dll activator . We propose that Dll induces early circular bab2 expression pattern , then EGFR signaling-induced C15 in the distalmost cells competes with Dll for LAE binding and resolves bab2 pattern as a ring . Taken together our data shed light on how the concerted action of a quartet of transcription factors reshapes gene expression during limb proximo-distal axis development . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"legs",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"limbs",
"(anatomy)",
"cloning",
"transcription",
"factors",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"musculoskeletal",
"system",
"bioinformatics",
"proteins",
"gene",
"expression",
"molecular",
"biology",
"homeobox",
"biochemistry",
"signal",
"transduction",
"immunostaining",
"anatomy",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"protein",
"domains",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling"
] | 2017 | Tissue-specific enhancer repression through molecular integration of cell signaling inputs |
During Saccharomyces cerevisiae mating-type switching , an HO endonuclease-induced double-strand break ( DSB ) at MAT is repaired by recombining with one of two donors , HMLα or HMRa , located at opposite ends of chromosome III . MATa cells preferentially recombine with HMLα; this decision depends on the Recombination Enhancer ( RE ) , located about 17 kb to the right of HML . In MATα cells , HML is rarely used and RE is bound by the MATα2-Mcm1 corepressor , which prevents the binding of other proteins to RE . In contrast , in MATa cells , RE is bound by multiple copies of Fkh1 and a single copy of Swi4/Swi6 . We report here that , when RE is replaced with four LexA operators in MATa cells , 95% of cells use HMR for repair , but expression of a LexA-Fkh1 fusion protein strongly increases HML usage . A LexA-Fkh1 truncation , containing only Fkh1's phosphothreonine-binding FHA domain , restores HML usage to 90% . A LexA-FHA-R80A mutant lacking phosphothreonine binding fails to increase HML usage . The LexA-FHA fusion protein associates with chromatin in a 10-kb interval surrounding the HO cleavage site at MAT , but only after DSB induction . This association occurs even in a donorless strain lacking HML . We propose that the FHA domain of Fkh1 regulates donor preference by physically interacting with phosphorylated threonine residues created on proteins bound near the DSB , thus positioning HML close to the DSB at MAT . Donor preference is independent of Mec1/ATR and Tel1/ATM checkpoint protein kinases but partially depends on casein kinase II . RE stimulates the strand invasion step of interchromosomal recombination even for non-MAT sequences . We also find that when RE binds to the region near the DSB at MATa then Mec1 and Tel1 checkpoint kinases are not only able to phosphorylate histone H2A ( γ-H2AX ) around the DSB but can also promote γ-H2AX spreading around the RE region .
Saccharomyces mating-type switching occurs through a DSB-initiated intrachromosomal gene conversion event at MAT , using one of two donors on chromosome III , HML and HMR ( Figure 1A ) [1]–[3] . Switching is initiated by expression of the site-specific HO endonuclease that cleaves only one site in the yeast genome , MATa or MATα . The unexpressed mating-type genes in HMLα and HMRa also contain HO cleavage sites , but they are not cut because these regions are heterochromatic [4]–[6] . Although either HMLα or HMRa can be used to repair a DSB at MAT , there is a strong mating type-dependent preference for the choice of the two donors . In MATa cells , HMLα is preferentially chosen for repair , about 85–90% of the time , whereas MATα cells strongly prefer HMRa , about 95% [3] , [7]–[9] . Donor preference is not altered if the mating-type genes encoded in the Y region are changed , e . g . if HMR carries Yα instead of Ya or if HML is replaced with HMR [7] , [8] . Donor preference in MATa depends on an approximately 275-bp Recombination Enhancer ( RE ) , located 17 kb to the right of HML [10] , [11] . One important aspect of donor preference is that MATa cells activate a large ( ∼40 kb ) region near the left end of chromosome III , so that a donor within this region is strongly preferred over HMR [8] . RE is responsible for this activation along the entire left arm of chromosome III [11] , [12] . Donor preference does not change if the cis-acting silencer sequences around HML or HMR are removed [13] . In addition , RE is not limited to the special features of MAT switching . If a leu2 allele is inserted in place of HML , its success in recombining with a different leu2 allele , either near MAT or even on another chromosome , is 20–30 times higher in MATa than in MATα and is RE-dependent [8] , [12] . RE is “portable;” that is , it will work in other chromosome contexts . When HML , HMR and MATa are all inserted on chromosome V , HML usage increases significantly when RE is inserted nearby [12] . In addition , in MATa cells where RE promotes HML , the usage of HMR can be markedly increased by placing a second RE near HMR [11] , [12] . In MATα cells , RE is inactivated by binding of the Matα2-Mcm1 repressor complex , which leads to formation of highly organized nucleosomes covering the RE region but not extending into adjacent gene regions [8] , [10] , [14] . In MATa cells , RE exhibits several nuclease hypersensitive sites when Mcm1 binds RE in the absence of the Matα2 protein ( which is not expressed in MATa cells ) . In addition to the Matα2-Mcm1 operator region , RE is composed of several evolutionarily conserved chromatin domains [14] , several of which were shown to contain putative binding sites for the Fkh1 transcription factor [15] . A conserved SCB ( Swi4/Swi6 cell cycle box ) is also present in Region C of RE [16] . Both Fkh1 and Swi4/Swi6 regulate donor preference by binding to RE in MATa cells [15]–[17] . Despite the presence of these transcription factors , there are no open reading frames adjacent to RE , although there is an adjacent noncoding RNA [18] . The DNA repair proteins Ku70 and Ku80 have a small effect on MATa donor preference that may be caused by the role of these proteins in localizing HML to the nuclear periphery [19] . Deleting the Chl1 helicase also causes a small reduction of MATa donor preference without affecting MATα choice [16] , [20] . Despite the identification of several proteins that bind to RE , it is still not clear how RE regulates donor preference . Previously we showed that RE could be deleted and replaced with small modules derived from RE . Notably 4 tandem copies of a 22-bp sequence containing a putative Fkh1 binding site were sufficient to increase HML usage to >60% ( where the use of HML in REΔ is 5% ) ; this increased preference for HML is abolished in fkh1Δ [15] . To further explore the mechanism of RE regulation , we replaced RE with four LexA operators and found that a LexA-Fkh1 fusion strongly promotes HML usage . Using this system , we dissect Fkh1 and find out that RE activity depends on the phosphothreonine binding motif of the FHA domain of Fkh1 and not on its forkhead domain . We show that LexA-FHAFkh1 becomes associated with the chromatin surrounding the MAT only after DSB induction . This interaction is seen even in a donorless strain , demonstrating that the FHA-mediated regulation is a break-dependent but repair-independent process . MATa donor preference is partially dependent on casein kinase II but not on two checkpoint kinases , Mec1 and Tel1 . We propose that the FHAFkh1 domain regulates donor preference by physically interacting with phosphorylated threonines on histones or other bound proteins surrounding the DSB during mating-type switch .
All strains in this study are derived from XW652 [11] , which carries HMLα , MATa and HMRα-B on chromosome III ( Figure 1A ) . HMRα-B contains a single base pair change that creates a BamHI site [8] . After galactose-induced expression of HO , MATa can be repaired to MATα or MATα-B , using HMLα or HMRα-B , respectively . Following HO induction for 60 min , HO expression was repressed by the addition of 2% dextrose and the ratio of switching to MATα or MATα-B was checked after 24 h . Donor preference could be measured either by Southern blot [8] or by a PCR-based assay in which the combination of MATα or MATα-B PCR products is digested with BamHI ( Figure 1B ) . PCR-based assay showed 85% usage of HMLα for XW652 but ≤10% for RE-deleted XW676 ( Figure 1C ) . Fkh1 is involved in the regulation of donor preference through direct interaction with RE [15] , [16] . To further explore the role of Fkh1 , we constructed a strain ECY406 by replacing RE with four LexA operators ( Figure 2A ) . In an otherwise wild type background , HML usage in ECY406 was less than 5% as expected for a deletion of RE ( Figure 2B ) . We then constructed a plasmid pEC16 that constitutively expresses a LexA-Fkh1 fusion protein from an ADH1 promoter of pAT4 [21] . The LexA-Fkh1 sequences from pEC16 were stably integrated at the arg5 , 6 locus of ECY406 to generate a new strain ECY457 ( Figure 2A ) . Expression of LexA-Fkh1 in ECY457 was able to up-regulate donor preference to around 32% presumably by binding to four LexA operators replacing RE ( Figure 2B ) , whereas the use of HML was less than 5% when LexA alone was expressed ( data not shown ) . This result demonstrates that regulation of donor preference by Fkh1 does not require the binding of Mcm1 or Swi4/Swi6 to their specific sites in the normal RE sequences . We noted further that the Fkh1 moiety in the LexA-Fkh1 fusion remained functional even with normal RE , as it could complement a fkh1Δ mutant in YJL017 by up-regulating donor preference to 68% ( Figure 2C ) . Fkh1 contains two conserved domains: a forkhead-associated ( FHA ) and a forkhead DNA binding domain ( Figure 3A ) [22] , [23] . To understand roles of different domains of Fkh1 in the regulation of donor preference , we prepared three plasmid constructs by fusing LexA of pAT4 with different regions of Fkh1: pJL4 for LexA-FHA ( aa 1–230 of Fkh1 ) , pJL5 for LexA-interdomain ( aa 163–302 ) , and pJL6 for LexA-forkhead ( aa 231–484 ) ( Figure 3A ) . The LexA fused sequences from these plasmids were integrated at arg5 , 6 locus of ECY406 to generate strains YJL019 , YJL020 , and YJL021 , respectively ( Figure 3A ) . These three strains and ECY457 ( Figure 2A ) all have a wild-type Fkh1 , which is not functional in donor preference because Fkh1 cannot bind to REΔ::LexABD4 . Southern blots revealed that only YJL019 could re-establish donor preference to 90% , whereas YJL020 and YJL021 failed to increase HML usage ( Figure 3B ) . This result suggests that the FHA domain may play a critical role in the regulation of donor preference . We noted that donor preference regulated by LexA-FHAFkh1 ( 90% donor preference for YJL019; Figure 3B ) was much higher than that by LexA-Fkh1 ( 32% donor preference for ECY457; Figure 2B ) . We suggest two possible explanations for this difference . First , two DNA binding domains ( LexA and the forkhead DNA binding domain ) are present in LexA-Fkh1 , whereas only one ( LexA ) is present in LexA-FHAFkh1 . Therefore , the LexA-Fkh1 fusion protein likely binds multiple sites in yeast genome , which could mean that less fusion protein is available for regulating donor preference . In contrast , because there is only one DNA binding domain for LexA-FHAFkh1 , all fusion protein will be available for donor preference regulation . A second possible reason is that the FHAFkh1 domain is more exposed in LexA-FHAFkh1 than in LexA-Fkh1 when both fusion proteins bind to four LexA operators replacing RE . The presence of a forkhead domain in LexA-Fkh1 could interfere with regulation of the FHAFkh1 domain in donor preference , whereas this kind of interference is not present in LexA-FHAFkh1 . The FHA ( forkhead-associated ) domain is a small protein module that can preferentially bind to phosphothreonine residues on proteins [22] , [24] , [25] . FHA domains have been found in a wide range of proteins , such as kinases , phosphatases and transcription factors [23] , [26] . To confirm that the FHAFkh1 domain was responsible for increasing HML usage , LexA-FHA-R80A from pJL8 was integrated into the arg5 , 6 locus of ECY406 to generate a strain YJL094 ( Figure 3A ) . Preferential usage of HML was completely abolished using LexA-FHA-R80A ( Figure 3C ) , which carried a non-functional FHA domain [22] , [23] . Thus , the phosphothreonine-binding motif of the FHA domain plays a critical role in the regulation of donor preference . We employed Chromatin Immunoprecipitation ( ChIP ) to ask if LexA-FHAFkh1 could associate with the region around MAT before or after induction of a DSB . Using an anti-LexA antibody , we showed that LexA-fused FHAFkh1 physically interacted with the MAT region after DSB induction in a strain lacking HML and HMR ( Figure 4A ) , so that DSBs could not be repaired by homologous recombination . We observed a >10-fold increase in ChIP signals within about 5 kb on either side of the HO cleavage site at the MAT , whereas no significant signal could be detected using primer pairs that amplify regions further away from the HO site ( Figure 4B ) . Therefore , the LexA-FHAFkh1 fusion protein physically interacted with the DSB-cut MAT through a repair-independent mechanism , which suggests that LexA-FHAFkh1 or RE can be used to stimulate recombination between any two homologous sequences in budding yeast [11] . We note that the localization of LexA-FHAFkh1 binding is quite different from the roughly 50-kb phosphorylation of histone H2A-S129 ( γ-H2AX ) on either side of the DSB [27] , [28] , although it is similar to a second damage-induced modification , which is the casein kinase II-dependent phosphorylation of H4-S1 ( Figure 4D ) [29] . Given that RE activity was completely abolished in the strain YJL094 carrying LexA-FHA-R80A ( Figure 3C ) , it was not unexpected that the ChIP assay showed no physical interaction between LexA-FHA-R80A and the MAT region after DSB induction ( Figure 4C ) ; However , the LexA-FHA-R80A fusion protein still strongly associated with REΔ::LexABD4 likely due to the presence of the LexA domain ( Figure 4C ) . These data strongly support the idea that the FHA domain of Fkh1 regulates donor preference by physically interacting with the MAT region during mating-type switch , and these interactions fully depend on the phosphothreonine binding motif of the FHAFkh1 domain . Because the FHAFkh1 domain regulates donor preference via a repair-independent but break-dependent mechanism , it suggests that FHAFkh1 domain or RE can be used to facilitate recombination between any homologous sequences in yeast genome . Previously we showed that RE stimulated leu2 heteroallele spontaneous recombination when one of the alleles was situated in place of HML [11] . In that case , the nature and position of the initiating DNA lesions were unknown . Here we integrated a leu2::HOcs construct at can1 locus on chromosome V , so HO-induced DSBs can recombine with a LEU2 locus placed near RE on chromosome III ( Figure 5 ) [30] . In one assay , LEU2 on chromosome III could be used as a donor to repair HO-induced DSBs on chromosome V in competition with a leu2-K donor inserted at ura3 , which is 85 kb from the leu2::HOcs ( Figure 5A ) . The leu2-K allele was created by ablating KpnI site in LEU2 [31] . As shown in Figure 5A , the proportion of repair events using the interchromosomal donor was more than 50% when RE was present but fell to less than 10% when RE was deleted . In a second assay , the LEU2 on chromosome III was the only possible donor for DSB repair . This construct allowed us to ask whether RE stimulated recombination by facilitating the earliest step , the search for homology by Rad51 recombinase bound to the resected end of the DSB . We measured the time at which Rad51 became associated with the donor ( i . e . when strand invasion had occurred ) by a ChIP assay analogous to that used to assay strand invasion kinetics during MAT switching [32] , [33] . As seen in Figure 5B , the kinetics of Rad51 association with the LEU2 donor was significantly faster when RE was present . The presence of RE also assured that the proportion of cells that completed repair was 72% compared to 37% when RE was deleted . The percentage of completed repair was determined by comparing survival on galactose plates with that on dextrose plates where HO was not induced . γ-H2AX rapidly forms around the site of a DSB , dependent on either Mec1 or Tel1 checkpoint protein kinase [27] , [28] . If RE bound to regions around the DSB , would γ-H2AX also form around RE region ? To address this question , we used ChIP with anti-γ-H2AX antibody to examine the phosphorylation of histones around RE following initiation of a DSB . γ-H2AX formed over a large domain around MAT following the induction of a DSB within 15–60 min ( Figure 6A ) . Surprisingly , γ-H2AX also appeared around RE at 1 hr after HO induction in MATa cells . As predicted , there was no similar modification around RE in MATα cells , where RE is repressed ( Figure 6B ) . Moreover , the kinetics of γ-H2AX modification around RE was slower than around MAT , consistent with the idea that RE first had to be recruited to the DSB before this modification could take place ( Figure 6A , 6C ) . Finally , by using both mec1Δ sml1Δ and tel1Δ derivatives of JKM139 , we showed that either checkpoint kinase was capable of carrying out γ-H2AX modification around RE ( Figure 6D ) . These data provide additional supporting evidence of a direct RE-to-MAT contact after DSB induction and support the model that the binding of RE to MAT is the basis of bringing HML into close proximity . In addition , these data show for the first time that a region not suffering a DSB can be modified by both checkpoint kinases if this region is brought close to the DSB site . Our data strongly argue that the FHA domain of Fkh1 , clustered at the normal RE or REΔ::LexABD4 , interacts with phosphorylated residues in the region surrounding the DSB . The most obvious candidates are histones that are phosphorylated after DSB induction , including H4-S1 [29] and histone H2A-S129 ( γ-H2AX ) . The possibility that H4-S1 could be involved was made more attractive by our finding that this modification is confined to the first 10 kb around a DSB , much more restricted than γ-H2AX ( Figure 4D ) . We constructed a strain YJL102 , carrying the h4-S1A in HHF2 and deleted for HHF1; however this alteration had no effect on donor preference ( Figure 7A ) . In addition , phosphorylation of H2A-S122 , H2A-T126 and H2A-S129 have been implicated after MMS-induced DNA damage [34] . To test these H2A modifications , we constructed a strain YJL121 by deleting endogenous HTA1-HTB1 and HTA2-HTB2 and complementing by a plasmid carrying hta1-S122A-T126A-S129A-HTB1 , but donor preference was not affected ( Figure 7A ) . We have directly tested whether post-translational modifications of the N-terminal tail of histones H3 or H4 are implicated in donor preference . In addition to H4-S1 , several other sites have been reported to be phosphorylated during the cell cycle , such as H3-T3 , H3-S10 and H3-S28 [35]–[37] , which might also be targets for modification after a chromosome break . In particular , we constructed a strain YJK340 , in which HHF1-HHT1 was deleted with NAT . Then , the remaining copy of H3 gene was modified to carry a deletion of the first 32 amino acids or HHF2 was modified to lack the first 16 amino acids of histone H4 . We found that the H3 tail deletion unsilenced HML but not HMR ( i . e . cells became non-mating by expressing both HMLα and MATa ) ; hence we replaced the Yα sequences at HML with HPH as previously described [13] . This modification also deleted part of the HO cleavage site at HML , so only MATa would be cleaved by HO . When HO was induced at MAT , there was no change in donor preference , as 73 of the 82 ( 89% ) switched products were derived from hml::HPH and only 9 ( 11% ) were MATα-B , derived from HMRα-B . In the case of the H4 N-terminal truncation , both HMLα and HMRα-B become unsilenced [38]; thus to look at donor preference , we replaced HML's Yα with HPH and HMR's Yα-B with KAN [13] . Among 39 colonies that switched from MATa , 34 ( 87% ) used hml::HPH whereas only 5 ( 13% ) recombined with hmr::KAN . Therefore , deleting the tail of histone H3 or H4 had no effect on donor preference in MATa cells . Although Mec1 and Tel1 can phosphorylate histone H2A in the region surrounding RE when it is brought in conjunction with MAT , these checkpoint kinases are not responsible for promoting MAT donor preference . We constructed a strain YJL054 ( mec1Δ tel1Δ sml1Δ ) derived from XW652 . We noted that because Mec1 or Tel1 was required for efficient clipping of the Ya tail to enable the completion of switching to MATα or MATα-B [39] , there was a strong reduction in the switching efficiency ( data not shown ) , but the proportion of MATα to MATα-B was unaltered in YJL054 ( Figure 7B ) . This conclusion that Mec1 and Tel1 are not involved in the regulation of donor preference was further supported by our data that donor preference was not altered in YJL121 , in which histone H2A-S129 was mutated to alanine ( Figure 7A ) . Casein kinase II phosphorylates serine 1 ( S1 ) of histone H4 after exposure to MMS- and phleomycin-induced DSBs [29] and after HO-induced DSBs ( Figure 4D ) . Casein kinase II is required for cell cycle progression in budding yeast and essential for cell viability [40] . We constructed a strain lacking the chromosomal CKA1 and CKA2 genes but carrying a pRS315 plasmid with a temperature-sensitive cka2-8 allele ( Figure 7B ) . Cells were grown overnight at the permissive temperature of 25°C and then shifted to the restrictive temperature of 37°C . Inactivation of Cka2 leads to 43% use of HML ( YJL119 ) compared to 87% donor preference in control YJL019 ( Figure 7B ) . This result indicates that casein kinase II activity is required for Fkh1-dependent regulation of donor preference . Because the N-terminal truncation of H4 ( including H4-S1 ) has no effect on HML usage , it is likely that casein kinase II phosphorylates some other targets , on a histone or another protein , which is involved in donor preference regulation . However , the fact that 43% donor preference is still significantly higher than 10% observed in RE-deleted strains ( Figure 1C ) suggests that multiple kinases may be involved in the regulation of donor preference .
We have shown that the phosphothreonine binding motif of the FHA domain of Fkh1 plays a critical role in the regulation of donor preference ( Figure 3 ) . A strong physical association between the FHAFkh1 domain bound at the RE region and MAT is readily seen , but only after a DSB is induced . This interaction is independent of the presence of an adjacent homologous HML donor ( Figure 4 ) . Conversely , the region surrounding RE can be phosphorylated by Mec1 and Tel1 kinases only after DSB induction in MATa but not in MATα strains ( Figure 6 ) , again suggesting that these regions can come into physical contact when there is a DSB at MAT and RE is active . RE's activity does not depend on any of the special features of MAT switching such as HML or HMR silencing [13] or HO cleavage [11] , [15] . Consequently RE is able to improve the use of an ectopic donor in repairing a DSB on a different chromosome . Normally , a DSB will be preferentially repaired by a donor on the same chromosome in competition with an ectopic donor , but if the ectopic donor is located near RE , more than half of gene conversions use the interchromosomal donor ( Figure 5A ) . Although our data and those from others show that HML is not constitutively much closer to MATa than HMR is ( i . e . in the absence of HO cleavage ) [41]–[43] , the data we present here suggest that such a reorganization will occur after a DSB is created . Taken together , our data suggest a simple model for RE action ( Figure 7C ) . After the induction of a DSB , casein kinase II and possibly other kinases modify some proteins bound near the DSB . These modifications , most likely phosphothreonines , are clustered near the DSB and can be bound by FHAFkh1 domains tethered at RE . This binding effectively tethers HML within about 20 kb of the DSB whereas HMR remains 100 kb away . Thermodynamic considerations argue that this close proximity is sufficient to explain why HML should be used 90% of the time for DSB repair in MATa cells [13] . This model also explains how RE can act over a long region of the left arm of chromosome III [8] , although with diminishing effect [12] , by this tethering mechanism . The model we propose argues that RE should be portable and able to stimulate the use of any homologous donor in a DSB repair mechanism . Our previous work has shown that RE is portable , as it is able to activate HML use when both are inserted on chromosome V [12] . Moreover , if a copy of RE is inserted near HMR in a MATa strain that also has RE near HML , then HMR usage is increased to about 50% ( E . C . , S . -Y . Tay and J . E . H . , unpublished ) . The ectopic recombination experiment presented here shows that RE can act efficiently on non-MAT sequences for DSB repair ( Figure 5A ) . We note that we have previously shown that RE could stimulate spontaneous recombination between leu2 heteroalleles when one of them was located close to the RE [11] , [12] . The results we report here suggest that a large proportion of spontaneous recombination events may be triggered by DSBs or that the same phosphorylated protein attracting the attention of RE during DSB repair also accumulates at the lesions that stimulate spontaneous recombination . At present , we have not yet identified the phosphothreonine target for the FHA domain of Fkh1 . We have ruled out a number of candidates , including γ-H2AX , N-terminal tails of histones H3 and H4 , as well as Mre11 and Sae2 , two proteins involved in DSB end-binding and initiating 5′ to 3′ resection ( C . -S . L . , J . E . H . , unpublished observations ) . Studies using peptide libraries and immunoprecipitation of the FHAFkh1 domain after DSB induction are underway . Aparicio group has recently made the intriguing finding that Fkh1 and Fkh2 proteins play a key role in the activation and clustering of early origins of replication in budding yeast [44] . This regulation involves a cis-acting association of these two forkhead proteins with proteins at origins . It will be interesting to ask if the FHA domain of Fkh1 plays an important role in this regulation . Another important finding emerging from our work is that two DNA damage checkpoint kinases , Mec1/ATR and Tel1/ATM , can act to phosphorylate distant DNA sequences when they are tethered in the vicinity of the DSB . As shown in Figure 6 , the γ-H2AX modification spreads around the RE region , but with significantly delayed kinetics compared with the modification around MAT , consistent with the idea that RE has to first recognize and bind to phosphorylated residues in the vicinity of the DSB at MAT . How these checkpoint kinases act on their target sequences is not yet firmly established . Mammalian ATM has been shown to be activated by intermolecular autophosphorylation and dimer exchange , which would suggest that activated ATM would initially form a “cloud” of activated kinases around the site where the kinases were associated with the DSB ends [45] , [46] . In the case of Tel1/ATM , the association with the DSB is via its association with the MRX/MRN proteins [47] , [48]; in the case of Mec1/ATR , by its association its partner protein Ddc2/ATRIP with RPA bound to ssDNA at the resected DSB end [49] , [50] . In budding yeast , the spreading of γ-H2AX from the DSB site is consistent with that the tethered kinases interact with phosphorylating histones on the adjacent chromosomal segment in a manner , which is similar to the contact of chromosomal regions as measured in chromosome conformation capture experiments [51] . Spreading of γ-H2AX further along the chromosome occurs more slowly and apparently depends on the continuing 5′ to 3′ resection of the DSB ends , generating ssDNA , as it depends only on Mec1 [27] , [28] . Here we show that histones in another distant chromosomal region , brought into proximity with the DSB by RE , can also be efficiently phosphorylated – and by both Mec1 and Tel1 . This result is different from the slow addition of γ-H2AX to regions further from the DSB , which depends on continuing 5′ to 3′ resection of the DSB ends and can only be performed by Mec1 [27] , [28] . We have also observed γ-H2AX spreading onto a different chromosome during the ectopic recombinational repair of a DSB , when these two regions are brought together by Rad51-mediated strand invasion ( K . L . and J . E . H . , unpublished observations ) .
All strains except when noted were derived from strain XW652 ( ho ade3::GAL::HO HMLα RE MATa HMRα-B ura3-52 lys5 leu2-3 , 112 trp1::hisG ) carrying a galactose-inducible HO endonuclease integrated at the ADE3 locus [11] . Strains are pre-cultured in YP-lactate medium until cell density reaches about 5∼8×106 per ml . Galactose induction is performed for 1 hour and stopped by the addition of 2% dextrose . Construction of ECY406 ( Figure 2A ) : Four LexA operators are amplified from pSH18-34 [52] using primers BglIILexAU ( 5′-cga cga gat cta tac ata tcc ata tct aat ctt acc-3′ ) and BglIILexAL ( 5′-gct gca gat ctc taa tcg cat tat cat ccc tcg a-3′ ) . Then PCR products were digested with BglII and subcloned into the BamHI site of pKS58 to generate pEC15 . The SphI-digested pEC15 ( marked with “LEU2” ) was transformed into XW676 ( ho ade3::GAL::HO HMLα REΔ::URA3 MATa HMRα-B ade1 leu2 trp1 ura3-52 ) to replace REΔ::URA3 with four LexA operators to generate a strain ECY405 . Then , REΔ::LexABD4-LEU2 from ECY405 was replaced with REΔ::LexABD4-KAN to generate a strain ECY406 ( Figure 2A ) via transformation using PCR fragments amplified from pJH1894 with primers leu2KanU ( 5′-gag aac ttc tag tat atc cac ata cct aat att att gcc tta tta aaa atc agc tga agc ttc gta cgc-3′ ) and leu2KanL ( 5′-tac gtc gta agg ccg ttt ctg aca gag taa aat tct tga ggg aac ttt cag cat agg cca cta gtg gat ctg-3′ ) . ECY457 ( Figure 2A ) is constructed by transforming ECY406 with PCR fragment arg5 , 6::LexA-Fkh1 obtained with primers pAT4UII ( 5′-atg cca tct gct agc tta ctc gtc tcg aca aag aga ctt aac gct tcc aaa ttc cat ttt gta att tcg tgt cg-3′ ) and pAT4LII ( 5′-tca gac acc aat aat ttt att ttc agg gat acc agc ata ctc tcc ata aca agg gaa caa aag ctg gag c-3′ ) on the plasmid pEC16 . Using a similar strategy , PCR products arg5 , 6::LexA-FHA ( from pJL4 ) , arg5 , 6::LexA-interdomain ( from pJL5 ) , arg5 , 6::LexA-forkhead ( from pJL6 ) and arg5 , 6::LexA-FHA-R80A ( from pJL8 ) are transformed into ECY406 to generate YJL019 , YJL020 , YJL021 and YJL094 , respectively ( Figure 3A ) . YJL084 was made by transforming YJL019 ( Figure 3B ) with BamHI digested pJH1250 to delete HML using the URA3 marker . YJL110 ( Figure 4A ) is made by transforming YJL084 with BsaI-digested pJH2039 to delete HMR using the NAT marker . Yeast strains with H3 or H4 N-terminal truncation were constructed by sequential transformations of JKM139 [32] . The HHF1-HHT1 allele of JKM139 was first knocked out by NAT-MX cassette to generate a strain YJK340 ( ho ade3::GAL::HO hmlΔ::ADE1 RE MATa hmrΔ::ADE1 ade1 leu2-3-112 lys5 ura3-52 trp1::hisG hhf1-hht1Δ::NAT ) . Then , YJK340 was transformed with linearized plasmid carrying hht2-hhf2 mutant alleles linked to URA3 marker to replace endogenous HHT2-HHF2 allele . HHT2 was modified to lack the first 32 amino acids of histone H3 or HHF2 was modified to lack the first 16 amino acids of histone H4 . To prepare for mating-type switching assay , HMLα and HMRα-B from XW652 were crossed into a yeast strain with H3 or H4 N-terminal truncation . All strains except when noted in this study are derived from XW652 ( ho ade3::GAL::HO HMLα RE MATa HMRα-B ura3-52 lys5 leu2-3 , 112 trp1::hisG ) [11] . The C→A change at position 658 of Yα creates a BamHI restriction site ( HMRα-B ) , which is absent in HMLα [8] . Donor preference ( HML usage ) is calculated using the formula ( MATα/ ( MATα+MATα-B ) for all XW652 derived strains ( Figure 1A ) . The measurement of donor preference via Southern blot was described previously [8] . Southern signals were quantified using ImageQuant V1 . 2 ( Molecular Dynamics ) . Because there is only 1-bp difference between two repaired products ( MATα and MATα-B ) , we have developed a PCR-based assay to measure donor preference . The presumption is that PCR amplification efficiency is almost identical for MATα and MATα-B because there is only 1-bp difference [8] . Around 10 ng of genomic DNA isolated from galactose-induced colonies will be used for PCR amplification . Two primers Yalpha105F ( 5′-gcc cac ttc taa gct gat ttc aat ctc tcc-3′ ) and MATdist-4R ( 5′-cct gtt ctt agc ttg tac cag agg-3′ ) can only amplify MATα or MATα-B , but not MATa , HMLα or HMRα-B due to sequence specificities of these two primers ( Figure 1B ) . Although amplified PCR products are the mixture of MATα-B and MATα , only one 1470-bp band can be visualized on DNA agarose gel prior to digestion . PCR products are then purified and subsequently digested with BamHI . The digested PCR products will be checked on DNA agarose gel . MATα product will remain as the 1470-bp band , whereas MATα-B product is digested into two smaller bands with different sizes ( 550-bp and 920-bp ) ( Figure 1C ) . Donor preference is determined by comparing intensities of MATα and MATα-B after the agarose gel is stained with ethidium bromide . To study if Fkh1 can regulate donor preference in our LexA system , we construct a LexA-Fkh1 fusion plasmid ( pEC16 ) carrying the coding sequence of Fkh1 . Fkh1 coding sequence is PCR amplified from XW652 genomic DNA using primers XmaIFkh1U ( 5′-tcg cga ccc ggg gat ccg tat gtc tgt tac cag tag gg-3′ ) and PstIFkh1L ( 5′-gca cga cct gca gtc aac tca gag agg aat tgt tca cg-3′ ) . The amplified PCR product is digested with XmaI and PstI and then subcloned into a pre-digested pAT4 [21] to generate the plasmid pEC16 . To address different roles of Fkh1 domains in the regulation of donor preference , three regions of Fkh1 are subcloned into pAT4 ( Figure 3A ) . The FHA domain of Fkh1 is amplified via PCR using primers XmaIFkh1U and PstIFkh1-690L ( 5′-gca cga cct gca gta ggt ggt cca gct gtt gta atc g-3′ ) . The interdomain region is amplified using primers XmaIFkh1-487U ( 5′-tcg cga ccc ggg gat cgg tgt gca aat gat ctt tat at-3′ ) and PstIFkh1-906L ( 5′-gca cga cct gca gga tat atc tgt ttt cat cca gc-3′ ) . The forkhead domain is amplified using primers XmaIFkh1-691U ( 5′-tcg cga ccc ggg gat cca cac ccc att atc gtc atc at-3′ ) and PstIFkh1L . These PCR products are then digested with XmaI and PstI , and subcloned into a pre-digested pAT4 , to generate three fusion plasmids pJL4 , pJL5 and pJL6 , respectively . Quickchange Multi Site-Directed Mutagenesis Kit ( Catalog # 200515 , Stratagene , La Jolla , CA ) was used to mutate the FHA domain of pJL4 . Two primers Fkh1-Arg80 ( 5′-tta gaa gtt acc att ggt gcg aac aca gac agc ttg aac-3′ ) and pAT4-940R ( 5′-ctt tgc cag aca aga aca ccg cat-3′ ) were used to synthesize mutant strand from pJL4 . Fkh1-Arg80 shares two-base mismatches with Fkh1 and pAT4-940R perfectly matches pJL4 . The mutated plasmid pJL8 ( pLexA-FHA-R80A ) was confirmed by direct sequencing . Procedures for ChIP analysis were described previously [15] . Rabbit anti-LexA polyclonal antibody ( Catalog no . 39184 ) used in ChIP assay is purchased from “Active Motif” company ( Carlsbad , CA ) . LexA ChIP signals are quantified with real-time PCR using a Chromo 4 machine from MJ Research . The linearity of PCR signals is monitored with r-square value of a calibration curve , which is prepared using a series of dilutions of the 0 hr input sample . IP signal is determined by comparing to the calibration curve , and then normalized to the IP signal of a control locus CEN8 . PCR primer sequences around the MAT , RE and the ectopic leu2::HOcs are available on request . | Mating-type gene switching occurs by a DSB–initiated gene conversion event using one of two donors , HML or HMR . MATa cells preferentially recombine with HML whereas MATα cells choose HMR . Donor preference is governed by the Recombination Enhancer ( RE ) , located about 17 kb from HML . RE is repressed in MATα cells , whereas in MATa RE binds several copies of the Fkh1 protein . We replaced RE with four LexA operators and showed that the expression of LexA-Fkh1 fusion protein enhances HML usage . Donor preference depends on the phosphothreonine-binding FHA domain of Fkh1 . LexA-FHAFkh1 physically associates with chromatin in the region surrounding the DSB at MAT . We propose that RE regulates donor preference by the binding of FHAFkh1 domains to phosphorylated sites around the DSB at MAT , thus bringing HML much closer than HMR . FHAFkh1 action partially depends on casein kinase II but not on the DNA damage checkpoint kinases Mec1 and Tel1 . We also find that , when RE binds to the MAT region , phosphorylation of histone H2A ( γ-H2AX ) by Mec1/Tel1 not only surrounds the DSB but also spreads around RE . This is the first demonstration that γ-H2AX can spread to contiguous , but undamaged , chromatin . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"model",
"organisms",
"genetics",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"biophysics",
"genetics",
"and",
"genomics"
] | 2012 | Regulation of Budding Yeast Mating-Type Switching Donor Preference by the FHA Domain of Fkh1 |
For all animals , the taste sense is crucial to detect and avoid ingesting toxic molecules . Many toxins are synthesized by plants as a defense mechanism against insect predation . One example of such a natural toxic molecule is l-canavanine , a nonprotein amino acid found in the seeds of many legumes . Whether and how insects are informed that some plants contain l-canavanine remains to be elucidated . In insects , the taste sense relies on gustatory receptors forming the gustatory receptor ( Gr ) family . Gr proteins display highly divergent sequences , suggesting that they could cover the entire range of tastants . However , one cannot exclude the possibility of evolutionarily independent taste receptors . Here , we show that l-canavanine is not only toxic , but is also a repellent for Drosophila . Using a pharmacogenetic approach , we find that flies sense food containing this poison by the DmX receptor . DmXR is an insect orphan G-protein–coupled receptor that has partially diverged in its ligand binding pocket from the metabotropic glutamate receptor family . Blockade of DmXR function with an antagonist lowers the repulsive effect of l-canavanine . In addition , disruption of the DmXR encoding gene , called mangetout ( mtt ) , suppresses the l-canavanine repellent effect . To avoid the ingestion of l-canavanine , DmXR expression is required in bitter-sensitive gustatory receptor neurons , where it triggers the premature retraction of the proboscis , thus leading to the end of food searching . These findings show that the DmX receptor , which does not belong to the Gr family , fulfills a gustatory function necessary to avoid eating a natural toxin .
Taste is essential to distinguish between nutritious and toxic substances . To avoid eating toxins , animals are able to detect them by using a repertoire of taste receptors [1] . Although it is recognized that a bitter taste sensation is critical to avoid toxic substances [2] , [3] , the cellular and molecular mechanisms that have been established during evolution to detect a toxin are not well understood . In particular , how a receptor becomes tuned to a toxin is not well documented , mainly because the structure of its ligand binding pocket ( LBP ) and the evolutionary relationship with the ancestor receptor are not known . In Drosophila , the family of gustatory receptors ( Grs ) is predicted to consist of 68 genes [4] , [5] . This family of receptors , which consist of seven transmembrane domain proteins , is characterized by a very high level of amino acid divergence , showing as little as 8%–12% amino acid identity [5] . Such diversity suggests that the Gr family could cover the entire range of taste-receptive capability of the fly . Nevertheless , the extreme divergence within this family does not exclude the possibility of evolutionarily independent insect taste receptors not belonging to the Gr family . To date , only few receptors of the Gr family have been associated with a specific taste molecule: for example , the receptor for the sugar trehalose , called Gr5a [6] , and the bitter compound caffeine coreceptors , called Gr66a and Gr93a [7] , [8] . Plants synthesize many toxic molecules as defense mechanisms against predation [9] , [10] . A number of such toxic compounds are nonprotein amino acids [11] , [12] . The best-characterized example of nonprotein amino acid that plays a defensive role is l-canavanine ( 2-amino-4-guanidinooxybutyric acid ) [13]–[15] , which is massively accumulated in the seeds of many legumes ( up to 143 mM in Medicago sativa [16] ) . l-Canavanine is a natural insecticide because it is structurally similar enough to l-arginine ( Figure S1 ) to interfere with l-arginine metabolism and to be incorporated by arginyl-tRNA synthase in de novo proteins resulting in dysfunctional proteins [17]–[19] . Thus , these properties of l-canavanine render it a highly toxic secondary plant constituent [15] . To deal with this natural poison , some insects have generated several adaptive strategies . Indeed , the tobacco budworm Heliothis virescens uses detoxification [20] and the beetle Caryedes brasiliensis feeds exclusively on l-canavanine–containing seeds but catabolizes l-canavanine to l-canaline and urea [21] . However , these two mechanisms to circumvent the toxic properties of l-canavanine are specific to few insect species . Thus , the evidence for a protective function against predation for such nonprotein amino acids , i . e . , whether and how insects are informed that plants contain l-canavanine , remains to be shown [15] . Amino acids are known to be the ligands of G-protein–coupled receptors ( GPCRs ) belonging to the family C [22] , [23] . All members of this family display a common structural architecture characterized by a long N-terminal extracellular domain containing a bilobular LBP [24] , a seven transmembrane domain , and an intracellular C-terminus . This family includes metabotropic glutamate receptors ( mGluRs ) . In mammals and in insects , mGluRs , which are activated by the neurotransmitter glutamate , play different roles in the central nervous system [25] , [26] . We have previously shown that one mGluR has diverged through evolution to give rise to the mX receptor , called DmXR in Drosophila [27] . Orthologs of DmXR are so far only found in insects [27] . DmXR differs from mGluRs in the distal part of the LBP , so that this receptor is an orphan receptor , which is not activated by glutamate [27] . However , we previously showed that the DmXR and mGluR LBPs share the crucial residues necessary to bind a ligand with amino acid structural properties [27] . To deorphanize the DmX receptor , we previously tested various molecules having such properties , including all the classical amino acids , and did not find any ligand [27] . Drosophilidae are saprophytic animals , and members of dipteran families such as Tephritidae or Scatophagidae are seed predators [28] , so we asked whether l-canavanine could activate DmXR . Here , we show that l-canavanine is a ligand of DmXR in vitro . We then wondered whether insects could be informed that plants contain l-canavanine via the mX receptor . We have addressed this question by using Drosophila as an insect model . First , we confirmed that l-canavanine is highly toxic when ingested . We then tested whether Drosophila avoid eating food containing l-canavanine . We found that l-canavanine is recognized by flies and mediates a behavioral avoidance response via a chemosensory mechanism . Hence , l-canavanine is a repellent . We then analyzed the molecular and cellular bases of l-canavanine–induced repulsive behavior , using gustatory behavior , pharmacology , and genetic approaches . We found that l-canavanine is detected in vivo by the DmX receptor . To control the l-canavanine avoidance behavior , the DmX receptor is expressed and required in bitter-sensitive gustatory receptor neurons ( GRNs ) . These findings show that the gustatory detection of a natural toxin relies on DmXR , a divergent mGluR not belonging to the Gr family .
To test whether l-canavanine could activate DmXR , we transiently expressed this receptor in human embryonic kidney ( HEK ) cells and assayed for l-canavanine–induced DmXR activation . We found that l-canavanine activated HEK cells expressing DmXR ( Figure 1A and Figure 1B ) . In contrast , the close structural homolog l-arginine showed no agonist or antagonist effect on DmXR ( Figure 1C ) . l-Canavanine did not activate HEK cells expressing the unique fly mGlu receptor , DmGluA ( Figure 1A ) . We searched for an antagonist and found that N-methyl-l-arginine ( NMA ) inhibited DmXR activation by l-canavanine ( Figure 1C and 1D ) . Previous sequence analysis , mutagenesis , and 3-D modeling studies had shown that the LBP of the DmXR is very homologous to the LBP of mGluRs [27] . The residues contacting the amino acid moiety of glutamate ( the α-COO− and NH3+ groups ) are conserved in DmXR ( e . g . , Thr-176 ) , whereas the residues interacting with the γ-carboxylic group are not [27] . The Thr-176 residue is conserved in all mGluRs , and its mutation strongly decreases the affinity of these receptors for glutamate [24] . Similarly , l-canavanine did not activate Thr-176–mutated DmX receptor ( DmXRT176A ) transfected in HEK cells ( Figure 1A ) , although this mutated receptor is actually localized at the plasma membrane ( as shown in [27] ) . This indicates that the plant amino acid binds into the LBP . Altogether our data show that DmXR is a l-canavanine receptor because of a partial modification of its LBP from the LBP of mGluRs . These results suggest that Drosophila may be able to detect l-canavanine in vivo through the DmX receptor . Since l-canavanine is described as a natural insecticide [15] , we first examined whether ingested l-canavanine is also toxic for Drosophila melanogaster . We maintained 50 young wild-type ( WT ) flies on Drosophila medium containing 10 mM l-canavanine and compared their viability and their fecundity to flies maintained on medium without l-canavanine ( n = 8 ) . When flies fed on 10 mM l-canavanine , we did not observe massive mortality or dramatic decrease of the lifespan . However , all the offspring of flies constrained to eat 10 mM l-canavanine died during larval stages ( number of offspring in control medium >1 , 000 , number of offspring in 10 mM l-canavanine = 0 ) . These results indicate that Drosophila is a l-canavanine–susceptible insect . Because of its toxicity , we hypothesized that Drosophila may avoid eating l-canavanine if they have the choice . To test this , we performed a two-choice feeding preference test . This behavioral assay measures the consumption of a sucrose solution ( 5 mM ) colored by two food dyes ( blue or red ) offered simultaneously to adult fly populations [29] . After 2 h in the dark , flies are counted on the color dye witnessed in their abdomen . In the control situation , WT flies preferred the blue solution , the preference index ( PI ) being 0 . 82±0 . 04 ( Figure 2 ) . We then added increasing concentrations of l-canavanine to the blue solution ( 1 mM to 40 mM ) . We found that l-canavanine inhibited the intake of the blue solution ( Figure 2 ) , leading to a symmetrical increase in the intake of the red solution ( unpublished data ) . The l-canavanine effect increased with concentration , reaching a plateau at 30–40 mM . At these concentrations , the PI for the blue solution is 0 . 13±0 . 06 ( Figure 2 ) . A similar repulsive effect of l-canavanine was also visible when the drug was added to the red solution ( unpublished data ) . To determine whether this repulsive effect was mediated by a chemosensory mechanism , we used flies carrying an adult-viable mutant allele of the pox-neuro ( poxn ) gene . In homozygous poxn flies , external chemosensilla are deleted or transformed into mechanosensilla [30] , [31] . poxn flies fed equally on the two colored 5 mM sucrose solutions ( PI for blue = 0 . 49±0 . 09 ) ( Figure 2 ) . When 40 mM l-canavanine was added to the blue solution , their feeding behavior did not differ ( PI for blue = 0 . 51±0 . 09 , Figure 2 ) . Thus , these results show that l-canavanine is a repellent molecule and that Drosophila uses a chemosensory mechanism to detect this plant insecticide . To study whether DmXR was involved in l-canavanine detection in vivo , we first used a pharmacological approach . We hindered DmXR function by using the NMA antagonist in the two-choice feeding test . When present in the medium , this drug should diminish the repulsive action of l-canavanine . We first tested whether NMA alone could influence the fly feeding behavior: WT flies were allowed to choose between a blue solution containing 30 mM NMA versus a red control solution , both containing 5 mM sucrose . We found that flies were insensitive to 30 mM NMA ( PI for blue = 0 . 80±0 . 04 and 0 . 83±0 . 04 in control and NMA-fed flies , respectively , Figure 3A ) . However , significantly more flies fed on 20 mM l-canavanine in presence of 30 mM NMA ( PI for blue = 0 . 68±0 . 07 ) than on 20 mM l-canavanine alone ( PI for blue = 0 . 36±0 . 07 ) ( Figure 3A ) . l-Arginine ( 30 mM ) , which is inactive on DmXR , had no effect on l-canavanine-induced repulsive behavior ( PI for blue = 0 . 41±0 . 05 , Figure 3A ) . Thus , blockade of DmXR with the antagonist NMA lowered the repellent effect of l-canavanine . We then used a genetic approach , taking advantage of two fly lines in which the DmXR encoding gene that we called mangetout ( mtt ) is disrupted . The f06268 line carries a piggyBac transposon inserted into the mtt gene as determined by Exelixis sequence analysis [32] , and the Df ( 2R ) Exel7096 line completely removes the mtt locus [33] ( Figure 4A ) . Both mutants are viable in homozygous conditions . We expected that the insertion of a transposon 35 bp downstream from the third exon of mtt would disrupt the transcription of the gene . Indeed , f06268 homozygous flies are mtt loss-of-function mutants since no RNA was detected by quantitative real-time reverse-transcriptase polymerase chain reaction ( QRT-PCR ) in adults ( Figure 4B ) . In homozygous Df ( 2R ) Exel7096 flies , mtt RNA was also not detectable by QRT-PCR ( Figure 4B ) . As a control for genetic background , we used flies homozygous for the f01266 piggyBac transposon line [32] ( Figure 4A ) , which express normal levels of mtt RNA ( Figure 4B ) . In control two-choice feeding tests without l-canavanine , homozygous mttf06268 , hemizygous mttf06268/Df ( 2R ) Exel7096 , and homozygous f01266 flies behaved as WT ( Figure 3B ) . When 30 mM l-canavanine was added to the blue solution , mttf06268/mttf06268 and mttf06268/Df ( 2R ) Exel7096 flies fed on this blue solution ( PI for blue = 0 . 71±0 . 06 and 0 . 73±0 . 09 , respectively ) , whereas WT and f01266 flies were repulsed ( PI for blue = 0 . 16±0 . 06 and 0 . 23±0 . 08 , respectively ) ( Figure 3B ) . Thus , mtt mutant flies are insensitive to 30 mM l-canavanine . Both pharmacological and genetic data lead to the conclusion that DmXR is required for the detection of l-canavanine and that its activation hinders the feeding . Since DmXR may be the l-canavanine–sensitive receptor in chemosensory organs , we wondered whether this receptor was actually localized in these organs . Drosophila , like other insects , base their feeding decisions on the presence or absence of specific volatile and nonvolatile chemicals present in the food . Volatile chemicals are in general detected by olfactory neurons , located mainly on the antenna , whereas nonvolatile chemicals like amino acids are detected by gustatory receptor neurons ( GRNs ) . GRNs are present in taste sensilla localized in the legs , the labial palps ( or labellum ) found on the tip of the proboscis , and within the pharynx ( called internal taste organs ) [34] . As flies walk on their food sources , tarsal gustatory sensilla evaluate their chemical contents . If phagostimulants are present , the fly extends its proboscis , enabling labellum sensilla to have contact with the food . In the labellum , gustatory chemosensilla house two to four GRNs as well as a single mechanosensory neuron [35]–[37] . In each sensillum , the different subsets of specialized taste neurons are activated by specific classes of tastants , allowing Drosophila to detect sugars , bitter compounds , and water [4] , [38] . To investigate whether mtt is expressed in gustatory sensilla , we first performed QRT-PCR experiments on dissected labellum and tarsi/tibiae of WT flies . As shown in Figure S2 , mtt RNA is expressed in both organs bearing gustatory sensilla . To assess whether mtt would be present in GRNs , we compared the expression level of mtt RNA in WT to that found in poxn mutants , where chemosensory neurons are transformed into mechanosensory neurons . A significant decrease of the amount of mtt RNA was detected in the labellum and the tarsi/tibiae of poxn mutant compared to WT ( Figure S2 ) , suggesting that mtt is expressed in some GRNs . To visualize whether mtt is indeed expressed in gustatory chemosensilla , we then performed in situ hybridization experiments on labellum of WT flies . We found that mtt riboprobe hybridized to a single neuron-like cell within some chemosensory sensilla ( Figure 5A–5D ) . These sensilla , which house five neurons , are clearly gustatory sensilla , because mechanosensory sensilla contain a single , mechanosensory neuron [39] . The in situ labeling appeared to be specific for mtt because chemosensory sensilla were not labeled in labellum from Df ( 2R ) Exel7096 homozygous mtt mutant flies ( Figure 5E ) . Altogether , QRT-PCR and in situ hybridization data indicate that mtt is expressed in only one GRN per labellar chemosensillum , consistent with a role of DmXR as a taste receptor . Due to its very low level of expression , we were unable to use the in situ hybridization technique combined with immunocytochemistry to further analyze the mtt expression pattern . Hence , in order to investigate the nature of the GRNs expressing mtt , we took advantage of a GAL4 enhancer trap line , NP4288-GAL4 , inserted 3 . 1 Kb upstream from the transcription start site of mtt ( Figure 4A ) . A similar strategy had been undertaken for many other Grs , also reported to have a low level of expression [40] . The expression patterns of these receptors were analyzed using GAL4 transgenes containing taste receptor promoters [29] , [40]–[42] or enhancer traps such as NP1017 [43] . These studies have shown that Gr66a-GAL4 , Gr5a-GAL4 , and NP1017-GAL4 lines drive specific expression in bitter- , sugar- , and water-sensitive GRNs , respectively [29] , [42] , [43] . When NP4288-GAL4 was crossed with a green fluorescent protein ( GFP ) reporter line , we observed GFP-positive neurons in taste organs . In the labellum , chemosensory sensilla contained one GFP-positive neuron ( Figure 5F–5I ) , in accordance with the in situ hybridization data . We observed around 28 GFP-positive neurons per labial palp , and two in each tarsus of the legs ( Figure 5J and 5K , and Table 1 ) . In addition , four GFP-positive neurons were present in the labral sense organs ( LSO ) and in the ventral cibarial sense organs ( VCSO ) ( unpublished data ) , which are bilaterally symmetrical internal taste organs located in the pharynx [44] . Interestingly , Gr66a-GAL4 , but not Gr5a-GAL4 or NP1017-GAL4 , drives also expression in the LSO and VCSO ( Table 1 ) , suggesting that NP4288 is expressed in bitter-sensitive GRNs . To determine whether NP4288-GAL4 and Gr66a-GAL4 are indeed coexpressed , we analyzed transgenic fly lines expressing UAS-nlsGFP under the control of both NP4288 and Gr66a GAL4 drivers , and then counted and compared the number of GFP-positive neurons to that of flies containing each driver alone . In flies that express either the NP4288-GAL4 or Gr66a-GAL4 driver , an average of 28 . 4 and 26 . 6 neurons were observed per labial palp , respectively ( Figure S3 ) . In flies that express both drivers , an average of 28 . 8 neurons were detected per palp ( Figure S3 ) . Furthermore , we also observed coexpression of both drivers in the LSO and in the foreleg tarsi ( table in Figure S3 ) . This indicates that most , if not all , GRNs that express Gr66a also express NP4288 , suggesting that this transgene reflects a role for DmXR in bitter-sensitive GRNs . Altogether , these results show that mtt is expressed in one GRN per sensilla that likely correspond to the bitter-sensitive GRNs . To investigate whether GRNs are sensitive to l-canavanine , we examined a direct behavioral measure of leg GRN stimulation by the proboscis extension reflex ( PER ) paradigm: when the leg tarsi encounter an attractive sugar solution , the proboscis often extends [42] , [43] , [45] . If a toxic or bitter compound is added to the sugar solution , the PER is inhibited [42] , [43] , [45] . This assay enables the application of the drugs only on the legs , which carry solely taste sensilla . In addition , we took care that the proboscis never touched the drugs when it extended , so that we were sure that there was no ingestion of drugs ( and consequently , no central effect of these drugs ) . Using the classical PER paradigm ( 5-s stimulation by touching the leg tarsi either with a 100 mM sucrose solution or with a 100 mM sucrose+40 mM l-canavanine solution ) , we found that the occurrence of PER was not affected by l-canavanine in WT or mtt mutant flies ( Figure 6 ) . However , after the PER , WT flies usually sustain their proboscis extension to search for food when their legs are maintained in contact with the sugar solution as shown in Video S1 . This sustained phase was strongly affected by l-canavanine , since significantly more flies prematurely retracted their proboscis ( 78±5% of proboscis retraction [PR] ) compared to 25±4% of PR for the sucrose solution , Figure 6 ) . This PR phenotype occurred generally just after the proboscis extension ( Video S1 ) . We then tested whether the l-canavanine–induced PR phenotype requires DmXR and found that this phenotype disappeared in the mtt loss-of-function mutants ( around 25% of PR , Figure 6 ) . Altogether , these data indicate that DmXR is required in leg GRNs for the l-canavanine detection . To determine in which GRNs mtt is required , we established transgenic flies carrying a mtt RNA interference ( RNAi ) construct under the control of UAS sequence [46] . We first expressed mtt RNAi with NP4288-GAL4 in heterozygous mttf06268 flies and tested the effect of l-canavanine by the PER/PR behavioral assay . The occurrence of PER was not affected by l-canavanine in controls and mtt-knockdown flies ( Figure S4A ) . However , RNAi knockdown of mtt suppressed the l-canavanine–induced premature PR phenotype ( Figure 7A ) in a comparable manner to that observed in mtt mutant flies . This indicates that mtt is expressed in NP4288-GAL4–positive cells , which overlap Gr66a-GRNs in taste organs . However , NP4288-GAl4 drives also expression in cells during development and in the adult brain ( unpublished data ) , precluding us from concluding that mtt is required only in GR66a-GRNs for l-canavanine detection . Thus , we next used the Gr66a-GAL4 driver to specifically express the mtt RNAi in Gr66a-GRNs of heterozygous mttf06268 flies . As already observed with NP4288-GAL4–driven knockdown of mtt , the occurrence of PER was unmodified in this genetic condition ( Figure S4A ) . Importantly , Gr66a-GAL4–induced knockdown of mtt significantly reduced the occurrence of l-canavanine–induced premature PR ( Figure 7A ) . This clearly indicates a requirement of mtt in Gr66a-GRNs for l-canavanine sensitivity . To clearly demonstrate that l-canavanine detection required the presence of DmXR only in Gr66a GRNs , we performed rescue experiments by targeting mtt expression in distinct types of GRNs of mtt homozygous mutants by using the GAL4/UAS system [46] . Indeed , several types of GRNs are present in the tarsi , such as sucrose- , bitter- , and water-sensitive GRNs ( Figure S5 ) . Expression of mtt in the different subsets of GRNs did not affect the percentage of PER ( Figure S4 ) . We then analyzed the PR and found that expression of mtt in sugar or water GRNs did not rescue the mutant phenotype ( Figure 7B ) . In contrast , expression of mtt in Gr66a-GRNs rescued l-canavanine sensitivity ( Figure 7B ) . Thus , it is the stimulation of DmXR , in Gr66a-GRNs , which is responsible for l-canavanine–induced PR . To further verify that only Gr66a-GRNs are necessary for l-canavanine sensitivity , we expressed the hid and rpr proapoptotic genes [47] in these neurons , or inhibited their neurotransmitter release with the tetanus toxin transgene [48] . In both cases , the PER was not affected ( Figure S6 ) , but the l-canavanine–induced PR was lost ( Figure S6 ) . Altogether , these results demonstrate that DmXR is expressed and required only in Gr66a-GRNs for l-canavanine detection . The sites of taste reception are localized to the dendrites of GRNs [36] , which are bipolar neurons containing a single dendrite and a single axon . To confirm that DmXR was actually the l-canavanine taste receptor and not a regulatory receptor modulating Gr66a-GRN synaptic transmission , we performed two kinds of experiments . First , we expressed a HA-tagged receptor in Gr66a-GRNs to determine its subcellular localization in the labellum . As shown in Figure 8A , the receptor was highly concentrated at the dendrite and not detected in the GRN axon in accordance with a gustatory function . Second , we tested whether DmXR could modify Gr66a activation by other repellents than l-canavanine . It has been shown that Gr66a-GRNs are required for caffeine-aversive behavior , Gr66a being a gustatory receptor for caffeine [8] . As already published , we could find that caffeine inhibited sucrose-induced PER ( Figure 8B ) . However , when the PER occurred , we noticed that there was a high rate of premature PR . The caffeine-induced PR phenotype occurred generally just after the PER , similar to what was observed with l-canavanine ( Video S2 ) . We then tested the caffeine-induced phenotypes in mtt mutants and did not find changes in caffeine-induced PER inhibition and caffeine-induced PR ( Figure 8B ) . These data clearly demonstrate that DmXR acts as a l-canavanine gustatory receptor in Gr66a-GRNs .
The ability to avoid ingestion of toxic plants compounds is crucial for insect survival . However , before the current study , only two receptors , Gr66a and Gr93a , which are essential for the caffeine response , were associated with a specific bitter tastant [7] , [8] . The nonprotein amino acid l-canavanine is known to be toxic to insects , when ingested ( [18] and this study ) . Here , our results show that l-canavanine is detected as a repulsive molecule . With a pharmacogenetic approach , we have shown that Drosophila uses a taste detection mechanism mediated by the orphan GPCR , DmXR , which is activated by l-canavanine to trigger this avoidance behavior . This process occurs in bitter-sensitive GRNs where this receptor is expressed . By using the two-choice feeding test , the repulsive effect of l-canavanine was clearly demonstrated . Contrary to the known repellents ( caffeine , quinine [8] , [49] ) , l-canavanine does not affect the PER . However , l-canavanine triggers the retraction of the proboscis following its initial extension impairing the food intake . Indeed , after the PER , WT flies usually sustain their proboscis extension to search for food when their legs are maintained in contact with the sugar solution . This sustained phase is strongly affected by l-canavanine , since significantly more flies prematurely retracted their proboscis at this stage . This inhibition of sustained proboscis extension is not specific for l-canavanine , but is also observed , with the same rate , in the presence of caffeine . Hence , caffeine induces a fast response , which is the PER inhibition , and a slow response , which is the PR , whereas l-canavanine only induces the slow response . One open question is to understand the molecular mechanisms responsible for such a difference between caffeine and l-canavanine in the response dynamics , knowing that both drugs act on the same cell type ( Gr66a-GRNs ) . GPCR-induced signal transduction pathways rely on the activation of intracellular heterotrimeric G-proteins [50]–[52] . To date , two G-proteins have been implicated in the taste pathway , and both are required for sugar perception [53] , [54] . However , a direct evidence for a coupling between Grs and G-proteins has not been demonstrated [38] . As we have shown in our cell transfection assay ( [27] and this study ) , DmXR is a genuine GPCR . It is thought that the Gr66a receptor , like other Grs , is a putative GPCR [8] , [38] . However , we do not know to which type of G-protein these two receptors are coupled in vivo . A first explanation about the different dynamics induced by l-canavanine and caffeine could be that that DmXR and Gr66a are coupled to distinct G-proteins or that both receptors are coupled to the same G-protein , but with different efficiencies . A more speculative explanation may be due to the different structural features of the two receptors , taking into account that Grs are structurally related to the olfactory receptors in Drosophila [40] . As was recently shown , Drosophila olfactory receptors may act as ligand-gated channels instead of being coupled to a G-protein [55] , [56] . Thus , Gr66a receptor may also be a ligand-gated channel . Because of the absence of any intermediate , changes in membrane excitability would be more rapid in presence of caffeine compared to l-canavanine . This may explain the difference in the response dynamics between these two drugs . This study shows that DmXR is expressed and required in bitter-sensitive leg GRNs . However , DmXR is also known to be expressed in the adult brain , in agreement with our observations that NP4288-GAL4 is expressed in this tissue ( unpublished data ) . This suggests the existence of an unknown endogenous ligand , different from l-canavanine , triggering DmXR activation in the brain . To exclude any action of l-canavanine in the brain , we took care that flies avoided ingesting the drug solutions by applying them only on legs during the PER analysis . In addition , we used GRN-restricted drivers , allowing us to specifically analyze the peripheral function of DmXR . Finally , the absence of any defects in the caffeine-induced response of mtt mutants flies excludes a role of DmXR in second and higher order neurons involved in the control of the studied gustatory behavior . As we observed mtt expression in the labellum and in internal taste organs ( LSO and VCSO ) , it is very likely that these taste sensilla also play a role in the l-canavanine–induced aversive behavior . In agreement with this , we observed that flies did not drink a l-canavanine/sucrose–containing solution when directly applied on the labellum ( unpublished data ) , confirming the presence of l-canavanine–sensitive GRNs . So , we assume that DmXR is a l-canavanine–tuned gustatory receptor in all these taste organs . Surprisingly , it is not a Gr member that has been selected to detect l-canavanine , despite the very high sequence diversity of this family . Indeed , DmXR belongs to the mGluR GPCR subfamily because of its close sequence relationship [27] . DmXR and mGluR LBP sequences and 3-D model comparisons have shown that DmXR has diverged only in the LBP part interacting with the γ-carboxylic group of glutamate [27] . These modifications have targeted and changed two residues that are conserved in all mGluRs and are crucial for glutamate-induced activation [24] , [27] . Our study shows that these structural changes are correlated with the ligand selectivity of the receptor . Indeed , DmXR has a divergent LBP so that glutamate is no more an agonist but l-canavanine is . Conversely , the Drosophila mGlu ortholog receptor , DmGluA , is not activated by l-canavanine . This suggests that the original conformation of the mGluR LBP was more adapted to diverge and to recognize l-canavanine than the one of the Grs . In addition , to give rise to this new type of gustatory receptor , appropriate GRN expression has also been added during the evolution of the DmXR function since mGluR expression is mainly found in the central nervous system [25] . Thus , our study suggests that other GPCRs , different from Grs , may have evolved in insects to recognize specific tastants . Finally , the insect-borne diseases are largely increasing , partly due to the development of insecticide resistance . Thus , it becomes urgent to identify insect-specific targets for the design of new drugs against insects . Our work illustrates that the pharmacological and functional characterization of the insect-specific GPCRs , which likely control insect-specific physiological processes , is a way to discover new protection or fighting strategies against harmful insects .
l-glutamate , l-canavanine , l-arginine , and γ-N-methyl-l-arginine ( NMA ) were from Sigma . Human embryonic kidney ( HEK 293 ) cells were cultured and transiently transfected by electroporation as previously described [27] . Carrier plasmid DNA ( pRK5 ) ( 14 μg ) , plasmid DNA containing HA-DmXR WT , HA-DmXRT176A mutant ( 4 µg ) , DmGluRA ( 2 µg ) , and plasmid DNA containing Gαqi9 ( 2 µg ) ( to enable the artificial coupling of DmXR and DmGluRA to phospholipase C , [57] ) were used for the transfection of 107 cells . Determination of inositol phosphate ( IP ) accumulation in transfected cells was performed as previously described [27] . Drosophila stocks were raised on standard fly food medium at 25°C on a 12-h light/dark cycle . WT Canton S flies were used as control flies in all behavioral assays . For experiments using pox-neuro ( poxn ) adult mutant flies , homozygous flies carrying the poxn70−23 mutant allele were used [30] . mttf06268 and f01266 lines carry PBac ( WH ) transposon and are described in [32] . The Df ( 2R ) Exel7096 line carries a small deficiency that completely removes the mtt locus and some adjacent genes ( CG8697 to CG2397 ) [33] . The p ( UAS-mtt ) transgene construct was generated by cloning the hemagglutinin N-terminally tagged full coding sequence of DmXR ( HA-mtt ) [27] into the pUAST transformation vector and injected into w1118 embryos . Several insertions lines were obtained . After QRT-PCR analysis ( unpublished data ) , the w1118;UAS-mttC5 line was chosen for this study . The mtt RNAi line was obtained after amplifying DmXR cDNA sequence with the sense primer 5′-ACT ACT TCT AGA GGC GAT GTG GCA ACA G-3′ and the antisense primer 5′-CCG GGC TCT AGA ATA AGT TTG TTT GCA G-3′ . This sequence was digested with the XbaI restriction enzyme and subcloned into AvrII-digested pWIZ . This new construct was then digested with the NheI restriction enzyme and ligated with the same XbaI-digested PCR product . A clone with the second insert oriented opposite to the first was then selected and used for injection of w1118 embryos . Several insertion lines were obtained . After QRT-PCR analysis ( unpublished data ) , the w1118;UAS-mtt RNAi1 line was chosen for this study . The Gr66a-GAL4 ( II chromosome ) and Gr5a-GAL4 ( II chromosome ) promoter GAL4 lines are generous gifts from H . Amrein ( Duke University , United States ) . The NP1017-GAL4 ( X chromosome ) enhancer trap line was kindly provided by T . Tanimura ( Kyushu University , Japan ) . The NP4288 enhancer trap was obtained from the GETDB Stock Center ( Kyoto , Japan ) [58] . The UAS-hid:UAS-rpr line was a gift from J . R . Martin ( Paris Sud University , France ) . The UAS-TeT line was kindly provided by C . J . O'Kane ( Cambridge University , England ) . The w;UAS-mCD8-GFP and the w;UAS-nlsGFP were obtained from the Bloomington Stock Center . Total RNAs were extracted from whole adult flies ( for the analysis of mtt mutants ) or dissected labella , tarsi , and tibiae ( for the analysis of mtt expression ) by using Trizol ( Sigma ) . cDNAs were generated from 1 µg of total RNAs treated with DNase I ( Ambion ) by using random decamers ( Ambion ) and Moloney murine leukemia virus reverse transcriptase ( Invitrogen ) . Real-time PCR was done using Applied Biosystems SYBR Green PCR mix according to the manufacturer's instructions . PCR was done as follows: 10 min at 95°C followed by 40 cycles: 15 s at 95°C , 60 s at 60°C . Housekeeping genes used to normalize DmXR expression levels were RpL13 , Tbp , and Pgk . Sequences of the primers are RpL13 5′-AGGAGGCGCAAGAACAAATC and 5′-CTTGCTGCGGTACTCCTTGAG , Tbp 5′-CGTCGCTCCGCCAATTC and 5′-TTCTTCGCCTGCACTTCCA , Pgk 5′-TCCTGAAGGTCCTCAACAACATG and 5′-TCCACCAGTTTCTCGACGATCT , and DmXR 5′-CGAATGCAACTGGTTCCTTCTC and 5′-TGAGGAAGTACTCCTCGAAC . Labella were dissected from flies and collected in 4% paraformaldehyde in PBS with 0 . 05% Triton X-100 on ice . After fixation overnight at 4°C , samples were washed 6×10 min in PTX ( PBS , 2% Triton X-100 ) at room temperature . Prehybridization was then done for 2 h at 55°C in hybridization buffer ( HB ) ( 50% formamide , 5× SSC , 0 . 5 mg/ml yeast tRNA , 0 . 1 mg/ml Salmon Sperm DNA , 0 . 05 mg/ml heparin , 0 . 3% Triton X-100 ) . Hybridization was performed overnight at 55°C with digoxigenin-labeled antisense mtt riboprobe derived from mtt cDNA and prepared according to the manufacturer's instructions ( Roche ) . Washes were performed at 58°C in HB followed by washes in HB/PTX mix ( 3/1 , 1/1 , and 1/3 , respectively ) . After blocking in 0 . 5% Blocking Reagent ( Roche ) in PTX , samples were then incubated with anti-Dig-AP ( Roche ) overnight at 4°C . Samples were then washed 6×10 min in PTX . NBT/BCIP mix ( Roche ) was used to visualize the digoxigenin-labeled probe . Samples were mounted in 90% glycerol . To visualize HA-DmXR protein expression , we dissected the labella from Gr66a-GAL4;UAS-HAmtt/+flies from adult head , fixed them overnight in 4% paraformaldehyde in 1× phosphate-buffered saline ( PBS ) , 0 . 3% Triton X-100 . Immunostaining was performed in 1× PBS 3% Triton X-100 and 0 . 5% Blocking Reagent ( Roche ) . The following antibodies were used: monoclonal rat anti-HA ( Roche; 1∶200 ) and Cy3-conjugated donkey anti-rat ( Jackson ImmunoResearch; 1∶500 ) . Samples were mounted in Vectashield . Images were acquired using a Leica microscope and CoolSNAP camera . | Plants evolve to fend off the insects that attack them , often by synthesizing compounds toxic to insects . In turn , insects develop strategies to avoid these plants or resist their toxins . Some plant toxins are nonprotein amino acids . For example , seeds from numerous legumes contain high amounts of l-canavanine , a nonprotein amino acid that is structurally related to l-arginine and is highly toxic to most insects . How insects can detect l-canavanine remains to be elucidated . Using pharmacology , genetics , and behavioral approaches , we show that flies sense l-canavanine using the receptor DmX , an orphan G-protein–coupled receptor that has diverged in its ligand binding pocket from metabotropic glutamate receptors . Disruption of the DmXR gene , called mangetout ( mtt ) , suppresses the l-canavanine repellent effect . DmXR is expressed and required in aversive gustatory receptor neurons , where it triggers the premature retraction of the proboscis , thus leading to the end of food searching . Our results indicate a mechanism by which some insects may detect and avoid a plant toxin . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/behavioral",
"neuroscience",
"neuroscience/sensory",
"systems",
"cell",
"biology/neuronal",
"signaling",
"mechanisms",
"genetics",
"and",
"genomics/gene",
"function",
"pharmacology"
] | 2009 | Plant Insecticide L-Canavanine Repels Drosophila via the Insect Orphan GPCR DmX |
Motor tics are a cardinal feature of Tourette syndrome and are traditionally associated with an excess of striatal dopamine in the basal ganglia . Recent evidence increasingly supports a more articulated view where cerebellum and cortex , working closely in concert with basal ganglia , are also involved in tic production . Building on such evidence , this article proposes a computational model of the basal ganglia-cerebellar-thalamo-cortical system to study how motor tics are generated in Tourette syndrome . In particular , the model: ( i ) reproduces the main results of recent experiments about the involvement of the basal ganglia-cerebellar-thalamo-cortical system in tic generation; ( ii ) suggests an explanation of the system-level mechanisms underlying motor tic production: in this respect , the model predicts that the interplay between dopaminergic signal and cortical activity contributes to triggering the tic event and that the recently discovered basal ganglia-cerebellar anatomical pathway may support the involvement of the cerebellum in tic production; ( iii ) furnishes predictions on the amount of tics generated when striatal dopamine increases and when the cortex is externally stimulated . These predictions could be important in identifying new brain target areas for future therapies . Finally , the model represents the first computational attempt to study the role of the recently discovered basal ganglia-cerebellar anatomical links . Studying this non-cortex-mediated basal ganglia-cerebellar interaction could radically change our perspective about how these areas interact with each other and with the cortex . Overall , the model also shows the utility of casting Tourette syndrome within a system-level perspective rather than viewing it as related to the dysfunction of a single brain area .
Tourette syndrome ( TS ) is a neuropsychiatric disorder characterized by the presence of sudden and repetitive involuntary movements or vocalizations , generally termed as “tics” , having differing degrees of intensity and frequency , and unpredictable duration [1 , 2] . Tics can be simple , for example involving eye blinking , facial grimacing , shoulder shrugging , sniffing , or complex , involving more elaborated manifestations like touching objects , clapping , obscene gestures , or repetition of words [3 , 4] . The typical age of onset of TS is around five to seven years and the course of the disease can be quite variable . In addition to tics , children with TS can show a variety of comorbid psychopathologies , including learning difficulties , sleep abnormalities , anxiety , obsessive-compulsive disorder ( OCD ) , and attention deficit hyperactivity disorder ( ADHD ) [5 , 6] ( see S1 Table in the Supporting Information for all the main abbreviations used in the article ) . Usually , most TS symptoms decline during adolescence or early adulthood [7] . Motor tics are a cardinal symptom of TS shared with several neurological impairments including dystonia [8] , Huntington’s disease [9 , 10] and OCD [11 , 12] . Traditionally , tics in TS are associated with basal ganglia abnormalities and in particular with a dysfunction of the striatal GABAergic networks leading to an excess of striatal dopamine [13–16] . This excess might cause an abnormal functioning of the basal ganglia-thalamo-cortical circuit leading to the production of tics [17] . To understand how this circuit may operate in TS , we first briefly describe how it typically works in healthy subjects ( see section “The basal ganglia and their loops with the thalamo-cortical system: anatomy and physiology” for more details ) . In general , the basal ganglia promote movement generation of some specific motor patterns within primary motor cortex via a double-inhibition mechanism while maintaining tonic inhibitory control over other patterns [18–21] . In non-pathological conditions , the inhibition of specific GABAergic output nuclei of the basal ganglia leads to release the activity within the target thalamus areas forming loops with primary motor cortex , thus allowing the focused disinhibition of specific motor patterns . The basal-ganglia double-inhibition mechanism also targets sub-cortical areas , although in this case without the mediation of the thalamus , for example the superior colliculus for eye movements [22 , 23] . An alteration in striatal dopamine release as in TS may induce the production of tics as a consequence of a focal excitatory abnormality in the striatum that causes an undesired disinhibition of thalamo-cortical circuits [15 , 17] whose effect is the production of tics . The basal ganglia are strongly linked , both anatomically and functionally , with several cortical regions and with the cerebellum . The basal ganglia and cerebellum receive input from , and send output to , cortex through multisynaptic anatomically partially segregated loops performing distinct functional operations within the motor and cognitive realms [24–27] . Studies in rats [28] and monkeys [29] have demonstrated that the cerebellum has a strong disynaptic projection to the striatum mediated by the intralaminar nuclei of the thalamus . Complementary to this , recent investigations on monkeys have shown that the subthalamic nucleus , an important component of the basal ganglia , has a disynaptic projection to the cerebellar cortex by way of the pontine nuclei [30] . Similar data have been found in humans [31] . These data have stimulated new research to investigate the role of the cerebellum and basal ganglia in functions typically associated with cortex ( e . g . , action understanding , [32–35] ) , and the involvement of cortical and cerebellar regions in impairments typically associated with basal ganglia such as Parkinson’s disease [36–46] and TS [47–49] . This system-level perspective [50 , 51] , according to which the basal ganglia work in concert with cortex and cerebellum to produce motor and cognitive behaviours of various complexity [26 , 35 , 52–55] , renders the whole picture of TS pathophysiology more complex [56] . In particular , the specific contribution of cerebellar and cortical areas to basal ganglia-mediated tic expression remains unknown . The cerebellar activation found in several studies on tics may reflect an increase of afferent sensory input driven by overt tic movements or , rather , may be due to the transmission of descending signals originating from primary motor cortex [57] . Another possibility is that cerebellar neurons fire before tic movements and their discharge takes place no later than that of primary motor cortex neurons [49] . Recently , McCairn and colleagues [49] have explicitly adopted a system-level approach to investigate the role of basal ganglia , cortical , and cerebellar areas in TS . The authors generated a pharmacologic motor tic/TS model with two monkeys by microinjecting the GABA antagonist bicuculline into the sensorimotor striatum ( putamen ) [57 , 58] . In this way , the increased striatal inhibition caused abnormalities in the dopamine release [3 , 59 , 60] that , in turn , led to motor tics [13–16] ( see section “Simulation settings” for more details ) . Neural activity was recorded from several areas of the basal ganglia , cerebellum , and primary motor cortex simultaneously to investigate their relationship . The results confirmed that aberrant activity leading to motor tics was initiated in the basal ganglia . However , they also showed how the occurrence of tics was closely associated with enhanced activity involving both the motor cortex and the cerebellum , implying that these may act in concert to produce overt tic movements . The time latencies of pathological activity in the cerebellum and primary motor cortex substantially overlapped and followed that of basal ganglia . This suggests that aberrant signals may travel along divergent pathways from the basal ganglia to the cortex and cerebellum . In this respect , the authors suggest that the basal ganglia might , presumably , influence cerebellar activity via the subthalamic-pons-cerebellar disynaptic link [30] , with a latency that is sufficiently short to allow cerebellum to affect abnormal movements . However , the authors did not support this claim empirically . Building on the results obtained in [49] , in this paper we propose a computational model reproducing key anatomical and functional features of the system formed by the basal ganglia , thalamus , primary motor cortex , and cerebellum to investigate within a system-level perspective how motor tics are generated in TS . The model yields several results and predictions . First , it reproduces the main results obtained in [49] about the differences in basal ganglia/primary motor cortex/cerebellum neural activity recorded during tic/no-tic events . Second , and remarkably , the model shows that in order to reproduce and explain these data it is important to study the interplay between striatal dopamine signals and cortical activity , and the role played by the recently discovered subthalamic-pons-cerebellar pathway [30] working in synergy with the cerebello-thalamo-cortical circuit . In particular , the model predicts that the interplay between dopaminergic signals and cortical activity may underlie the emergence of tic events , and that the anatomical connection linking subthalamic nucleus and cerebellum may support the involvement of the cerebellum in tic production . In this way , the model supports the claim of [49] about a possible involvement of the subthalamic-pons-cerebellar circuit in tic generation , while specifying what functions it might accomplish . These predictions could form the basis for future experiments . Third , the model predicts that tic production could be reduced by externally stimulating or inhibiting the primary motor cortex . These predictions could be important for identifying new target areas , aside the traditional ones [6 , 61 , 62] , to design innovative system-level therapeutic actions . Finally , the model investigates the role of the recently discovered disynaptic bi-directional connections linking the basal ganglia with the cerebellum [29 , 30] . To the best of our knowledge , there are no computational models investigating the role of these connections . Previous computational and conceptual models have , indeed , mainly studied the indirect interactions between basal ganglia and cerebellum mediated by cortical areas [63–69] . In view of recent empirical studies , attention to non-cortical-mediated basal ganglia-cerebellum interaction could radically change our perspective about how these subcortical areas interact with each other and with the cortex to regulate motor and non-motor behaviours [31 , 35 , 43 , 55] . The computational model proposed here starts to address this issue by developing a simplified computational implementation of such links and by suggesting the possible involvement of the subthalamic-pons-cerebellar circuit in motor tic production . Fig 1 summarizes the brain areas mainly involved in tic production . The rest of the paper is organized as follows . Section “Methods” describes the computational features of the model and the biological support of its assumptions . Section “Results” illustrates the results of the target empirical experiments with monkeys performed in [49] and how the model reproduces and explains them . It also presents the predictions of the model . Section “Discussion” discusses the system-level mechanisms through which the model explains the motor tic production and presents some limitations of the model while also suggesting possible future work to overcome them .
The system-level architecture of the model is formed by four main components ( see Fig 2 ) : the basal ganglia component ( BG ) reproduces the key anatomical and functional features of the basal ganglia building on the computational models proposed in [21 , 70–72]; the cerebellum component ( Cer ) captures some critical anatomical and functional aspects of the cerebellum pivoting on the models proposed in [68 , 73 , 74]; the motor thalamus and the primary motor cortex components ( respectively Th and M1 ) , which do not focus on anatomical features , only reproduce functional aspects related to the activity of distinct neural populations . Indeed , as it was non-trivial to reproduce the dynamics of the complex system formed by the basal ganglia-thalamo-cortical loops , the loops linking the cerebellum with the cortex through thalamus , and the circuits linking the basal ganglia with the cerebellum , we used simplified models of the primary motor cortex and thalamus that allowed an easier study of the structures considered important for the generation of tics . This follows a strategy previously proposed for building system-level models more amenable to analysis [75] ( cf . also [76 , 77] ) . At the same time , due to the key role of the basal ganglia in triggering motor tics in TS [49] we considered a more sophisticated model of these nuclei with respect to the other components of the model . The possible effects of introducing finer grained anatomical and physiological details in the model are discussed in section “Conclusions and future work” . With the exception of Cer components , each of the other model components is formed by three neural units representing three distinct neural populations encoding different information contents . From a behavioural point of view , it would have been sufficient to include just one neural unit for each component to address the target experiment of McCairn and colleagues [49] . Indeed , this experiment involved monkeys not solving any specific task but rather producing motor tics as spontaneous input-free behaviors under neural noise ( as detailed below , in the model such noise is intended to capture the spurious effects on neural activation due to the signals supplied by other cortices as well as the effect of intrinsic neural noise [77–80] ) . However , it was important to include a larger number of neural units to reproduce in a realistic way the circuitry implementing the competitive dynamics typical of some components of the model , in particular of the BG [21 , 81] , relevant to the production of tics ( see sections “The basal ganglia and their loops with the thalamo-cortical system: anatomy and physiology” and “The model predicts that the interplay between dopaminergic signal and cortical activity triggers the tic event” ) . The neural units within each component of the model are represented by leaky integrator units [82 , 83] . The activation of a single leaky unit represents the average firing rate of a population of real neurons . The neural population approach based on leaky integrator units is suitable for representing system-level features that are not immediately apparent at the level of individual neurons but manifest at higher levels [77] . This approach facilitates the comparison between the data on neural activation recorded in the model and the data obtained in the target experiment proposed in [49] . In addition , it allows a dimensionality reduction that increases the computational efficiency of simulations [84] , and this is important for running sensitivity analyses of large models such as the one performed here . The chosen granularity of the model was also suitable for this work since it did not aim to reproduce detailed neural spatio-temporal patterns supporting the selection and performance of specific movements ( cf . section “Simulation settings” ) . The model has been implemented , as described here , based on a technique that was proposed in [85 , 86] ( see also [87] ) . This technique , suitable to illustrate neural system-level models formed by homogeneous neurons , aims to standardise all equations of the model so as to simplify its explanation , understanding , implementation , analysis , and reproducibility . The model is in particular fully described by the few equations presented in this section , the values of the equation parameters reported in the S2 Table ( see Supporting Information ) , and the diagram of Fig 2 showing the architecture and connectivity of the model . Each leaky integrator unit of the model components has an activation a and an activation potential ( hereafter “potential” ) u at time t having the following dynamics [82 , 83]: τ u ˙ = - u + I ( 1 ) a = f ( u ) ( 2 ) where τ is the unit decay coefficient; I is the input to the unit that , depending on the component to which the unit belongs , could take into account the effects of the different pre-synaptic connections received from other components , the effects of noise , and the effects of dopamine . In particular , the term I of the post-synaptic unit j of the component post is computed as follows ( the effects of dopamine are discussed below ) : I p o s t j = r p o s t j + ∑ p r e ∑ i w p r e i → p o s t j · a p r e i + n ( 3 ) where rpostj is the resting potential of the post-synaptic unit j of the component post; wprei → postj is the weight of the connection from the pre-synaptic unit i of the component pre to the post-synaptic unit j of the component post; aprei is the activity of the pre-synaptic unit i of the component pre computed according to Eq 2 , and n is a noise value independently sampled from a Gaussian distribution for each unit . The pre-synaptic and post-synaptic units are those respectively sending and receiving signals as indicated in Fig 2 . The function f ( . ) = [tanh ( . ) − thr]+ is the activation function of neural units , where tanh ( . ) is the hyperbolic tangent function , whose values were remapped to the range [−400 , 400] , thr is a parameter used to reproduce the effects of the threshold potential of real neurons [88] , and [ . ]+ is a function returning the value of the function argument if this is positive , and zero otherwise . The differential equations related to the u of all units are numerically integrated using the Euler method . Before presenting the computational details of the model components , this section highlights some features of the anatomy and physiology of the basal ganglia , and their loops with the thalamo-cortical system , as they are particularly important for tic production . The description uses the same abbreviations adopted for the model components shown in Fig 2 . In the model , the BG component includes five regions , each formed by a layer of three leaky integrator units . The two main inputs of the BG component are Str and STN . Str is formed by two subregions , StrD1 and StrD2 , with units expressing D1R and D2R dopamine receptors . STN works in a loop with GPe and receives most of its afferent projections from M1 . Similarly , StrD1 and StrD2 receive afferent projections from M1 and Th . StrD1 , StrD2 , STN and GPe send efferent projections to the GPi or SNr , which are the GABAergic output nuclei of the BG ( hereafter , GPi and SNr , represented as one component in the model , will be indicated as GPi/SNr ) . The excitatory and inhibitory connections between the regions of the BG component are feedforward links between one unit and the topologically corresponding unit in the following layer ( thin lines in Fig 2 ) . This connectivity reproduces in an abstract fashion the structure of the BG channels ( one-to-one connections ) . The units of STN are connected with all GPi and GPe units ( all-to-all connections ) . This simulates the diffused action of the STN over its target regions [24 , 70] . BG project to the Th through inhibitory links ( GPi/SNr-Th ) [21] and to Cer through excitatory connections ( STN-Cer ) [30 , 35] . For the striatal sub-component StrD1 , the I term is calculated by multiplying the right side of Eq 3 for the dopaminergic term aDAD1 used to account for the dopaminergic modulation on the activity of StrD1 and computed as follows: a D A D 1 = b S t r D 1 + d S t r D 1 · a D A ( 4 ) where bStrD1 is a baseline StrD1 potential modulation not due to DA , dStrD1 is the StrD1 DA factor amplitude , and aDA is the activity of a leaky integrator unit ( Eq 2 ) used to simulate the dopamine efflux . The dopamine efflux was simulated through an activation potential uDA of the DA leaky unit that rapidly reaches a maximum level DAMAX = 0 . 5 around 1 sec from the beginning of each trial , and then decays toward DAMIN = 0 . 01 . Similarly , for the striatal sub-component StrD2 the I term is calculated by multiplying the right side of Eq 3 by the dopaminergic term aDAD2 used to account for the dopaminergic modulation on the activity of StrD2 and computed as follows: a D A D 2 = a D A D 1 b S t r D 2 + d S t r D 2 · a D A ( 5 ) where bStrD2 is a baseline StrD2 potential modulation not due to DA and dStrD2 is the StrD2 DA factor amplitude . While the contribution of the dopaminergic efflux on the activity of StrD1 units was implemented as a multiplicative excitatory effect ( Eq 4 ) , the modulation of dopaminergic efflux on the activity of StrD2 units was implemented as a multiplicative inhibitory effect ( Eq 5 ) . It has been shown that these two different types of dopaminergic modulations reflected what happens in the real BG ( cf . [103] ) . Hence the term aDAD1 in the Eq 5 takes into account the recent data showing a possible combined effect of D1 and D2 receptors [89] . For the other sub-components of BG ( STN , GPe and GPi ) the I term was computed by simply using the Eq 3 . The Th component is formed by two regions: ThBC , representing the thalamic parts where both BG and Cer project; ThC , representing the thalamic areas where only Cer projects . Each region includes three leaky integrator units . This organization in two subregions is based on anatomical data showing the presence of both partially segregated and overlapping projections from the BG and Cer output regions to Th [104 , 105] . ThBC receives inhibitory signals from the BG component ( GPi/SNr region ) and excitatory signals from the Cer component [106 , 107] . By contrast , ThC only receives excitatory signals from the Cer component [26 , 105] . In addition , ThBC and ThC send excitatory signals to the input stages of the BG component ( StrD1 , StrD2 , STN ) [29 , 55 , 108 , 109] and are bi-directionally connected with M1 through excitatory links [26 , 27 , 53] . The I terms of ThBC and ThC were computed using Eq 3 . The Cer component was built starting from the Marr-Albus type of model [110 , 111] proposed in [68 , 73 , 74] , as these are implemented with a level of abstraction that was similar to the one of the BG component . In particular , the Cer includes four regions , each formed by a layer of leaky integrator units: the granule cells ( GC ) formed by 100 units; the Golgi cells ( GO ) formed by one inhibitory unit; the Purkinje cells ( PC ) formed by three units; the dentate nuclei ( DN ) formed by three units . These numbers approximate the proportion of neurons observed in the real Cer [110–112] . There is also a mossy fibers ( MF ) layer which receives excitatory connections from M1 and STN . These circuits reproduce the functional effects of the M1 and STN activities on the cerebellar areas due to the pons-cerebellar link [26 , 30] . GC transform the signal from MF for further processing by the PC . According to the Marr-Albus theory , GC provide a sparse code , that is , a code with only a small fraction ( less than 10% in the model used here ) of cells active at any time . In this way , the functioning of the cerebellum is facilitated because different MF inputs create highly dissimilar sparse GC activity patterns , which are easily recognizable by PC . GO receives excitatory input from MF and GC , and provides a feedback inhibition to GC . GO firing suppresses MF excitation of GC and thus tends to shorten the duration of bursts in the connections linking GC to PC . This mechanism further supports the sparse coding of the input [73] . PC show a spontaneous activity [112] that is influenced by parallel fibers—these are excitatory afferent inputs from GC . PC also receive an input signal from M1 through the inferior olive-climbing fiber system—a climbing fiber is an axon of a neuron of the inferior olive . This circuit is important for implementing Cer learning processes [113] . In this respect , the inferior olive is commonly thought to compute an error signal conveyed to PC through nucleo-olivary projections ( refer to [114] for a detailed computational model ) . In particular , in a model that would take into account the Cer learning processes , the output of DN should be subtracted from the M1 input to PC . The inferior olive-climbing fiber system is also relevant to managing the timing of the input [115] . Since the model did not aim to study the effects of Cer learning processes on tics , we abstracted the timing effect of such a system with a simple connection from M1 to PC ( see Fig 2 ) . This link contributes to modulate the PC activity in a synchronous way with respect to the M1 activity . The activity of the units of DN is modulated by the inhibitory connections from the corresponding units of PC ( one-to-one connections ) and by the excitatory collaterals from MF supplying a baseline activation for DN [116] . The three units of DN , in turn , send excitatory signals to M1 ( through Th ) [26] and to StrD1 and StrD2 ( through ThC ) [29 , 105] . The basic functioning of the Cer component is organized around the inhibitory PC , whose axons provide the only output of the cerebellar cortex . Each unit of PC modulates the selection of a particular motor pattern within the dentate-thalamo-cortical system [117] . In other words , similarly to what happens to BG , parallel sub-loops with the Cer component independently modulate a motor pattern allowing the selective facilitation of one response and the concurrent suppression of the others [26 , 35 , 45] . When MF are silent ( i . e . , no input is received by Cer ) , PC show spontaneous activity and their inhibitory output prevents DN cells from firing . This in turn prevents the selection of responses at such times . We assumed that a previous learning process based on long-term depression ( LTD ) and long-term potentiation ( LTP ) [68 , 118] has led to having the GC-PC connections assume a high negative value when a motor pattern has to be selected by the input , and a small negative or positive value when a motor pattern should be inhibited . The high negative value for the GC-PC synapse assures that the activity of the corresponding PC unit is close to zero and this in turn makes the corresponding DN unit positively activated [111] . Consequently , excitation from MF collaterals predominates over inhibition from PC to DN related to the correct response . DN neurons excite the thalamus that , in turn , excites the region in the motor cortex related to the correct response . The I terms for the units of the Cer were computed using Eq 3 . The noise term n was set to zero for GC , GO , PC and DN . We set by hand the value of the elements of wGC → PC by assuming that a previous learning process had led activity from GC to PC having a zero value when a motor pattern has to be selected by the input , and a positive value when no motor pattern has to be selected . The values of the parameters of the equations are shown in the S2 Table ( see Supporting Information ) . The activation recorded in the GC and PC layers is assumed to correspond to the firing rates measured within the cerebellar cortex of the monkeys ( labeled as “CbllCx” in [49] ) . The M1 component is formed by three leaky units whose activity is assumed to correspond to the firing rate recorded in the primary motor cortex of the monkeys in the target experiment of McCairn and colleagues [49] . M1 is bi-directionally connected with Th and projects to BG and Cer through excitatory links [26] . The I term of the units of the M1 component was computed using Eq 3 .
This section compares the data on neural activity collected in [49] in the brain of one monkey and the data on neural activity collected in the brain of one subject simulated with the model ( data for other subjects are qualitatively similar ) . Figs 3 and 4 show respectively the firing rate in the BG and in the M1 and Cer during TIC and NO-TIC trials ( i . e . , intertic intervals ) recorded in the monkey and in the model . The model curves are obtained with ( a ) an activation of cortex affected by the intrinsic neural noise of the various regions of the model; ( b ) a further activation mimicking possible inputs to M1 from other cortical areas ( here captured , in the case of no-tic and tic cases , with a Gaussian-like input with a height of respectively 30 and 17 , and a standard deviation of respectively 0 . 040 and 0 . 250 sec ) ; ( c ) dopaminergic bursts that capture the possible dopamine dysregulation caused by bicuculline ( here captured , in the case of no-tic and tic cases , with a Gaussian-like input with a height of respectively 1 and 50 and a standard deviation of respectively 0 . 600 and 0 . 020 sec ) . The figures show that real and simulated data are very similar . In both cases , in the Dorsal putamen and GPi there are no relevant differences in the firing rate amplitudes between the tic and no-tic state whereas there is a partial preservation of the response for GPe , with the early inhibitory peak maintained and the later excitatory peak increased during a tic . By contrast , for M1 and Cer ( CbllCx in the figure ) the firing rate amplitudes during the tic state are greater than those measured during the no-tic state . The model allows the simulation of the activity of other key areas not monitored in the target experiment [49] . In particular , we measured the activity in STN and Th based on the hypothesis that these regions might be involved in tic production due to their potential role as mediators between M1 , BG , and Cer signals [26 , 55 , 105] . Fig 5 shows that , similarly to what happens for M1 and CbllCx , in the STN and Th there is a remarkable difference in the activity amplitudes between tic and no-tic states . This result represents a prediction of the model that could be tested in new experiments . The abnormal activation of M1 in case of a tic supports the increase of activity in STN and Th . The enhanced activity of STN , in turn , contributes to get a larger excitatory peak in GPe ( cf . section “Propagation of aberrant basal ganglia activity to primary motor cortex and cerebellum” ) . In section “Discussion” , we further discuss the possible neural processes based on which STN and Th may be involved in tic production . The results obtained with the model and presented in sections “The model reproduces data on firing rate during tic/intertic intervals” and “The model predicts an abnormal tic-related activity in the subthalamic nucleus and in the thalamus” are supported by statistical analysis of the data collected across 40 simulated subjects . For each subject , we considered one trial randomly selected from the 10 trials . In this way , we got 40 different measures across all the simulated subjects . A two-way analysis of variance ( ANOVA ) was performed using the function aov of the statistical analysis software R . In more detail , the ANOVA was performed with two factors , namely the peak activity in the different areas ( i . e . , Dorsal putamen , GPi , GPe , STN , Th , M1 , CbllCx ) and the movement state ( i . e . , NO-TIC vs . TIC ) . A post hoc test was also applied using the function TukeyHSD of R . A result was considered statistically significant if the p value was less than 0 . 001 . The average value of the peak amplitude of the activity and its standard deviation for the areas of the model in TIC and NO-TIC trials are reported in the S3 Table , visually summarised in Fig 6 . The ANOVA shows a statistically significant interaction between the activity in the different areas and the movement state ( p < 0 . 001 ) . In addition , the post hoc tests show that , as in the experiment of McCairn and colleagues [49] , the differences in the activity amplitudes between TIC and NO-TIC trials are not statistical significant for the Dorsal putamen ( p = 0 . 990 ) and GPi ( p = 0 . 970 ) , whereas they are statistically significant for all other regions , in particular GPe , STN , Th , M1 , and CbllCx ( p < 0 . 001 for all of them ) . Fig 7 shows the firing rate of the Dorsal putamen and M1 cells presented in [49] and obtained by recording neuron activity in the monkey model of tics . The authors found that the striatal burst occurs 0 . 29 sec before the tic initiation . This is followed by the activation of GPe and GPi , occurring respectively 0 . 26 sec and 0 . 19 sec before the tic onset , and by the activation of Cer and M1 respectively happening 0 . 11 sec and 0 . 12 sec before the tic onset . The authors also found significant differences in the latency distribution of BG areas versus M1 and CbllCx , whereas they did not find significant differences in this distribution between M1 and CbllCx . Overall , these findings suggest that in the animal model of [49] the tic event is triggered by the putamen as the activation of BG precedes that of M1 and Cer . We obtained similar results in the model . In more detail , to study the causality of the signal propagation in the model we computed the delay of the onset of the average activity in M1 with respect to the onset of the average activities in the other areas . The delay was calculated by using the cross-correlation function ccf of the statistical analysis software R applied to the derivative of the signals . The results of the cross-correlations are summarized in Fig 8 . The figure shows that in the tic state the onset of the average activity in the Dorsal putamen takes place 0 . 126 sec before the onset of the same signal in M1 . Similarly , the onset of the average activity in GPe and GPi anticipates the onset of the same signal in M1 of respectively 0 . 116 sec and 0 . 124 sec . By contrast , the delays between the onset of the average activity in STN , Th , CbllCx and in M1 are small . We performed a statistical analysis over the data collected in 40 simulated subjects ( the data for the analysis were collected as described in section “Statistical analysis” ) . The one-way ANOVA ( having as a factor the means of the delays for each area ) shows that there are significant differences in the means of the delays resulting from the cross-correlations between M1 and the other areas ( p < 0 . 001 ) . The post hoc tests show that there are statistically significant differences between the means of the delays related to Dorsal putamen , GPi , and GPe and the mean related to M1 ( p < 0 . 001 for all comparisons ) . By contrast , there are no statistically significant differences between the means of the delays related to STN , Th , and CbllCx and the mean related to M1 ( M1 vs . STN: p = 0 . 988; M1 vs . Th: p = 0 . 991; M1 vs . CbllCx: p = 0 . 156 ) . The clustering in two groups of the delays is apparent from Fig 8 . Overall , these results suggest that in the model the abnormal tic-related activity in M1 is triggered in the Dorsal putamen and propagates through GPe , GPi , STN , Th , and CbllCx . The data shown in the previous section suggest that the tic activity first emerges in the BG , in particular in the Dorsal putamen , and then propagates towards the other regions of BG , and to Th , Cer , and M1 . However , these data do not answer the question: why in some trials is there a tic event while in others there is not ? The model suggests a possible answer to this question . In more detail , the model predicts that in the case of trials where a tic is exhibited there is a conjunction of two events: ( i ) a dopaminergic burst; ( ii ) M1 neurons activation happening at a time close to the dopaminergic burst and due to noise and inputs from other cortical areas . To further investigate this mechanism of tic generation , we ran further simulations where we explicitly simulated different random events possibly affecting dopamine , representing the effects of dopamine dysregulation caused by bicuculline , and M1 , possibly representing inputs from other cortical regions . The interaction of cortical and dopamine events having different intensities are shown in Fig 9 . The graphs of the figure have been obtained with three increasing levels of M1 activation ( simulated with a Gaussian-like input with a height measuring respectively 0 , 17 , and 30 , and a standard deviation measuring respectively 0 . 250 and 0 . 040 sec for the two non-zero height cases ) . Such input was also multiplied by a random number drawn from a uniform distribution ranging in ( 0 , 1 ) before being sent to each of the three channels of M1 , so as to capture differential inputs received by the three channels . For dopamine , we simulated three dopaminergic bursts with increasing intensities ( simulated with a Gaussian-like input with a height measuring respectively 0 , 1 , and 50 , and a standard deviation measuring respectively 0 . 600 and 0 . 020 sec for the two non-zero height cases ) . The figure shows that when the two events occur together and have a sufficient intensity , the activity of the Dorsal putamen triggers the BG selection process ( see section “The basal ganglia component ( BG ) ” ) . In this way , the noisy signal conveyed to one of the three channels is possibly amplified so that it wins the neural competition . The signal of the winner channel is then transmitted to the Cer through the STN-pons-Cer circuit and further modulated through the Cer-Th-M1 circuit , contributing to an abnormal activity in STN , Cer and M1 . The activity within M1 is also amplified through the recurrent excitatory M1-Th loop . Fig 10a shows the effects on the activation of the three M1 channels and tic production caused by the concomitant occurrence of M1 activation and a dopamine burst . By contrast , the model does not exhibit tics if there is no relevant activity of M1 ( Fig 9 ) . Indeed , in this case even if the Dorsal putamen might have an activity due to the striatal noise and the dopamine production , the BG selection process cannot select any signal within the BG-Th-M1 channels as the thalamic activity is substantially zero . Similarly , the model does not exhibit tics if there is a non-zero activity of M1 but there is not a dopaminergic burst . The reason is that the activity of Dorsal putamen depends on the presence of dopamine . Dopamine modulates the signals conveyed by the direct and indirect pathways in different ways: it has a multiplicative effect on StrD1 ( cf . Eq 4 ) and an inhibitory effect on StrD2 ( cf . Eq 5 ) . If the dopamine burst is zero ( or close to zero ) the Dorsal putamen shows a very low activity ( Fig 9 ) and cannot support the selection process . In particular , the signal transmitted by the indirect pathway leads to a strong net inhibitory effect on the signals conveyed by the three BG-Th-M1 channels . This implies that no channel can win the neural competition and so no tics are generated ( Fig 10b ) . As mentioned , Fig 9 has been obtained by directly activating M1 units through an external signal . This simulated process might be thought to mimic a real situation where cortex is activated through an external stimulation . As an example , this stimulation might be performed through transcranial direct-current stimulation ( tDCS ) , which can be used to either enhance or inhibit cortical activity . The model hence suggests the possibility of designing tDCS non-invasive treatments targeting M1 and directed to induce suitable plasticity processes possibly reducing tic generation [132] . We further investigated the role of dopamine in tic generation by running simulations where we gradually increased the level of the dopamine bursts . In particular , we considered dopaminergic bursts having a peak that increased from 0 to 100 in 17 steps ( the burst had a Gaussian shape with 0 . 020 sec of standard deviation ) . M1 received random inputs simulating afferent signals received from other cortical areas . The inputs had a Gaussian-shape in correspondence to the dopamine bursts , in particular they had a height randomly drawn from the range of ( 0 , 900 ) and had a standard deviation of 0 . 040 sec . For each level of dopamine , we analysed the data collected in 30 trials of 40 simulated subjects . The results are shown in Fig 11 . The model predicts that the number of tics progressively increases with the size of the dopamine bursts . This result confirms what was said in relation to Fig 9 , showing how stronger dopaminergic bursts lead to a higher probability of producing tics . A statistical analysis supports the results shown in Fig 11 . In particular , the data were analysed through a one-way ANOVA having as factor the dopamine levels . The ANOVA shows that the dopamine level has a significant effect on the number of tics ( p < 0 . 001 ) . We evaluated how the data collected with the model were sensitive to the variations of the values assumed by the parameters when running the optimisation procedure discussed in section “Simulation settings” . In particular , we restricted the analysis to the best parameter sets found by the optimisation procedure during the whole search , namely to those that produced a high fit of the model to the target empirical data . To this purpose , we selected the parameter sets having a fitting data error within the first quartile ( this amounted to selecting the parameter sets having a simulated-real data error smaller than 0 . 08 ) . The analyses focused on the standard deviation of the ( normalized ) values of the parameters sets selected in such a way . The focus on the standard deviation was based on the idea that a small variance of a parameter indicated a great influence of the parameter on the model behaviour: indeed , the values of the parameter which were far away from its mean were associated , with a high probability , with a worse data fitting by the model and so were discarded by the procedure illustrated above related to the selection of the best parameter sets . Fig 12 shows the standard deviation of parameters computed with such procedure . The figure shows that the most important parameters to ensure a good fit of the target data by the model are those involving the ThBC efferent connections reaching M1 ( wThBC → M1 ) : this indicates that ThBC might be important as it integrates information from BG ( dishinibition ) and from Cer ( activation ) and we have seen that a concurrent activation of BG and M1 , supported by Cer , is important for the production of tics . In this respect , note the lesser importance of the Th reached only by Cer ( wThC → M1 ) . The STN efferent connections reaching GPi ( wSTN → GPi ) , GPi parameters ( rGPi ) and other parameters related to the indirect pathway ( wGPe → GPi , rGPe ) are also very important , stressing the relevance of BG activation to trigger tics . The efferent connections of M1 towards the ThC ( wM1 → ThC ) , Cer ( wM1 → Cer ) and BG ( wM1 → STN ) , as well as those from the Cer towards Th ( wCer → ThBC , wCer → ThC ) , have a medium importance . At the opposite side of the spectrum , we find the parameters related to the efferent connections of M1 towards the ThBC ( wM1 → ThBC ) and towards Dorsal putamen ( wM1 → StrD2 , wM1 → StrD1 ) and the connection linking BG to Cer ( wSTN → Cer ) .
This work proposes a computational system-level model that reproduces the recent data obtained in [49] , proposes a detailed hypothesis of the brain mechanisms that might possibly underlie them , and produces predictions that point to new brain areas as targets for future therapeutic interventions . In particular , the model furnishes an explanation of the neural mechanisms underlying Tourette syndrome that pivots on integrated basal ganglia-thalamo-cortical action selection processes and on the recently discovered subthalamic nucleus-pons-cerebellar connection [29 , 30] . Notwithstanding its novelty , the model has relevant limitations that represent starting points for future research . First , future versions of the model could include more detailed versions of thalamus and cortex ( e . g . , [76 , 77 , 139 , 140] ) in order to study more in detail the mechanisms through which such brain components contribute to tic production . Indeed , the dynamics of the thalamo-cortical subsystem are very important for the production of motor movements [72] , and so their better understanding might also be important for a better view of the production of dysfuntional motor tics . Similarly , future versions of the model could also study the effects of dopamine in the subthalamic nucleus in Tourette syndrome [141] as well as the results of recent data on the role of the nucleus accumbens and the related limbic network in tic generation [142] . Second , future research could investigate how the increased cerebellar activation by the subthalamic nucleus could modulate the tic intensity through the cerebello-thalamo-cortical circuit . This view is in line with previous theoretical proposals [46 , 65 , 66] and empirical evidence [37 , 143] highlighting the role of the basal ganglia for triggering movements and of the cerebellum for motor pattern amplification . Empirical evidence indicates that an increased activation of the cerebellum tends to cause an increased activity to the primary motor cortex through the thalamo-cortical pathway [44 , 144] . Moreover , evidence also indicates that cerebellar hyperactivity may indeed contribute to tic production in Tourette syndrome patients [48 , 145] . Third , the model could be modified to reproduce the dopamine-based learning processes of basal ganglia , [146] , and also the inferior olive-climbing fibers circuit believed to provide error signals to the cerebellar cortex [147] , to investigate how these plasticity events may affect tic emergence in Tourette syndrome . In this respect , the model could be used to address data suggesting that unmedicated individuals with Tourette syndrome learn better from rewards than from punishments [148 , 149] . Along the same line , the model could be used to study recent findings suggesting that the involuntary and recurrent nature of tics could be a manifestation of overlearned motor patterns due to excessive LTP in the cerebellar cortex [147] , an hypothesis supported by the fact that such abnormal learning processes can be interrupted by modulation of cerebellar activity through non-invasive brain stimulation [150] . Lastly , the model proposes one possible role of the subthalamic-pons-cerebellar circuit [29 , 30] . The discovery of these connections has raised fundamental questions on how basal ganglia and cerebellum might directly influence each other [35 , 55] . In this respect , the model represents the first computational proposal suggesting a possible role of the basal ganglia-cerebellum connection , in particular assigning to the cerebellum a role in tic production . However , alternative possible roles based on the literature should be compared to this one . For example , the subthalamic nucleus is part of the indirect pathway of the basal ganglia and is implicated in action inhibition and aversive learning [151 , 152] . Thus , another possible role of the subthalamic-pons-cerebellar circuit might be to provide a stop signal to the cerebellum for withholding ongoing movements [45] . Another possibility might be that the subthalamic nucleus signals the cerebellum an “off-line” status of the system , in particular that the subthalamic nucleus itself is withholding motor programs via the excitation of globus pallidus and substantia nigra reticulata in turn inhibiting respectively the thalamus and midbrain motor nuclei . The purpose of this would be to allow the cerebellar internal models to be safely used for off-line mental simulation [45 , 153 , 154] . Further research will be necessary to understand the functions of the newly discovered pathway [35 , 55] . Notwithstanding the need for these further studies , we think the model offers a system-level framework supporting our understanding of the brain mechanisms underlying tic production . This framework is expected to support an increasingly integrated interpretation of existing data , and also the design of novel empirical experiments and therapeutic interventions under the guidance of a wider systemic perspective . | Tourette syndrome is a neuropsychiatric disorder characterized by vocal and motor tics . Tics represent a cardinal symptom traditionally associated with a dysfunction of the basal ganglia leading to an excess of the dopamine neurotransmitter . This view gives a restricted clinical picture and limits therapeutic approaches because it ignores the influence of altered interactions between the basal ganglia and other brain areas . In this respect , recent evidence supports a more articulated framework where cerebellum and cortex are also involved in tic production . Building on these data , we propose a computational model of the basal ganglia-cerebellar-thalamo-cortical network to investigate the specific mechanisms underlying motor tic production . The model reproduces the results of recent experiments and suggests an explanation of the system-level processes underlying tic production . Moreover , it furnishes predictions related to the amount of tics generated when there are dysfunctions in the basal ganglia-cerebellar-thalamo-cortical circuits . These predictions could be important in identifying new brain target areas for future therapies based on a system-level view of Tourette syndrome . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"neuropsychiatric",
"disorders",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"dopaminergics",
"brain",
"vertebrates",
"neuroscience",
"organic",
"compounds",
"animals",
"mammals",
"hormones",
"primates",
"amines",
"neurotransmitters",
"cerebellum",
"catecholamines",
"dopamine",
"monkeys",
"basal",
"ganglia",
"neurochemicals",
"chemistry",
"motor",
"cortex",
"thalamus",
"mental",
"health",
"and",
"psychiatry",
"biochemistry",
"organic",
"chemistry",
"anatomy",
"tourette",
"syndrome",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"amniotes",
"cerebral",
"cortex",
"organisms"
] | 2017 | Dysfunctions of the basal ganglia-cerebellar-thalamo-cortical system produce motor tics in Tourette syndrome |
Mycobacterium ulcerans is the causative agent of Buruli ulcer ( BU ) . This nontuberculous mycobacterial infection has been reported in 34 countries worldwide . In Australia , the majority of cases of BU have been recorded in coastal Victoria and the Mossman-Daintree areas of north Queensland . Mosquitoes have been postulated as a vector of M . ulcerans in Victoria , however the specific mode of transmission of this disease is still far from being well understood . In the current study , we trapped and analysed 16 , 900 ( allocated to 845 pools ) mosquitoes and 296 March flies from the endemic areas of north Queensland to examine for the presence of M . ulcerans DNA by polymerase chain reaction . Seven of 845 pools of mosquitoes were positive on screening using the IS2404 PCR target ( maximum likelihood estimate 0 . 4/1 , 000 ) . M . ulcerans DNA was detected from one pool of mosquitoes from which all three PCR targets: IS2404 , IS2606 and the ketoreductase B domain of mycolactone polyketide synthase gene were detected . None of the March fly samples were positive for the presence of M . ulcerans DNA .
Buruli ulcer ( BU ) , also known regionally as Daintree ulcer in north Queensland , Australia or Bairnsdale ulcer in Victoria , Australia , is an emerging disease of skin and underlying tissue , with a potential to lead to permanent disability , particularly if treatment is inadequate or delayed . The causative agent of this disease , M . ulcerans secretes a polyketide exotoxin , mycolactone , the production of which requires expression of a series of contiguous genes on the large pMUM001 plasmid . This exotoxin is the main virulence determinant of the bacteria [1] . The outbreaks of BU have been consistently linked with wetland or coastal regions [2] . Environmental samples such as water , aquatic plants , soil at endemic areas has been found PCR-positive for M . ulcerans DNA [3 , 4] . Insects such as mosquitoes and aquatic bugs has been proposed as a vital ecological niche for the maintenance of pathogen in environment [5 , 6] . The detection of M . ulcerans DNA in insects does not prove their ability to transmit M . ulcerans but could indicate potential to act as either biological or mechanical vector . A study conducted by Marsollier and his colleagues provided evidence of the presence of M . ulcerans DNA in the salivary gland of wild caught Naucoridae ( aquatic bug ) . They successfully isolated the pathogen by culture from the salivary glands of aquatic bugs and suggested aquatic insects as having an important ecological niche in the maintenance of the organism in the environment . They were also able to demonstrate transmission to mice in a laboratory environment [6] . Similarly , a study conducted by Wallace et al . provided evidence of the ability of mosquitoes to act as a mechanical vector of M . ulcerans [7] . Studies conducted in endemic areas of Africa suggest that conducting farming activities close to rivers [8] and swimming in rivers located in endemic areas [9] are risk factors for exposure to M . ulcerans . In Australia , foci of BU infection have been found in tropical north Queensland [10 , 11] , the Capricorn Coast region of central Queensland [10] , the Northern Territory [12] and temperate coastal Victoria [5] . In Queensland , Australia , cases of Daintree ulcer have been reported primarily in Douglas Shire , exclusively in the vicinity of Wonga , Miallo and Daintree [10 , 11] . A few cases has also been reported from Capricorn Coast region of central Queensland [10] . The Douglas Shire covers an area of 2 , 445 sq . Kms and the total population is around 11 , 000 . A majority of the population ( around 70% ) reside in Port Douglas and Mossman . Thus , the Daintree ulcer endemic areas in north Queensland is sparsely populated . There has been a significant decrease in human cases of BU in north Queensland , since a large outbreak in 2011–2012 , when more than 60 cases were reported . This outbreak occurred after prolonged and heavy rainfall in 2010–2011 [11] . The average reported rate over fifteen years period from 2002–2016 was 0 . 2 cases/100 , 000 population per year [13] . Victorian researchers detected the presence of M . ulcerans DNA in five different species of mosquito during a BU outbreak in an endemic area of Victoria , Australia . They demonstrated the absence of M . ulcerans in a neighboring area , where BU did not occur [5] . Together , the evidence was proposed to support a link with mosquitoes in the ecology of BU in Victoria [5 , 14] . More recently , a small study conducted in the BU endemic region of north Queensland , found that of 35 insect/insects pools , one sample of an individual mosquito and one pool of two mosquitoes were positive for IS2404 . The IS2404 positive mosquito pool contained DNA of a closely related M . ulcerans subspecies that had a low copy number for IS2606 which does not commonly cause disease in human . The individual mosquito had insufficient DNA for detection of the additional gene targets . The study highlighted a need to examine a larger sample size to gauge the significance of the role of mosquito in ecology of BU in Northern Queensland [15] . An additional suggestion proposed by the local population ( including people with a history of BU ) was that March flies ( Tabanidae ) might have a role in transmission . We therefore aimed , in this study to capture and screen mosquitoes and March flies for the presence of M . ulcerans DNA in the BU endemic area of Northern Queensland .
Mosquitoes were captured using a model 512 “CDC miniature light trap” ( John W . Hock Company , Gainesville Florida USA ) baited with 1 kg of dry ice as the source of CO2 . This trap is the most reliable , efficient and portable device for trapping mosquitoes and sand flies [17] . This trap consists of an electric light and fan just over the collection container and is operated by a 12V battery . A two liter insulated container was used to hold dry ice and a pipe was attached to release CO2 over the trap to attract mosquitoes ( Fig 2 ) . Thirty overnight trapping sessions were conducted starting from September 2016 through to February 2018 , with at least 4 CDC traps placed within a 1 kilometer radius of each-other . Of the 30 trapping sessions , 14 were conducted at eight different sites within region-1 , nine at six different sites within region-2 and seven at five different sites of region-3 ( Fig 1 ) . Traps were placed at different sites after obtaining permission to access properties from the owners and selection of sites were based on history of BU cases in humans in nearby households . Geographical Information System ( GIS ) coordinates of each trap was recorded . On each occasion , traps were set before dusk and checked for mosquitoes after dawn the next morning . After each occasion of trapping , catches were transported to the Mosquito Research Facility , Australian Institute of Tropical Health and Medicine ( AITHM ) , James Cook University , Cairns , Australia where they were counted , sorted and pooled by genus , with each pool containing ≤ 20 mosquitoes of same genus and collected from the same site . The key of Russell was used to identify the genus of mosquitoes trapped [18] . Several attempts were made to trap march flies from endemic areas with an investigator wearing dark clothes to attract them , or with the use of an insect net sprayed with insecticide . These attempts occurred from February 2016 through September 2016 . The yield from these attempts were very low . A request was made to residents of region-1 through the local State School to collect march flies . This effort was successful and large numbers of March flies of genus Tabanus were collected by the local community . The addresses of properties from which March flies were collected were recorded . Sampling of March flies was restricted to region-1 . The molecular analyses were performed using the protocol available on given link: dx . doi . org/10 . 17504/protocols . io . vqbe5sn . DNA was extracted from each pools of ≤ 20 mosquitoes of the same genus by using the FastPrep Instrument ( MP Biomedicals , Solon , OH , USA ) as per manufacturer’s instruction with FastDNA Kit ( MP Biomedicals ) . Using the same instrument , DNA from individual March fly was extracted with FastDNA Spin Kit ( MP Biomedicals ) . One sterile water sample in each batch of extractions was used as a negative control to identify the possible contamination during the process of extraction of DNA . Extracted DNA was stored at -20 oC . The extracted DNA samples were screened for the presence of M . ulcerans DNA by using a semi-quantitative real-time PCR adapted from a method for the detection of M . ulcerans DNA from environmental samples [19] . To rule-out the possibility of contamination , three negative controls ( double deionized water , MilliQ ) and three positive controls ( purified M . ulcerans DNA obtained from Victorian Infectious Disease Reference Laboratory ) were used during qPCR assay run . All of the extracted DNA samples were initially screened for the M . ulcerans insertion sequence ( IS ) element IS2404 . Samples positive for IS2404 were re-analyzed by a second real-time PCR for the detection of two additional regions in the genome of M . ulcerans: IS2606 and ketoreductase B domain ( KR ) . This screening process has been validated by Fyfe et al . to differentiate M . ulcerans from other mycolactone producing mycobacteria ( MPM ) [19] . They suggested that the difference in real-time PCR cycle thresholds ( Ct ) between IS2606 and IS2404 ( ΔCt [IS2606 –IS2404] ) allows for the differentiation of M . ulcerans strains commonly causing disease in human from other MPM ( which are also considered members of the species M . ulcerans ) that contain IS2404 but which have fewer copy numbers of IS2606 . Samples containing all three independent DNA sequences and with expected Ct values were considered positive for M . ulcerans DNA . The software recommended by Centers of Disease Control and Prevention ( Atlanta , GA , USA ) was used to calculate the maximum likelihood estimate ( MLE ) per 1 , 000 mosquitoes tested ( bias corrected MLE ) [20] . The Genebank accession number of nucleotide sequence on M . ulcerans gene IS2404 , IS2606 and KR have been allocated as BX649209 , BX649209 and BX649209 respectively .
A total of 16 , 900 mosquitoes were captured over the course of the study from 30 occasions of trapping at three different regions of northern Queensland . Total mosquitoes captured from region-1 , region-2 and region-3 were 7880 , 5100 , and 3920 , respectively . The majority of captured mosquitos belonged to the Verrallina genus ( specifically Verrallina lineata ) 82% , followed by Coquillettidia ( 9% ) and Mansonia ( 3% ) . The remaining 6% consisting seven other genera that were classified as “other” for screening . See Fig 3 below . Of a total of 16 , 900 mosquitoes screened ( 845 pools ) , seven pools were positive for IS2404 . Three of those seven pools were Verrallina sp . from region-1 , two pools were Coquillettidia sp . one each from capture region-1 and 3 and the remaining two pools were Mansonia sp . from region-1 . Of the seven pools positive for IS2404 , two pools had a high cycle threshold ( Ct ) values for IS2404 and did not contain sufficient amount of DNA to detect IS2606 and KR . IS2606 was not detected from four pools , despite of having desired Ct values for IS2404 . All three targets were detected from remaining pool ( Table 1 ) . Thirty pools of mosquitoes which were negative for IS2404 were tested for IS2606 and KR . None of them were positive for these probes signifying the dependent nature of existence of IS 2606 and KR with IS2404 . Similar findings were reported during the Victorian outbreak [5] . The bias corrected MLE value for all mosquitoes collected from study site ( region-1 , region-2 and region-3 ) was 0 . 06 M . ulcerans PCR-positive mosquitoes per 1 , 000 tested ( 95% confidence interval , 0 . 00–0 . 29 ) . Only Region-1 had M . ulcerans PCR-positive mosquitoes and calculated MLE value was 0 . 13 ( 95% confidence interval , 0 . 01–0 . 61 ) /1 , 000 mosquitoes tested .
Mosquitoes serve as important biological vectors for a variety of pathogens . The movement of pathogens from the gastro-intestinal tract after ingestion to the salivary glands for subsequent transmission is well documented for many diseases . However , this phenomenon has not been demonstrated for M . ulcerans . A study conducted by Wallace and colleagues ( 2010 ) provided evidence on the maintenance of M . ulcerans throughout larval development without further passage of the organisms into pupa or adult mosquitoes [21] . They concluded that mosquitoes were an unlikely biological vector of M . ulcerans . Wallace et al ( 2017 ) subsequently provided evidence of mechanical transmission of M . ulcerans via anthropogenic skin puncture or mosquito bites [7] . For mechanical transmission , insect vectors such as mosquitoes must acquire the pathogen either from the environment or an infected host . For this to occur efficiently , the organism must be abundantly present in the environment . A survey in Victoria , Australia has confirmed a strong correlation between mosquitoes found to test positive for carrying M . ulcerans DNA and the number of human cases of BU occurring [5 , 22] . The group found a significantly higher number of mosquitoes screened positive for M . ulcerans DNA during an intense outbreak of BU in endemic areas , in comparison to areas with a lower incidence of human cases . The number of human cases of BU has decreased in Northern Queensland , Australia since the largest recorded outbreak in 2011 ( > 60 cases ) . The majority of the cases during the 2011 outbreak were from Wonga and the Wonga beach area , referred as region-1 in the study by Steffen and Freeborn ( 2018 ) [23] . Out of 394 pools collected in region 1 , six pools were positive for IS2404 DNA in this study . Interestingly , three pools mosquitoes of these positive pools were trapped in the backyard of a property in Wonga Beach area ( region-1 ) where two human cases of BU were confirmed in 2017 . All other pools of mosquitoes and march flies collected from that properties negative for M . ulcerans DNA . As shown in the result , seven pools of mosquitoes were positive for IS2404 . However , all three targets with expected Ct value were detected from only one of these seven pools . Samples that were positive for only IS2404 were not considered further . In north Queensland , the Daintree River arises in mountainous rainforest region around the town of Mossman and flows into the sea at Cape Tribulation . The wet season starts normally from November/December and continues up to April , and the dry season starts from May and continues up to October/November . Outbreaks of human cases of BU in north Queensland have been linked with heavy rainfall and flooding . This survey was conducted from September 2016 through to February 2018 , when dryer environmental conditions prevailed . Out of seven M . ulcerans DNA positive pools of mosquitoes , five were collected in wet season and two were collected in dry season . A majority of cases of Daintree ulcer are reported after rainy season ends [13] . The estimated mean incubation period of Daintree ulcer is 4 . 5 months [24] . Thus , it is more likely that the transmission occurs in the wet season which justifies the detection of M . ulcerans DNA from the pools of mosquitoes that were captured in wet season in this study . In a separate study conducted in North Queensland , Australia , one sample of a single mosquito and one pool of two mosquitoes was found positive for IS2404 . [15] . However , it must be noted that this study was conducted soon after 2011 which raises the possibility that sampling should occur as close as possible in time to when transmission is thought to be occurring . M . ulcerans is an environmental pathogen and detection of M . ulcerans DNA positive mosquitoes may only be an indicator for the presence of the organism in the environment . A significant decrease in human cases of BU in Northern Queensland in recent years could be due to a lower load of bacteria in the environment . This may explain the low detection of M . ulcerans DNA positive mosquitoes and March fly populations in the study sites . However , the detection of M . ulcerans DNA even in a single pool of mosquitoes from the endemic areas of Northern Queensland is significant , as it corroborates findings in Victoria where five different species of mosquitoes captured from BU-endemic regions during human outbreaks were positive for M . ulcerans . Our detection of M . ulcerans DNA in mosquitoes in Northern Queensland does support the earlier report from Victoria in Australia [5] . The Victorian study provides evidence for high detection rates of M . ulcerans positive mosquitoes if captured during peak times of outbreaks . Our study found that it is less likely to find M . ulcerans positive mosquitoes if they are trapped from areas where human incidence of BU is currently low . We hypothesize that mosquitoes and perhaps other biting insects , such as March flies may have a significant role in the ecology and transmission of M . ulcerans in endemic areas during outbreaks and that the level of detection of M . ulcerans positive mosquitoes in the environment could be an indicator for disease outbreaks .
Our study confirms the presence of M . ulcerans DNA in the mosquitoes samples captured from the BU-endemic regions of North Queensland , Australia . Lower detection of M . ulcerans positive mosquitoes in BU-endemic areas in North Queensland may partially explain low endemicity of the disease . | The causative agent of Buruli ulcer is Mycobacterium ulcerans . This destructive skin disease is characterized by extensive and painless necrosis of skin and underlying tissues usually on extremities of body due to production of toxin named mycolactone . The disease is prevalent in Africa and coastal Australia . The exact mode of transmission and potential environmental reservoir for the pathogen still remain obscure . Aquatic and biting insects have been identified as potential niche in transmission and maintenance of pathogen in the environment . In this study we screened mosquitoes and march flies captured from endemic areas of northern Queensland for the presence of M . ulcerans DNA . We found seven pools of mosquito out of 845 pools positive for IS2404 . In only one of the seven samples were the additional targets IS2606 and KR detected . None of the March fly samples were positive . The results could indicate a low burden of the bacteria in the environment coinciding with a comparatively low number of human cases of M . ulcerans infection seen during the trapping period of the study . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion",
"Conclusions"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"population",
"genetics",
"geographical",
"locations",
"tropical",
"diseases",
"australia",
"animals",
"gene",
"pool",
"bacterial",
"diseases",
"signs",
"and",
"symptoms",
"ulcers",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"population",
"biology",
"insect",
"vectors",
"bacteria",
"extraction",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"buruli",
"ulcer",
"artificial",
"gene",
"amplification",
"and",
"extension",
"actinobacteria",
"molecular",
"biology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mycobacterium",
"ulcerans",
"dna",
"extraction",
"mosquitoes",
"eukaryota",
"diagnostic",
"medicine",
"polymerase",
"chain",
"reaction",
"oceania",
"genetics",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"evolutionary",
"biology",
"organisms"
] | 2019 | A survey on Mycobacterium ulcerans in Mosquitoes and March flies captured from endemic areas of Northern Queensland, Australia |
How small heat shock proteins ( sHsps ) might empower proteostasis networks to control beneficial prions or disassemble pathological amyloid is unknown . Here , we establish that yeast sHsps , Hsp26 and Hsp42 , inhibit prionogenesis by the [PSI+] prion protein , Sup35 , via distinct and synergistic mechanisms . Hsp42 prevents conformational rearrangements within molten oligomers that enable de novo prionogenesis and collaborates with Hsp70 to attenuate self-templating . By contrast , Hsp26 inhibits self-templating upon binding assembled prions . sHsp binding destabilizes Sup35 prions and promotes their disaggregation by Hsp104 , Hsp70 , and Hsp40 . In yeast , Hsp26 or Hsp42 overexpression prevents [PSI+] induction , cures [PSI+] , and potentiates [PSI+]-curing by Hsp104 overexpression . In vitro , sHsps enhance Hsp104-catalyzed disaggregation of pathological amyloid forms of α-synuclein and polyglutamine . Unexpectedly , in the absence of Hsp104 , sHsps promote an unprecedented , gradual depolymerization of Sup35 prions by Hsp110 , Hsp70 , and Hsp40 . This unanticipated amyloid-depolymerase activity is conserved from yeast to humans , which lack Hsp104 orthologues . A human sHsp , HspB5 , stimulates depolymerization of α-synuclein amyloid by human Hsp110 , Hsp70 , and Hsp40 . Thus , we elucidate a heretofore-unrecognized human amyloid-depolymerase system that could have applications in various neurodegenerative disorders .
Amyloid fibers are thread-like protein polymers with cross-β structure . These unusually stable , self-templating structures were first identified in various systemic amyloidoses and neurodegenerative disorders such as Alzheimer's disease [1] . In isolation , many proteins can form amyloid fibers , suggesting that amyloidogenesis is an intrinsic property of polypeptides [1]–[3] . Indeed , amyloid conformers have been captured during evolution for various beneficial purposes , including prion-based transmission of advantageous phenotypes , long-term memory formation , melanosome biogenesis , drug resistance , and biofilm formation [4]–[14] . Moreover , as stable self-organizing polymers , amyloids are interesting nanomaterials [15]–[19] . Thus , in diverse fields there is an urgent need to understand how we can promote or reverse amyloidogenesis as necessary . We hypothesized that small heat shock proteins ( sHsps ) might enable control of amyloidogenic trajectories . sHsps are the most widespread family of molecular chaperones [20] , [21] . sHsps protect cells from diverse environmental stresses by suppressing amorphous aggregation of denatured proteins [20]–[23] . All sHsps harbor a conserved C-terminal α-crystallin domain of ∼90 residues , but are otherwise diverse in size and sequence . Typically , sHsps form large dynamic oligomers and function as ATP-independent chaperones that bind denatured proteins to prevent aggregation [20]–[23] . sHsps maintain proteins in a soluble form that can be reactivated by Hsp70 [24]–[26] . The yeast cytosol harbors two sHsps: Hsp42 and Hsp26 . Both form large dynamic oligomers of 24 subunits [27]–[30] . Hsp42 is more abundant and prevents protein aggregation at physiological and heat shock temperatures [28] . By contrast , Hsp26 is activated as a chaperone at elevated temperatures via complex changes in the quaternary dynamics of its oligomer [27] , [29] , [31] , [32] . Hsp26 and Hsp42 display overlapping and broad substrate specificity [28] . Incorporation of Hsp26 into denatured aggregates can promote their dissolution and renaturation by Hsp104 and Hsp70 [26] , [33] . However , if Hsp26 is added to preformed denatured aggregates , then it cannot assist Hsp104 and Hsp70 [33] . Much less is known about Hsp42 , which might be involved in aggregate partitioning in vivo [34] . Whether Hsp42 interacts directly with Hsp104 or Hsp70 is unknown . Despite these advances in understanding how sHsps handle denatured proteins , much less is known about how sHsps might interface with amyloidogenic folding pathways . sHsps might inhibit different steps in amyloidogenesis of various disease proteins , such as α-synuclein ( α-syn ) , polyglutamine , or Aβ40 [23] , [33] , [35]–[38] . Yet it is unknown whether sHsps enable the proteostasis machinery to disassemble pathological amyloid . Furthermore , it is unknown whether sHsps regulate beneficial amyloid . In this study , we address these issues by first employing the yeast translation termination factor , Sup35 ( Figure 1 ) . Sup35 forms infectious amyloids ( prions ) that transmit heritable reductions in translation termination fidelity and comprise the yeast prion [PSI+] [8] . [PSI+]-encoded reductions in translation termination fidelity confer phenotypic diversity and selective advantages to yeast in diverse environments [8] , [10] , [11] , [13] , but can be deleterious in other settings [39] . Sup35 is a valuable paradigm for studying prion-folding events , with analytical tools that are unavailable for other amyloids or prions . Indeed , these tools have helped clarify how Sup35 prions assemble [40] , [41] . They have also revealed various aspects of Sup35 prion structure at the resolution of spatial arrangements of individual amino acids involved in inter- and intra-molecular contacts ( Figure 1 ) [40]–[42] . These tools provide unique opportunities to establish a detailed understanding of how sHsps affect prion assembly and disassembly . The two N-terminal domains of Sup35 , termed NM , confer all the properties needed to form a stable prion in yeast ( Figure 1 ) [8] . In isolation , NM spontaneously forms prions by a well-defined mechanism that involves a lag phase and assembly phase [40] , [43] , [44] . Early in lag phase , NM partitions between a monomeric ( ∼90% total NM ) and oligomeric ( ∼10% total NM ) pool ( Figure 1 , step 1 ) [40] , [43] , [45] , [46] . NM monomers are largely unstructured and populate multiple transient conformations [47] . However , the specific intermolecular contacts required for prion formation ultimately form in molten NM oligomers . NM monomers within structurally fluid oligomers gradually reorganize ( Figure 1 , step 2 ) to form amyloidogenic oligomers ( Figure 1 , step 3 ) , which are structurally distinct to fibers [40] , [43] , [48] , [49] . The intermolecular contacts that define prions form very rapidly once these obligate , transient intermediates appear ( Figure 1 , step 4 ) [40]–[42] , [48] , [49] . Amyloid fibers then seed their own rapid bidirectional assembly by capturing and converting monomers to the cross-β form ( Figure 1 , step 5 ) [43] , [46] , [50] . Short prion recognition elements within the N-terminal domain ( N ) , termed the “Head” and “Tail” , are proposed to make homotypic intermolecular contacts in assembled prions [40]–[42] , [51] , [52] . Thus , prions are maintained by alternating Head-to-Head and Tail-to-Tail contacts that separate a central core ( Figure 1 ) . Both the Head and Tail regions can nucleate prion assembly , although the rate-limiting step of lag phase is the establishment of the Head-to-Head contact [40]–[42] , [51] . This well-defined sequence of prion-folding events provides an unparalleled opportunity to understand how sHsps affect prion formation at a molecular level . How Hsp26 and Hsp42 might affect prion-folding events in yeast is unclear . Both Hsp26 and Hsp42 are found to be associated with ex vivo SDS-resistant prion aggregates [53] , but deletion of Hsp26 does not affect [PSI+] propagation [12] and overexpression of Hsp26 or Hsp42 does not cure [RNQ+] [33] . However , beyond these observations nothing is known about how these sHsps might affect prion-folding events . It is also unclear whether sHsps contribute to the dissolution of amyloid or prion conformers by the proteostasis network . In yeast , the protein disaggregase and AAA+ ATPase , Hsp104 , can rapidly disassemble amyloid conformers [48] , [49] , [54]–[58] . Overexpression of Hsp26 or Hsp42 together with Hsp104 can increase soluble levels of polyglutamine in yeast , but whether this reflected enhanced disaggregation or inhibition of aggregation remains unknown [33] . Curiously , metazoan proteostasis networks lack an Hsp104 homologue [59] . Thus , it is unclear how amyloid dissolution is catalyzed in these systems [60] . We have recently defined a mammalian disaggregase machinery composed of Hsp110 ( Apg-2 ) , Hsp70 ( Hsc70 or Hsp70 ) , and Hsp40 ( Hdj1 ) , which resolves denatured aggregates , but does not rapidly remodel amyloid [61] . Yet amyloid fibers are dynamic entities and monomers at fiber ends can slowly dissociate and rapidly reassociate [62]–[68] . Whether proteostasis networks capitalize on this molecular recycling to promote gradual amyloid depolymerization is unknown . Here , we define how sHsps regulate beneficial Sup35 prions and potentiate amyloid dissolution .
Using complementary methods , including Thioflavin-T ( ThT ) fluorescence and SDS-resistance , we demonstrated that Hsp42 potently inhibited ( IC50∼0 . 67 µM of Hsp42 monomer ) spontaneous NM fibrillization ( Figure 2A , B blue markers ) . Marked inhibition was observed at a ratio of NM∶Hsp42 of 10∶1 and assembly was abolished at an NM∶Hsp42 ratio of 1 . 67∶1 ( Figure 2A , B ) . Hsp26 also inhibited ( IC50∼1 . 1 µM ) spontaneous NM fibrillization ( Figure 2A , B red markers ) . Hsp26 was not as effective as Hsp42 , but inhibition was observed at a ratio of NM∶Hsp26 of 4∶1 ( Figure 2A , B ) . A 5-fold molar excess of Hsp26 was needed to completely block NM fibrillization . These inhibitory effects truly precluded prion formation because NM incubated in the presence of sHsps failed to transform [psi−] cells to [PSI+] ( Figure 2C ) . Typically , sHsps bind 1 substrate per ∼2–3 sHsp monomers [24] , [27] . Thus , the strong inhibition at substoichiometric concentrations indicates that the sHsps might inhibit a rare or transient NM conformer that is critical for prion formation . Remarkably , the inhibitory activities of Hsp26 and Hsp42 were synergistic . The combination of an equimolar mixture of Hsp26 and Hsp42 was a more potent inhibitor ( IC50∼0 . 16 µM , i . e . 0 . 08 µM of each sHsp ) of NM assembly than either sHsp alone ( Figure 2A–C ) . Hsp26 and Hsp42 also synergized to inhibit spontaneous prionogenesis of full-length Sup35 ( Figure 2D ) . To the best of our knowledge , this is the first example of two distinct sHsps working together in a synergistic manner to prevent prion formation . The synergistic inhibition of Sup35 prionogenesis by Hsp26 and Hsp42 suggested that the two sHsps might inhibit prion formation by distinct mechanisms . The ability of the sHsps to abrogate prion formation at substoichiometric concentrations also suggested interference with a specific conformer or intermediate that is initially present at low concentrations . Two non-mutually exclusive possibilities emerge . First , sHsps might mask the ends of de novo formed NM fibers and thus prevent seeded assembly ( Figure 1 , step 5 ) . Second , sHsps might antagonize the formation ( Figure 1 , step 1 ) or reorganization ( Figure 1 , steps 2–4 ) of transient molten oligomers . Only a small fraction ( ∼10% ) of the total NM accesses molten oligomeric forms [45] , which are obligate reaction intermediates for spontaneous fibrillization ( Figure 1 ) [48] , [49] . These malleable NM oligomers possess a hydrodynamic radius of 50–130 nm , form extremely rapidly , and can be recovered by ultracentrifugation [45] , [55] . Importantly , neither Hsp26 nor Hsp42 alone or in combination inhibited the formation ( Figure 1 , step 1 ) of NM oligomers ( Figure 3A ) . By contrast , the combination of Ssa1 ( an Hsp70 ) and Ydj1 ( an Hsp40 ) inhibited oligomer formation ( Figure 3A ) [55] . After their initial formation , molten NM oligomers gradually reorganize into amyloidogenic forms ( Figure 1 , steps 2 and 3 ) that ultimately elicit assembly phase ( Figure 1 , step 4 ) . This maturation process can be tracked using single cysteine NM mutants labeled with acrylodan at specific positions [40] , [41] . Sequestration of labeled sites from solvent yields increases in acrylodan fluorescence . Specific portions of N ( ∼residues 21–106 ) gradually become solvent inaccessible in molten oligomers prior to fiber assembly ( Figure 1 , steps 2 and 3 ) . This process begins immediately and the maximal increase in acrylodan fluorescence signals the end of lag phase and the start of assembly phase [40] . To determine whether Hsp26 or Hsp42 interfered with this process we utilized NM with acrylodan attached to cysteines replacing Asn21 , Gln38 , Gly96 , or Tyr106 . These mutated and labeled NM variants retain wild-type assembly kinetics and ability to access infectious forms [40]–[42] . We measured acrylodan fluorescence after 15 min , when the assembly reaction remained in lag phase ( Figure 3B ) . Hsp26 had only a slight inhibitory effect on increases in acrylodan fluorescence at all positions tested , suggesting that Hsp26 does not interfere with oligomer maturation ( Figure 3B; Figure 1 , steps 2 and 3 ) . By contrast , Hsp42 inhibited increases in acrylodan fluorescence at all positions tested ( Figure 3B ) . We conclude that Hsp42 inhibits spontaneous NM fibrillization by preventing the maturation of molten NM oligomers ( Figure 1 , steps 2 and 3 ) . Remarkably , Hsp26 and Hsp42 together caused the greatest inhibition of increased acrylodan fluorescence ( Figure 3B ) . Next , we determined how Hsp42 or Hsp26 affected an obligate on-pathway oligomeric intermediate in spontaneous Sup35 assembly , which accumulates during lag phase and is specifically detected by the conformation-specific antibody A11 [48] , [49] , [69] . This oligomeric species is most abundant at the end of lag phase and then rapidly disappears during assembly phase ( Figure 1 ) [48] , [49] . We performed a kinetic experiment where NM assembly was initiated for 10 min at which time either buffer , Hsp26 , or Hsp42 were added ( arrow in Figure 3C , D ) and the reaction was then allowed to continue . After 10 min , A11-reactive species had already accumulated ( Figure 3C ) , whereas no fibers had assembled as determined by the lack of ThT fluorescence ( Figure 3D ) . Addition of buffer after 10 min had no effect on assembly and A11-reactive species continued to accumulate until the end of lag phase ( ∼30 min , Figure 3C , grey markers ) . A11-reactive species then rapidly declined as fiber assembly initiated ( ∼40 min , compare grey markers in Figure 3C to black markers in Figure 3D ) . Addition of Hsp42 after 10 min caused a rapid disappearance of A11-reactive species ( Figure 3C , cyan markers ) and fiber assembly was blocked ( Figure 3D , blue markers ) . Thus , Hsp42 reversed the formation of A11-reactive conformers . By contrast , addition of Hsp26 had no effect on the accumulation of A11-reactive species during lag phase or their rapid decline after 40 min ( Figure 3C , orange markers ) . Yet very little fiber assembly occurred ( Figure 3D , red markers ) . Thus , in contrast to Hsp42 , Hsp26 does not inhibit or reverse oligomer maturation events ( Figure 1 , steps 1–4 ) . Rather , these data suggest that Hsp26 inhibits the growth of newly formed fibers ( Figure 1 , step 5 ) . Indeed , electron microscopy revealed that a few very short fibers assembled in the presence of Hsp26 ( arrow in Figure 3E ) . By contrast , oligomeric structures persisted in the presence of Hsp42 , or Hsp26 and Hsp42 ( Figure 3E ) . To pinpoint which steps of Sup35 prionogenesis are antagonized by Hsp42 or Hsp26 , we experimentally bypassed the requirement for oligomer maturation in spontaneous prion formation ( Figure 1 , steps 2–4 ) . Thus , we crosslinked single cysteine NM mutants in specific positions with the flexible 11 Å crosslinker: 1 , 4-bis-maleimidobutane ( BMB ) . NM that is BMB-crosslinked at cysteines at position Gly25 or Gly31 assembles into fibers without a detectable lag phase [40] , [41] . Hsp26 or Hsp42 potently inhibited the assembly of uncrosslinked NM , and the combination of Hsp26 and Hsp42 was the most effective ( Figure 3F ) . By contrast , Hsp26 but not Hsp42 inhibited assembly of NM that was BMB-crosslinked at position 25 or 31 ( Figure 3F ) . The combination of Hsp26 and Hsp42 was less effective against NM that had been BMB-crosslinked at position 25 or 31 ( Figure 3F ) . These data suggest that Hsp26 selectively antagonizes events after lag phase ( Figure 1 , step 5 ) , whereas Hsp42 selectively antagonizes oligomer-remodeling events during lag phase ( Figure 1 , steps 2–4 ) . Thus , Hsp26 antagonizes events that occur after prion recognition elements have initially formed intermolecular contacts , whereas Hsp42 prevents initial formation of these contacts . These data suggest that Hsp26 and Hsp42 synergize to directly antagonize Sup35 prionogenesis prior to intermolecular contact formation by prion recognition elements . These findings suggested that Hsp26 and Hsp42 might antagonize [PSI+] induction in vivo . Prion nucleation involves protein–protein interactions [40] , [43] . Thus , [PSI+] induction frequency is very low unless Sup35 or NM is overexpressed [70] , [71] . Indeed , when Sup35 was transiently overexpressed , [PSI+] induction increased from barely detectable levels ( <1 in 1 , 000 ) to ∼20% of cells ( Figure 4A ) . [PSI+] induction was modestly increased by ∼1 . 3-fold in Δhsp26 cells and ∼1 . 6-fold in Δhsp42 cells ( Figure 4A ) . In a double deletion Δhsp26Δhsp42 strain , [PSI+] induction was increased by ∼2 . 1-fold ( Figure 4A ) . Importantly , immunoblots revealed that neither Hsp104 nor Hsp70 expression were affected by the sHsp deletion ( Figure 4B ) . Thus , increased [PSI+] induction observed in the sHsp deletion strains ( Figure 4A ) is likely to be a direct effect of reduced sHsp activity . Elevated expression of Hsp26 or Hsp42 antagonized [PSI+] induction by Sup35 overexpression ( Figure 4C ) . The combination of Hsp26 and Hsp42 was more effective than either sHsp alone , and almost as effective as the protein disaggregase Hsp104 ( Figure 4C ) . The expression of Hsp104 and Hsp70 was not altered by sHsp overexpression ( Figure 4D ) , indicating that these effects are likely due to sHsp activity . Accordingly , Hsp26 and Hsp42 overexpression prevented the formation of NM-YFP foci , a reporter of [PSI+] induction , when NM-YFP was overexpressed ( Figure 4E ) [72] . Here too , the combination of Hsp26 and Hsp42 yielded the greatest inhibition ( Figure 4E ) . Importantly , overexpression of Hsp26 , Hsp42 , or both has no effect on the [RNQ+] prion [33] , which is critical for [PSI+] induction by Sup35 overexpression [73] . Taken together , these data suggest that Hsp26 and Hsp42 work together to directly antagonize [PSI+] induction . The induction of [PSI+] by Sup35 overexpression depends on the presence of another prion [RNQ+] , which is comprised of infectious Rnq1 amyloid [73] . Rnq1 prions are proposed to template the initial formation of Sup35 prions in vivo , and this activity has been reconstituted in vitro [74] . Thus , we tested whether Hsp26 and Hsp42 inhibited NM fibrillization cross-seeded by Rnq1 fibers in vitro . Hsp26 or Hsp42 inhibited NM assembly that was cross-seeded by Rnq1 fibers , whereas a control protein ( BSA ) had no effect ( Figure 5 ) . The ability of Hsp42 to inhibit this seeding reaction was unexpected and might suggest that Rnq1 fibers accelerate events in the lag phase of NM assembly ( Figure 1 , steps 1–4 ) , rather than acting as a direct template for NM fibrillization ( Figure 1 , step 5 ) [74] . Hsp26 and Hsp42 might interact directly with Rnq1 fibers to prevent interactions with NM . Alternatively , interactions between Hsp26 or Hsp42 and NM might prevent interactions with Rnq1 that drive cross-seeding . Importantly , the combination of Hsp26 and Hsp42 was more potent than either sHsp alone ( Figure 5 ) , suggesting that the two sHsps work together to prevent cross-seeding . Thus , Hsp26 and Hsp42 antagonize de novo formation of Sup35 prions in vitro and in vivo . The ability of Hsp26 to inhibit the assembly of BMB-crosslinked NM ( Figure 3F ) suggested that Hsp26 might inhibit fibrillization of NM seeded by preformed NM fibers . Indeed , Hsp26 potently inhibited ( IC50∼1 . 9 µM ) NM fibrillization seeded by preformed NM fibers ( 5% wt/wt ) ( Figure 6A , B ) . Hsp42 also inhibited seeded assembly ( Figure 6A , B ) . However , Hsp42 was less effective ( IC50>24 µM ) and inhibition was only observed at high concentrations ( Figure 6A , B ) . In contrast to spontaneous assembly ( Figure 2A , B ) , the combination of Hsp26 and Hsp42 did not yield stronger inhibition ( Figure 6A , B ) . Thus , Hsp26 and Hsp42 did not synergize to prevent seeded assembly in vitro ( Figure 6A , B ) , which is consistent with results obtained with BMB-crosslinked NM ( Figure 3F ) . Very similar results were obtained using full-length Sup35 ( Figure 6C ) . The strong inhibition of Sup35 prion formation by Hsp26 at 25°C ( Figures 2A , 6A ) was unexpected because exposure to elevated temperature is required for Hsp26 to bind unfolded polypeptides and prevent their aggregation [27] , [31] , [32] . Indeed , as expected , pretreatment of Hsp26 for 10 min at 45°C increased the ability of Hsp26 to suppress aggregation of glutamate dehydrogenase ( GDH ) ( Figure 6D ) . By contrast , the same pretreatment reduced the ability of Hsp26 to inhibit spontaneous NM assembly ( Figure 6D ) . Thus , at elevated temperatures Hsp26 might switch from inhibiting prion assembly to suppressing aggregation of denatured substrates . As with spontaneous NM assembly ( Figure 6D ) , pretreatment of Hsp26 at 45°C reduced its ability ( IC50>24 µM ) to inhibit seeded fibrillization ( Figure 6A–C ) . Thus , conditions that enable Hsp26 to prevent the aggregation of environmentally denatured proteins reduce the ability of Hsp26 to antagonize Sup35 prionogenesis . These data suggest that Hsp26 uses a distinct mechanism to antagonize prion formation . Prion recognition elements termed the “Head” ( residues ∼21–38 ) and “Tail” ( residues ∼91–106 ) in NM fibers formed at 25°C make homotypic intermolecular contacts such that fibers are constructed by “Head-to-Head” and “Tail-to-Tail” contacts ( Figure 1 ) [40] , [41] . We asked whether Hsp26 might inhibit either “Head-to-Head” or “Tail-to-Tail” intermolecular contact formation or both to inhibit seeded fibrillization ( Figure 1 , step 5 ) . Thus , we employed six different individual NM single cysteine mutants labeled with pyrene in either the head ( G25C , G31C , or Q38C ) or the tail region ( G86C , G96C , or Y106C ) . Upon intermolecular contact formation and fibrillization , pyrene molecules form excimers ( excited-state dimers ) that produce a strong red shift in fluorescence [40] , [41] . Hsp42 only partially inhibited seeded “Head-to-Head” or “Tail-to-Tail” contact formation ( Figure 6E ) . By contrast , Hsp26 strongly inhibited seeded “Head-to-Head” or “Tail-to-Tail” contact formation , and a 45°C pretreatment reduced this inhibitory activity ( Figure 6E ) . To determine whether Hsp26 inhibited seeded assembly by interacting with NM fibers or monomers or both , we pretreated NM fibers with Hsp26 or Hsp42 . We then recovered the fibers by centrifugation , washed , and resuspended the material to use as seed . Hsp26 or Hsp42 did not disassemble NM fibers in this timeframe and equal amounts of NM were recovered for each condition . NM fibers pretreated in this way with BSA or Hsp42 could still seed , whereas those pretreated with Hsp26 were unable to seed ( Figure 7A ) . Very similar results were obtained using pyrene-labeled NM ( Figure 7B ) . Both Hsp42 and Hsp26 were recovered in the pellet with NM fibers ( Figure 7C ) and thus were present in the seeded assembly reaction . The residual concentration was estimated to be ∼50 nM for Hsp26 and ∼20 nM for Hsp42 . These concentrations are not sufficient to cause significant inhibition of seeded assembly without pretreatment of fibers ( Figure 6A ) . Thus , Hsp42 binds to NM fibers in a manner that does not affect seeded assembly . By contrast , Hsp26 binds to NM fibers and occludes prion recognition elements to inhibit seeded assembly . These results do not exclude that Hsp26 or Hsp42 might also interact with monomeric NM to inhibit prion formation . Thus , we pretreated soluble NM-his with Hsp26 or Hsp42 ( or BSA ) . We then recovered the NM-his using Ni-NTA superflow and used it as substrate for preformed NM fibers . Hsp26 did not bind NM-his monomers ( Figure 7D ) , in contrast to the interaction between Hsp26 and NM fibers ( Figure 7C ) . We detected minimal binding of Hsp42 to NM-his ( Figure 7D ) , but again this interaction was not as pronounced as the interaction between Hsp42 and NM fibers ( Figure 7C ) . The NM-his recovered under each of these conditions was readily converted to amyloid by NM fibers ( Figure 7E ) . These data indicate that the direct interaction between Hsp26 and NM fibers is critical for the inhibition of seeded assembly . Next , we tested whether the sHsps could inhibit seeding by ex vivo Sup35 prions . Several proteins bind Sup35 prions in situ , most notably Ssa1 [75] , which might affect how Hsp42 or Hsp26 influence seeding . Hence , we isolated Sup35 prions from [PSI+] cells [75] and used them to seed the assembly of full-length Sup35 in the absence or presence of Hsp26 or Hsp42 . Ex vivo Sup35 prions effectively seeded Sup35 fibrillization and Hsp26 potently inhibited this process ( Figure 7F ) . Hsp42 also inhibited seeding by ex vivo Sup35 prions ( Figure 7F ) but only at high Hsp42 concentrations and was comparable to the inhibition observed with NM ( Figure 6A ) . The inhibition of seeded assembly by Hsp26 ( Figures 6A–C , E , 7F ) suggested that Hsp26 might interfere with [PSI+] propagation in vivo . Indeed , elevated expression of Hsp26 effectively cured [PSI+] , whereas [PSI+] curing was not detected in the vector control ( Figure 8 ) . Surprisingly , even though Hsp42 did not effectively inhibit seeded assembly ( Figures 6A–C , E , 7F ) , elevated expression of Hsp42 also effectively cured [PSI+] ( Figure 8 ) . Moreover , elevated expression of both Hsp26 and Hsp42 cured [PSI+] just as effectively as overexpression of Hsp104 ( Figure 8 ) , which is a potent method of [PSI+] curing [76] . Thus , Hsp26 and Hsp42 work together to eliminate Sup35 prions in vivo . The magnitude of the [PSI+] curing effect by Hsp42 overexpression ( Figure 8 ) was unexpected because it is much less effective in inhibiting seeded assembly than Hsp26 ( Figures 6A–C , E , 7F ) . We reasoned that Hsp42 might collaborate with other chaperones to antagonize seeded Sup35 assembly . Ssa1 , an Hsp70 chaperone , can collaborate with Hsp40 partners ( Sis1 or Ydj1 ) to inhibit seeded fibrillization of NM [55] . This inhibition is due to an interaction between the fiber and Hsp70 and Hsp40 because neither Ssa1:Ydj1 nor Ssa1:Sis1 interact with NM monomers directly [55] . We tested whether the seeding activity of NM fibers that had been pretreated with Hsp42 were more sensitive to inhibition by increasing concentrations of Ssa1:Sis1 or Ssa1:Ydj1 . We first confirmed that adding Ssa1 , Ydj1 , or Sis1 alone had no effect on seeding by NM fibers or seeding by NM fibers pretreated with Hsp42 ( Figure 9A–C ) . Next , we titrated Ssa1:Sis1 or Ssa1:Ydj1 into seeded assembly while maintaining the Ssa1∶Hsp40 ratio at 1∶1 . As expected , both Ssa1:Sis1 and Ssa1:Ydj1 inhibited seeded assembly of buffer-treated NM fibers with an IC50 of ∼1 . 1 µM and ∼1 . 6 µM , respectively ( Figure 9D , E ) . However , Hsp42-treated NM fibers were more susceptible to inhibition by Ssa1:Sis1 and Ssa1:Ydj1 ( Figure 9D , E ) . The IC50 was reduced about 5-fold to ∼0 . 21 µM for Ssa1:Sis1 and about 3-fold to ∼0 . 49 µM for Ssa1:Ydj1 . These marked decreases in IC50 suggest that Hsp42 binds NM fibers and promotes interactions between the prion and Ssa1:Sis1 or Ssa1:Ydj1 that preclude seeded assembly . Thus , Hsp42 may direct Hsp70 and Hsp40 to fiber ends to prevent assembly . These results were corroborated in vivo . Two titers of synthetic NM prions were transformed into [psi−] cells overexpressing the indicated combination of Hsp26 , Hsp42 , Ydj1 , or Sis1 ( Figure 9F ) . Importantly , [PSI+] induction by NM fibers was reduced by the overexpression of either sHsp or either Hsp40 especially at lower prion titers ( Figure 9F ) . The greatest inhibition was observed when Hsp26 and Hsp42 were combined ( Figure 9F ) . At higher prion titers , the combination of sHsp and Hsp40 was more potent than the sHsp alone ( Figure 9F ) . Specifically , the combination of Hsp26 plus Ydj1 was more effective in preventing infection than Hsp26 alone ( two-tailed Student's t test; p = 0 . 005 ) , as was Hsp26 plus Sis1 ( two tailed Student's t test; p = 0 . 0012 ) . Similarly , the combination of Hsp42 plus Ydj1 was more effective in preventing infection than Hsp42 alone ( two-tailed Student's t test; p = 0 . 006 ) , as was Hsp42 plus Sis1 ( two-tailed Student's t test; p = 0 . 0025 ) . These data suggest that Hsp42 can collaborate with Hsp70 and Hsp40 to prevent seeding by NM prions in vivo . They also suggest that Hsp26 can prevent seeding by preformed NM prions in vivo . Does Hsp26 or Hsp42 binding alter prion structure ? Although Hsp26 and Hsp42 did not disassemble NM fibers after a brief incubation , they both induced a slight decrease in the thermal stability of NM fibers in 1 . 6% SDS . The melting temperature was slightly reduced from 78±1°C to 71±1°C for Hsp26 and to 72±1°C for Hsp42 ( Figure 10A ) , whereas Ssa1 plus Sis1 had no effect ( Figure 10A ) . Thus , sHsp binding to NM fibers might weaken or subtly alter the intermolecular contacts between NM protomers , and shift the monomer-fiber equilibrium in favor of dissociation . In this way , sHsps might render amyloid forms more susceptible to dissolution by protein disaggregases . To monitor intermolecular contacts directly , we independently assembled 17 individual single cysteine NM mutants labeled with pyrene [40] , [41] . We then determined how Hsp26 and Hsp42 binding affected excimer fluorescence at these positions . Excimer fluorescence detects intermolecular contact integrity and the proximity of residues within different protomers of the assembled prion [40] , [41] . Ssa1 and Sis1 had no effect on excimer fluorescence ( Figure 10B ) , whereas Hsp26 and Hsp42 caused subtle but significant alterations in excimer fluorescence at almost every position tested ( Figure 10B ) . Excimer fluorescence in the Head ( amino acids 21–38 ) and Tail ( amino acids 86–106 ) regions was slightly reduced ( Figure 10B ) . The most drastic alteration was observed in the extreme N-terminal positions 2 and 7 , where excimer fluorescence was reduced ∼2-fold ( Figure 10B ) . Thus , Hsp26 or Hsp42 binding alters prion architecture in a way that weakens intermolecular contacts and forces residues that are N-terminal to the Head contact ( amino acids 21–38 ) further apart . This sHsp-induced weakening of prion architecture may promote dissolution by prion disaggregases such as Hsp104 . The destabilization of Sup35 prions by Hsp26 or Hsp42 binding could be exploited by Hsp104 to rapidly disaggregate prions . Thus , we assembled NM fibers and incubated them with increasing concentrations of Hsp104 in the presence or absence of Hsp26 or Hsp42 . In the absence of other components Hsp104 effectively disassembled NM fibers , with an EC50 of ∼0 . 15 µM ( Figure 11A ) . Remarkably , Hsp42 enabled Hsp104 to disassemble NM fibers at concentrations where it would usually be inactive ( 0 . 01–0 . 1 µM; Figure 11A ) . Hsp42 reduced the Hsp104 EC50 to ∼0 . 075 µM . By contrast , Hsp26 inhibited Hsp104 activity ( Figure 11A ) . Addition of Ssa1:Sis1 had little effect on Hsp104 activity in the absence of sHsps ( the Hsp104 EC50 was ∼0 . 14 µM; Figure 11B ) . However , Ssa1:Sis1 enhanced Hsp104 activity in the presence of Hsp42 and reduced the Hsp104 EC50 to ∼0 . 05 µM ( Figure 11B ) . Ssa1:Sis1 relieved the inhibition of Hsp104 by Hsp26 and potentiated Hsp104 remodeling activity , reducing the EC50 to ∼0 . 06 µM . We confirmed that prions had been protected or eliminated by transforming products into [psi−] cells ( Figure 11C ) . These data suggest that sHsps enhance the ability of Hsp104 to eliminate Sup35 prions . Next , we tested whether overexpression of Hsp26 or Hsp42 increased [PSI+] curing by elevated Hsp104 concentrations . Consistent with our in vitro observations , Hsp26 and Hsp42 synergized with Hsp104 to promote [PSI+] curing ( Figure 11D ) . These data reinforce our in vitro observations that sHsps potentiate Hsp104 activity against Sup35 prions . Next , we tested whether the sHsps potentiated Hsp104 activity against α-syn amyloid , which is connected to Parkinson's disease [1] , [77] , [78] , and polyglutamine amyloid , which is connected to Huntington's disease [1] , [79] . α-Syn and polyglutamine ( Q62 ) fibers were assembled and preincubated with either Hsp26 or Hsp42 . The indicated combination of Hsp104 , Ssa1 , and Sis1 was then added . sHsps promoted rapid disassembly of α-syn and polyglutamine fibers by Hsp104 ( Figure 11E ) . Indeed , preincubation with Hsp26 enabled Hsp104 to catalyze more α-syn fiber disassembly ( two-tailed Student's t test; p = 0 . 0031 ) and more Q62 fiber disassembly ( two-tailed Student's t test; p = 0 . 0037 ) than Hsp104 alone . For these amyloid substrates , Hsp26 alone did not antagonize Hsp104 activity ( Figure 11E ) . Preincubation of fibers with Hsp42 also enabled Hsp104 to catalyze more α-syn fiber disassembly ( two-tailed Student's t test; p = 0 . 0209 ) and more Q62 fiber disassembly ( two-tailed Student's t test; p = 0 . 0019 ) than Hsp104 alone . Optimal disaggregation was achieved with sHsp plus Hsp104 , Ssa1 , and Sis1 ( Figure 11E ) . Thus , sHsps potentiate Hsp104 activity against disease-associated amyloid . Next , we asked whether Hsp42 could function like Hsp26 to promote the disaggregation of disordered aggregates by Hsp104 [33] . In contrast to Hsp26 , Hsp42 did not promote the disaggregation of heat-denatured luciferase aggregates by Hsp104 , Hsp70 , and Hsp40 ( Figure 11F ) . These data help explain why Hsp26 , but not Hsp42 , assists Hsp104 in promoting luciferase disaggregation and thermotolerance in vivo [33] . Thus , Hsp42 selectively promotes the disassembly of amyloid conformers by Hsp104 , whereas Hsp26 promotes Hsp104-catalyzed disaggregation of both amyloid and non-amyloid aggregates . Metazoa lack an Hsp104 orthologue and how amyloid might be disaggregated in animal systems remains unknown [60] . In general , monomers at fiber ends are more likely to be susceptible to disaggregation because they are only restrained by one intermolecular contact ( e . g . , Head or Tail , Figure 1 ) . Indeed , fiber ends are dynamic and monomers slowly dissociate within a biologically relevant timeframe ( days ) and rapidly reassociate in a process termed molecular recycling ( Figure 12A ) [62]–[64] , [67] , [68] . Dissociation is the rate-limiting step in recycling and reassociation is rapid . A homogeneous population of fibers formed by a SH3 domain with an average length of 100 nm , recycle ∼50% of monomers within 2 to 20 d [62] . Thus , agents that accelerate monomer dissociation or prevent monomer reassociation or both could drive fiber depolymerization on a timescale similar to that of molecular recycling . For example , Hsp26 and Hsp70:Hsp40 pairs ( e . g . , Ssa1:Sis1 ) prevent seeded assembly ( Figures 6A–C , 9D , 9E ) and might inhibit monomer reassociation events ( Figure 12A ) . We were particularly interested in Hsp110 in this context . Hsp110 can synergize with Hsp70 and Hsp40 to extract and refold proteins from denatured aggregates [61] . Thus , the combination of Hsp110 , Hsp70 , and Hsp40 might even accelerate dissociation of monomers from fiber ends . We assembled and sonicated NM fibers to generate a uniform population of short fibers [43] . NM fibers were stable for 28 d alone ( black filled squares , Figure 12B , C ) or in the presence of molecular chaperones that alone do not affect seeded assembly , including Hsp42 ( blue filled triangles , Figure 12B , C ) , Sis1 ( black open squares , Figure 12B , C ) , Ydj1 ( red open circles , Figure 12B , C ) , Ssa1 ( black open triangles , Figure 12B , C ) , or Sse1 ( blue filled squares , Figure 12B , C ) . In remarkable contrast , Hsp26 ( red filled circles , Figure 12B , C ) , Ssa1:Sis1 ( blue open triangles , Figure 12B , C ) , and Ssa1:Ydj1 ( black open circles , Figure 12B ) , which inhibit seeded assembly ( 6A , 6B , 9D , 9E ) , slowly disassembled preformed NM fibers over a time period of 28 d ( Figure 12B , C ) . Consistent with their efficacy to inhibit seeded assembly , Ssa1:Sis1 ( blue open triangles , Figure 12B , C ) was more effective than Ssa1:Ydj1 ( black open circles , Figure 12B , C ) or Hsp26 ( red filled circles , Figure 12B , C ) . Notably , the combination of Sse1 ( Hsp110 ) , Ssa1 and Sis1 ( green filled squares , Figure 12B , C ) , or Sse1 , Ssa1 , and Ydj1 ( cyan filled triangles , Figure 12B , C ) yielded more rapid disassembly , whereas Sse1 combined with Ssa1 ( red filled squares , Figure 12B , C ) , Sis1 ( black filled circles , Figure 12B , C ) , or Ydj1 ( red filled triangles , Figure 12B , C ) had no effect . Electron microscopy ( Figure 12D ) and prion transformation ( Figure 12E ) confirmed that prions had been eliminated by Sse1:Ssa1:Sis1 , Sse1:Ssa1:Ydj1 , Ssa1:Sis1 , Ssa1:Ydj1 , and Hsp26 , but not by Sse1:Ssa1 , Sse1:Sis1 , Sse1:Ydj1 , Sse1 , Ssa1 , Ydj1 , Sis1 , or Hsp42 . Disassembly was contingent on the number of fiber ends , as unsonicated fibers were more refractory to disassembly ( Figure 12F ) . Indeed , we confirmed that disassembly entailed depolymerization from fiber ends using “capped” fibers . Thus , NM fibers comprised of untagged NM were resuspended in buffer containing high concentrations of C-terminally his-tagged NM . This procedure allowed NM fibers to be rapidly elongated creating NM fibers with NM-his “caps” ( Figure 13A ) . We established conditions where ∼50% of the total NM in fibers was his-tagged ( Figure 13A ) . If NM fibers with NM-his “caps” were treated for extended periods with Sse1:Ssa1:Sis1 , Sse1:Ssa1:Ydj1 , Ssa1:Sis1 , Ssa1:Ydj1 , or Hsp26 , then only NM-his was released into the soluble fraction ( Figure 13B , C ) . Conversely , if NM-his fibers were capped with untagged NM ( Figure 13A ) , then only untagged NM was released into the soluble fraction ( Figure 13D , E ) . When capped fibers were sonicated prior to incubation to randomize the form of NM at fiber ends , approximately equal amounts of NM and NM-his were released ( Figure 13F , G ) . Taken together , these data suggest that Sse1:Ssa1:Sis1 , Sse1:Ssa1:Ydj1 , Ssa1:Sis1 , Ssa1:Ydj1 , and Hsp26 slowly depolymerize NM fibers from their ends . The most effective depolymerization is promoted by the combination of Hsp110 , Hsp70 , and Hsp40 . Hsp110 , Hsp70 , and Hsp40 might exploit the destabilization of amyloid by sHsp binding ( Figure 10A , B ) to promote amyloid depolymerization . Thus , NM fibers were pretreated with Hsp26 or Hsp42 prior to addition of Sse1:Ssa1:Sis1 or Sse1:Ssa1:Ydj1 . Pretreatment with Hsp26 or Hsp42 , which subtly alters NM fiber structure and stability ( Figure 10A , B ) , facilitated more rapid depolymerization by Sse1:Ssa1:Sis1 ( Figure 13H ) or Sse1:Ssa1:Ydj1 ( Figure 13I ) . For Sse1:Ssa1:Sis1 , the D1/2 ( the 50% disassembly time ) was reduced from ∼12 . 8 d to ∼7 . 4 d by Hsp26 and to ∼7 . 7 d by Hsp42 . For Sse1:Ssa1:Ydj1 , D1/2 was reduced from ∼16 . 1 d to ∼10 . 5 d by Hsp26 and ∼10 . 9 d by Hsp42 . Thus , sHsps render amyloid forms more susceptible to depolymerization by the Hsp110 , Hsp70 , and Hsp40 disaggregase machinery . Our newly discovered amyloid-depolymerase activity was not restricted to yeast chaperones and yeast prions . Indeed , human Hsp70 ( Hsc70 ) and Hsp40 ( Hdj1 ) slowly disassemble preformed α-syn fibers ( Figure 14A , B ) . This activity was stimulated by addition of human Hsp110 ( Apg-2 ) or the human sHsp , HspB5 ( Figure 14A , B ) . HspB5 potentiated α-syn fiber disassembly by Apg-2 , Hsc70 and Hdj1 ( Figure 14A , B ) . Indeed , HspB5 reduced the D1/2 to ∼14 d for Apg-2 , Hsc70 , and Hdj1 . We confirmed that the combination of Apg-2 , Hsc70 , Hdj1 , and HspB5 depolymerized fibers from their ends by employing α-syn fibers capped with his-α-syn . Thus , during the initial disassembly phase of unsonicated fibers , Apg-2 , Hsc70 , Hdj1 , and HspB5 liberated only his-α-syn into the soluble fraction ( Figure 14C ) , which indicates that disassembly proceeds via depolymerization . These data suggest that the human proteostasis network , like its yeast counterpart , is equipped with an amyloid-depolymerase modality . Although depolymerization is relatively slow , it occurs on a biologically relevant timescale , especially considering the lifespan of neurons in the human brain . Could Hsp104 interface with the human sHsp , HspB5 , and the Hsp110 , Hsp70 , and Hsp40 disaggregase machinery ? Remarkably , the combination of Apg-2 , Hsc70 , Hdj1 , and HspB5 enabled even more effective and rapid disaggregation of α-syn fibers by Hsp104 than with just Apg-2 , Hsc70 , and Hdj1 ( Figure 14D ) [61] . The ability of Hsp104 to interface effectively with the human disaggregase machinery and enable effective clearance of α-syn amyloid suggests that Hsp104 might be developed further to target pathological α-syn conformers . Collectively , our studies suggest that sHsps are ubiquitous potentiators of amyloid disassembly by the proteostasis network .
We have established that the sHsps from yeast , Hsp26 and Hsp42 , exert tight control over the formation of beneficial Sup35 prions . Both sHsps exerted a strong and direct inhibitory effect on Sup35 prion formation at substoichiometric concentrations . These results were surprising because sHsps commonly bind 1 substrate per ∼2–3 sHsp monomers [24] , [27] . Thus , the strong inhibitory effect at substoichiometric concentrations indicates that the sHsps might inhibit a rare or transient NM conformer that is critical for prion formation . Surprisingly , our results suggested that this conformer was different for each sHsp . Hsp42 targeted molten Sup35 oligomers , whereas Hsp26 targeted the self-templating ends of newly assembled prions ( Figure 1 ) . Although little was known about Hsp42 , it had been suggested to work by a mechanism similar to Hsp26 to inhibit protein aggregation [28] . Surprisingly , however , Hsp42 inhibited spontaneous Sup35 prionogenesis by a distinct mechanism to Hsp26 . Hsp42 specifically antagonized events in the lag phase of prion formation . Hsp42 prevented and reversed the maturation of Sup35 oligomers into prion-nucleating species ( Figure 1 , steps 2 and 3 ) . By contrast , Hsp26 bound to newly formed prions and inhibited their seeding activity ( Figure 1 , step 5 ) . These two activities synergized to inhibit de novo Sup35 prionogenesis in vitro and in vivo . To the best of our knowledge , this is the first example of two distinct sHsps working together in a synergistic manner to prevent prion formation . The mechanistic differences between Hsp26 and Hsp42 are likely conferred by their divergent N-terminal domains . Hsp42 has an extended N-terminal domain [80] , which displays no homology to other sHsps . The extended N-terminal domain of Hsp42 might enable insertion into molten Sup35 oligomers in a way that precludes prion formation . Hsp26 chaperone activity is usually activated at heat shock temperatures [27] , [29] , [31] , [32] . Unexpectedly , we found that pretreatment of Hsp26 at high temperature reduced its ability to inhibit Sup35 prionogenesis , while simultaneously enhancing its ability to prevent aggregation of a chemically denatured substrate . Our result thus reveals a fundamental difference in how Hsp26 antagonizes the aggregation of a denatured protein ( GDH ) and a yeast prion ( Sup35 ) . Hsp26 conformations that are ineffective against heat-denatured substrates are effective against Sup35 prions and vice versa . This difference might reflect distinct driving forces of GDH and Sup35 aggregation . GDH aggregation likely involves inappropriately exposed hydrophobic surfaces , whereas NM fibrillization likely involves polar interactions or backbone interactions or both because polar residues outweigh hydrophobic residues by ∼16 to 1 . At physiological temperatures , Hsp26 may be primed to inhibit prion formation , but at elevated temperatures , Hsp26 loses this ability and switches to inhibiting the aggregation of heat-denatured proteins . This switch in Hsp26 activity likely contributes to the increased levels of [PSI+] induction at elevated temperatures [11] . Both Hsp26 and Hsp42 bind to preformed Sup35 fibers , but only Hsp26 binding inhibited seeding activity . However , Hsp42 increased the ability of Hsp70 and Hsp40 ( Ssa1:Sis1 or Ssa1:Ydj1 ) to inhibit seeded assembly , potentially by recruiting Hsp70 to fiber ends . These data help explain why overexpression of Hsp26 or Hsp42 cures cells of [PSI+] . Unexpectedly , Hsp26 or Hsp42 binding destabilized Sup35 prions . Hsp26 and Hsp42 binding reduced excimer fluorescence at intermolecular contact regions . The most marked effect was at residues N-terminal to the Head region , which appeared to be forced further apart in adjacent protomers by Hsp26 or Hsp42 binding . These data suggest that Hsp26 or Hsp42 harness binding energy to alter prion architecture . Notably , these sHsp-induced alterations facilitated the disaggregation of Sup35 prions by Hsp104 . Pretreatment of Sup35 prions with Hsp42 rendered them more susceptible to rapid disassembly by Hsp104 . Curiously , Hsp26 alone inhibited Hsp104 . However , Ssa1 and Sis1 alleviated this inhibition and promoted more effective prion disassembly . These findings might suggest that the mechanism of Sup35 prion disassembly by Hsp104 is different in the presence of Hsp26 versus Hsp42 . Further experiments are needed to explore this possibility . Importantly , Hsp26 and Hsp42 promoted elimination of Sup35 prions by Hsp104 in vivo , as overexpression of Hsp26 or Hsp42 increased [PSI+] curing by elevated Hsp104 concentration . Hsp26 and Hsp42 also promoted rapid Hsp104-catalyzed disassembly of α-syn fibers that are connected with PD . We further demonstrated that Hsp104 directly disassembles polyglutamine fibers that are connected with HD . Hsp26 or Hsp42 boosted this activity and disaggregation was maximal in the presence of Hsp104 , an sHsp , Ssa1 , and Sis1 . We have established an important dichotomy between how Hsp26 and Hsp42 collaborate with Hsp104 . Hsp26 promotes the disaggregation of both amyloid and non-amyloid substrates by Hsp104 in the presence of Hsp70 and Hsp40 . By contrast , Hsp42 selectively promotes the disassembly of amyloid substrates by Hsp104 . Thus , Hsp42 is an amyloid-specific adaptor for Hsp104 . In yeast , Hsp42 appears to preferentially localize to peripheral inclusions [34] that might harbor amyloid conformers that can be solubilized by Hsp104 [53] , [56] , [81] , [82] . We have shown that in the absence of Hsp104 , the Hsp110 , Hsp70 , and Hsp40 disaggregase system [61] can slowly depolymerize amyloid fibers . Depolymerization was a slow process that required many days to complete and appeared to occur on a timescale similar to molecular recycling within amyloid fibers [62] , [67] . Thus , the proteostasis network might exploit this process to slowly eradicate amyloid by either accelerating monomer dissociation from fiber ends ( i . e . , increasing koff , Figure 12A ) or inhibiting monomer reassociation with fiber ends ( i . e . , decreasing kon , Figure 12A ) or both . Consistent with the possibility of inhibiting monomer reassociation ( decreasing kon , Figure 12A ) , agents that inhibit seeded polymerization of Sup35 prions ( e . g . , Hsp26 or Ssa1:Sis1 ) slowly depolymerized them over the course of many days . The relatively low number of Hsp26 monomers per molecule of substrate required for Hsp26 disaggregation activity might indicate that Hsp26 acts selectively at fiber ends . The combination of Sse1 , Ssa1 , and Sis1 yielded the most effective depolymerization . Given the capability of this disaggregase system to extract and refold proteins from large denatured aggregates [61] , we suggest that Hsp110 , Hsp70 , and Hsp40 might also accelerate monomer dissociation events ( increasing koff , Figure 12A ) . Importantly , destabilization of NM fibers by Hsp26 or Hsp42 accelerated prion depolymerization by Hsp110 , Hsp70 , and Hsp40 . Intriguingly , this activity is not confined to yeast but is conserved to humans . Thus , the human sHsp , HspB5 , accelerated the depolymerization of α-syn amyloid ( which is connected with PD ) by human Hsp110 ( Apg-2 ) , Hsp70 ( Hsc70 ) , and Hsp40 ( Hdj1 ) . Collectively , these data suggest that in metazoa , which lack an Hsp104 homologue , Hsp110 , Hsp70 , and Hsp40 can slowly eliminate amyloid forms by specifically exploiting the molecular recycling process ( Figure 12A ) . Although amyloid depolymerization is slow and requires many days to complete , it occurs on a biologically relevant timescale , especially considering the lifespan of humans . Indeed , a massive therapeutic advance will likely come with the ability to stimulate the proteostasis network to dissolve α-syn fibers in a few days in Parkinson's patients . Our data provide proof of principle that this may indeed be possible and that pure , individual components can drive this process . Although released monomers could have a chance to reassemble into toxic oligomers , we suspect that components of the proteostasis network would prevent toxic oligomer formation . Shutting down expression of an amyloidogenic protein enables mammalian cells to slowly clear protein aggregates [83] , [84] . Our findings suggest that sHsps and the Hsp110 , Hsp70 , and Hsp40 disaggregase system might play a crucial role in this clearance . Moreover , they suggest that potential RNA interference therapies to deplete the aggregating protein should be combined with targeted upregulation of sHsps and the Hsp110 , Hsp70 , and Hsp40 disaggregase system to promote clearance of existing aberrant conformers . Another way to accelerate the disaggregation of α-syn fibers is to introduce Hsp104 [54] , [61] . Indeed , the combination of Hsp104 with Apg-2 , Hsc70 , Hdj1 , and HspB5 disaggregated α-syn fibers most effectively and rapidly . Importantly , Hsp104 expression counteracts neurodegeneration associated with α-syn misfolding and polyglutamine misfolding in rodents [54] , [59] , [85] , [86] . Thus , our findings suggest that boosting sHsp levels or activity might provide a powerful strategy to facilitate clearance of deleterious amyloid by either the endogenous human Hsp110 , Hsp70 , and Hsp40 disaggregase machinery [61] or by Hsp104 in targeted therapeutic strategies [54] , [59] , [85]–[87] .
Hsp26 [27] , Hsp42 [28] , Ssa1 , Sis1 , Ydj1 , NM [48] , NM-his [51] , Sup35 [49] , Hsp104 [88] , Apg-2 [89] , Sse1 [90] , Rnq1 [74] , polyglutamine ( GST-Q62 ) [91] , and α-syn [92] were purified as described . Hsc70 and Hdj1 were from Enzo Life Sciences . HspB5 was from ProSpec . BSA and firefly luciferase were from Sigma and GDH was from Roche . Single cysteine NM mutants were labeled with pyrene-maleimide or acrylodan ( Invitrogen ) or crosslinked with BMB ( Pierce ) under denaturing conditions as described [40] . Throughout the manuscript , protein concentrations refer to the monomer , with the exception of Hsp104 , where it refers to the hexamer . The plasmids used for overexpression of Hsp26 , Hsp42 , Hsp104 , Ydj1 , Sis1 , Sup35 , and NM-YFP were as described [33] , [41] , [90] , [93] . The aggregation of denatured GDH was monitored by turbidity at 395 nm [31] . In some experiments , Hsp26 was thermally activated by incubation at 45°C for 10 min prior to addition to aggregation assays . NM ( 5 µM ) fibrillization was conducted in assembly buffer ( AB ) ( 40 mM HEPES-KOH , pH 7 . 4 , 150 mM KCl , 20 mM MgCl2 , 1 mM DTT ) . For Sup35 ( 5 µM ) fibrillization , AB was supplemented with 1 mM GTP and 10% glycerol . Unseeded reactions were agitated at 1 , 400 r . p . m . ( for NM ) or 700 r . p . m . ( for Sup35 ) in a thermomixer ( Eppendorf ) for the indicated time at 25°C . Seeded assembly was unagitated and performed for the indicated time at 25°C . The amount of seed is indicated as % ( wt/wt ) . In some experiments ( Figure 7A ) , NM fibers ( 5 µM NM monomer ) were pretreated for 60 min at 25°C without or with BSA , Hsp26 , or Hsp42 ( 10 µM ) . NM fibers were then recovered by centrifugation at 16 , 000 g , gently washed ( without resuspending the pellet ) , and then resuspended in AB . Ex vivo Sup35 prions for seeding experiments were isolated as described [75] and the amount of Sup35 in the isolated fraction was determined by immunoblot in comparison to known quantities of pure Sup35 . Rnq1 fibers were assembled as described [74] . For assembly reactions containing Ssa1:Sis1 or Ssa1:Ydj1 , ATP was added ( 5 mM ) plus an ATP regeneration system comprising creatine phosphate ( 40 mM ) and creatine kinase ( 0 . 5 µM ) . The extent of fiber assembly was determined by ThT fluorescence , electron microscopy , or by the amount of SDS-resistant NM as described [49] , [55] . The oligomer-specific A11 antibody was used to detect amyloidogenic NM oligomers by ELISA as described [69] . Importantly , Hsp26 and Hsp42 did not cross-react with A11 . For NM disassembly reactions , NM ( 5 µM ) was assembled with agitation for 6 h in AB as described above . Wild-type or A53T α-syn fibers were assembled as described [54] . Polyglutamine ( GST-Q62 ) ( 10 µM ) was incubated for 1 h at 25°C with thrombin in AB to separate GST from Q62 , and then incubated for a subsequent 16 h with agitation to generate fibers . Q62 fibers were recovered by centrifugation and resuspended at 5 µM . “Capped” NM fibers ( Figure 13A ) were generated by incubating preformed NM fibers ( 2 . 5 µM monomer ) with NM-his ( 5 µM ) , and the seeding reaction was allowed to proceed until 50% of NM-his had been converted to amyloid . This was verified empirically by determining the amount of SDS-resistant NM-his by quantitative immunoblot using an anti-Penta-His antibody in comparison to known quantities of NM-his . Fibers were recovered by centrifugation and washed ( without resuspending the pellet ) prior to disassembly reactions . Capped α-syn fibers were generated in the same way . Assembled NM , α-syn , or Q62 fibers ( 0 . 5–2 . 5 µM monomer ) were then incubated in AB with the indicated components and times ( refer to figure legends ) . ATP was added ( 5 mM ) plus an ATP regeneration system comprising creatine phosphate ( 40 mM ) and creatine kinase ( 0 . 5 µM ) . For long-term incubations ( Figures 12–14 ) , reactions were conducted in AB supplemented with sodium azide ( 0 . 001% ) and protease inhibitors ( Complete , Roche ) . Sodium azide and protease inhibitors were removed by dialysis prior to transformation into yeast cells . For short-term incubations ( Figure 11 ) , sodium azide and protease inhibitors were omitted . Immunoblot analysis confirmed that for long-term incubations the total amount of NM or α-syn remained constant throughout the incubation . Fiber disassembly was assessed by ThT fluorescence , electron microscopy , or by sedimentation analysis ( 436 , 000 g for 10 min at 25°C ) followed by determination of the amount of SDS-soluble protein in the supernatant or the amount of protein in the pellet fraction by quantitative immunoblot [49] , [54] , [55] , [61] . For disassembly of “capped” NM or α-syn fibers , an anti-Penta-His antibody ( Qiagen ) was used to detect the his-tagged protein , which migrates slower than untagged protein by SDS-PAGE . Thus , untagged and his-tagged protein could be readily distinguished and quantified in comparison to know amounts of untagged or his-tagged protein . Luciferase reactivation was performed as described [33] . Briefly , aggregation of firefly luciferase was elicited by heating at 45°C for 15 min in the absence or presence of indicated concentrations of Hsp26 or Hsp42 . Aggregates were then incubated in the presence of Hsp104 , Ssa1 , and Ydj1 . Luciferase reactivation was assessed using the luciferase assay system ( Promega ) . Acrylodan and pyrene fluorescence were measured as described [40] . The thermal stability of NM fibers ( Figure 10A ) was determined by incubation of fibers at increasing temperatures ( 25°C to 100°C in 10°C intervals ) for 5 min in 1 . 6% SDS , followed by SDS–PAGE and quantitative immunoblot to determine the amount of SDS-soluble NM [44] . Yeast cells from a W303-derived strain ( MATα leu2-3 , -112 his3-11 trp1-1 ura3-1 ade1-14 can1-100 [rnq−] [psi−] [ure-o] ) that contained an ADE1 nonsense mutation suppressible by [PSI+] were transformed with the indicated NM or Sup35 conformers and a URA3 plasmid . The proportion of Ura+ transformants that acquired [PSI+] was determined as described [44] , [49] . For transformations into [psi−] yeast cells expressing high levels of the indicated combination of Hsp26 , Hsp42 , Sis1 , and Ydj1 , a HIS3 or LEU2 plasmid was utilized . Δhsp26 , Δhsp42 , or Δhsp26Δhsp42 yeast strains were as described [33] . Yeast cells from a W303-derived strain ( MATa leu2-3 , -112 his3-11 trp1-1 ura3-1 ade1-14 can1-100 [RNQ+] [psi−] ) were transformed with plasmids for the overexpression of Hsp26 , Hsp42 , Ydj1 , and Sis1 together with a plasmid for the overexpression of NM fused to the yellow fluorescent protein ( NM-YFP ) or Sup35 . All the chaperone constructs were in 2 micron plasmids under the control of the constitutive GPD promoter for high expression . The NM-YFP or Sup35 construct was under the control of the inducible Gal1 promoter . Four colonies of each of the transformants were restreaked on fresh selective plates . Only colonies that presented the correct color for [psi−] , [PIN+] cells ( i . e . , red colonies ) were chosen . For each [PSI+]-induction experiment at least three independent transformants were incubated in three replicates each in 3 ml of selective liquid medium containing glucose as the sole carbon source overnight . The next day , the yeast cells were washed three times with sterile water before transferring them to selective liquid media containing galactose as the sole carbon source . The cells were incubated in the galactose media for 4 h ( for NM-YFP ) or 16 h ( for Sup35 ) at 30°C before they were diluted to an OD600 of 0 . 002 and 80 µl of these diluted cultures were evenly plated on 25% YPD plates . The plates were then incubated for 3 d at 30°C followed by an overnight incubation at 4°C for better color development . [PSI+] induction was scored as the number of white and pink ( [PSI+] colonies ) ADE+ colonies divided by the total number of colonies . Three independent yeast transformants expressing the indicated chaperones together with NM-YFP were incubated overnight in liquid selective media containing glucose as the sole carbon source . The next day , the cells were recovered and washed three times with sterile water and then transferred to selective liquid media containing galactose as the sole carbon source . Cells were incubated for 4 h at 30°C and then inspected by fluorescence microscopy using a Nikon Eclipse 300 microscope with the appropriate filters . Yeast cells from a W303-derived strain ( MATa leu2-3 , -112 his3-11 trp1-1 ura3-1 ade1-14 can1-100 [rnq−] [PSI+] ) were transformed with plasmids for the overexpression of Hsp26 , Hsp42 , or Hsp104 . All these constructs were in 2 micron plasmids under the expression control of the constitutive GPD promoter for high expression . Four colonies of each of the transformants were restreaked on fresh selective plates . Only colonies that presented the correct color for [PSI+] cells ( i . e . , white colonies ) were chosen . For each [PSI+]-curing experiment at least three independent transformants were incubated in three replicates each in 3 ml selective liquid medium containing glucose as the sole carbon source overnight . The next day , the cultures were diluted to an OD600 of 0 . 2 and incubated for 6 h . The yeast cultures were then diluted to an OD600 of 0 . 002 and 80 µl of these diluted cultures were spread on 25% YPD plates ( resulting in ∼700 colonies per plate ) . The plates were then incubated for 3 d at 30°C followed by an overnight incubation at 4°C for better color development . [PSI+] curing was scored as the proportion of red ade− [psi−] colonies . | Amyloid fibers are protein aggregates that are associated with numerous neurodegenerative diseases , including Parkinson's disease , for which there are no effective treatments . They can also play beneficial roles; in yeast , for example , they are associated with increased survival and the evolution of new traits . Amyloid fibers are also central to many revolutionary concepts and important questions in biology and nanotechnology , including long-term memory formation and versatile self-organizing nanostructures . Thus , there is an urgent need to understand how we can promote beneficial amyloid assembly , or reverse pathogenic assembly , at will . In this study , we define the mechanisms by which small heat-shock proteins synergize to regulate the assembly and disassembly of a beneficial yeast prion . We then exploit this knowledge to discover an amyloid depolymerase machinery that is conserved from yeast to humans . Remarkably , the human small heat shock protein , HspB5 , stimulates Hsp110 , Hsp70 , and Hsp40 chaperones to gradually depolymerize amyloid fibers formed by α-synuclein ( which are implicated in Parkinson's disease ) from their ends on a biologically relevant timescale . This newly identified and highly conserved amyloid-depolymerase system could have important therapeutic applications for various neurodegenerative disorders . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"protein",
"chemistry",
"proteins",
"protein",
"folding",
"macromolecular",
"assemblies",
"biology",
"biophysics",
"neuroscience"
] | 2012 | Small Heat Shock Proteins Potentiate Amyloid Dissolution by Protein Disaggregases from Yeast and Humans |
Plague is a life-threatening disease caused by the bacterium , Yersinia pestis . Since the 1990s , Africa has accounted for the majority of reported human cases . In Uganda , plague cases occur in the West Nile region , near the border with Democratic Republic of Congo . Despite the ongoing risk of contracting plague in this region , little is known about Y . pestis genotypes causing human disease . During January 2004–December 2012 , 1 , 092 suspect human plague cases were recorded in the West Nile region of Uganda . Sixty-one cases were culture-confirmed . Recovered Y . pestis isolates were analyzed using three typing methods , single nucleotide polymorphisms ( SNPs ) , pulsed field gel electrophoresis ( PFGE ) , and multiple variable number of tandem repeat analysis ( MLVA ) and subpopulations analyzed in the context of associated geographic , temporal , and clinical data for source patients . All three methods separated the 61 isolates into two distinct 1 . ANT lineages , which persisted throughout the 9 year period and were associated with differences in elevation and geographic distribution . We demonstrate that human cases of plague in the West Nile region of Uganda are caused by two distinct 1 . ANT genetic subpopulations . Notably , all three typing methods used , SNPs , PFGE , and MLVA , identified the two genetic subpopulations , despite recognizing different mutation types in the Y . pestis genome . The geographic and elevation differences between the two subpopulations is suggestive of their maintenance in highly localized enzootic cycles , potentially with differing vector-host community composition . This improved understanding of Y . pestis subpopulations in the West Nile region will be useful for identifying ecologic and environmental factors associated with elevated plague risk .
Yersinia pestis is the etiological agent of plague , a severe and often fatal disease in humans [1] . Transmission of Y . pestis to humans most often occurs through the bite of an infectious flea or by direct exposure to an infected mammalian host . Less frequently , human infection is the result of inhaling infectious respiratory droplets . The three primary clinical forms of human disease are bubonic , septicemic , and pneumonic plague . Bubonic plague is the most common and is characterized by one or more swollen and painful lymph nodes ( buboes ) , while pneumonic plague is the most severe , with fatality rates approaching 100% in patients with untreated disease [2] . In nature , Y . pestis is maintained by various small mammal hosts and their associated fleas . Epizootics among susceptible mammals often precede human plague cases . The high host mortality forces infected fleas to bite alternate hosts , including humans [3] . The geographic distribution of plague is widespread with foci in the Americas , Africa , and Asia [1] . In recent decades , the majority of human cases have been reported from East Africa ( Uganda , Tanzania , Democratic Republic of Congo ) and Madagascar , in resource limited areas , where the proximity to commensal rats and other small mammals increases the likelihood for human contact with infected animals or their fleas [1 , 4 , 5 , 6] . In Uganda , the current plague focus encompasses an area of approximately 900 km2 in the West Nile region , situated above the Rift Valley escarpment and within the districts of Arua and Zombo [7 , 8] . Human cases in this region are concentrated in the counties of Okoro and Vurra and occur primarily between the months of September and December each year [9] . For the time period of 1999–2007 , an incidence rate of >5 cases per 1 , 000 individuals was reported [7] . The risk for plague in this region is greater at elevations above 1 , 300 meters , and positively correlates with higher amounts of rainfall as compared to lower elevations . Ecologic differences above 1 , 300 meters compared with below include an increase in both abundance and diversity of small mammals and a greater diversity of flea species on mammalian hosts [4 , 8–11] . Y . pestis genotypes that cause human disease in the West Nile region of Uganda have not been previously described . Worldwide , limited nucleotide variation is observed between strains and Y . pestis is considered a genetically monomorphic pathogen [12] . Traditionally , three biovars of Y . pestis , antiqua , medievalis , and orientalis , have been differentiated based on their ability to metabolize glycerol or reduce nitrate [13] . With the advent of the genomic era and a rise in the number of sequenced Y . pestis genomes , single nucleotide polymorphisms ( SNPs ) have been used as the basis for defining worldwide populations of Y . pestis [14] . A global SNP analysis , based on 17 whole genome sequences to discover SNPs , demonstrated that Y . pestis strains separate into several populations with distinctive geographic patterns , including 1 . ORI ( orientalis; North and South America , Madagascar , Southeast Asia ) , 2 . MED ( mediaevalis; Asia ) , 1 . ANT ( antiqua; East and Central Africa ) , and 2 . ANT ( antiqua; Asia ) [14] . Although SNPs are useful phylogenetic markers , their discovery from comparison of a limited number of whole genome sequences has only limited resolving power . For higher level differentiation , other mutation types have been exploited in Y . pestis genomes , including tandem repeats , insertion sequence ( IS ) mediated rearrangements , clustered regularly interspaced short palindromic repeats ( CRISPRs ) , and insertions/deletions ( INDELs ) [15–18] . Strain typing methods capitalizing on these mutations include multi-locus variable number of tandem repeat ( VNTR ) analysis ( MLVA ) , pulsed field gel electrophoresis ( PFGE ) , CRISPR genotyping , and restriction fragment linked polymorphism ( RFLP ) of IS elements [15 , 17 , 19–22] . Here , three molecular typing methods detecting different genome mutations , SNPs , MLVA , and PFGE , were used to characterize 61 Y . pestis isolates recovered from human plague patients in the West Nile region of Uganda over a 9 year time span from 2004–2012 . Identified subpopulations were analyzed in the context of associated geographic , temporal , and clinical data for source patients .
Y . pestis isolates ( not de-linked from originating patients ) were cultured from samples collected with documented informed consent during three research studies approved by institutional review boards at the U . S . Centers for Disease Control and Prevention ( CDC ) , the Uganda Virus Research Institute , and the Uganda National Council for Science and Technology . A small number of archived clinical isolates banked for public health purposes were also included . In total , 61 Y . pestis isolates were cultured from human plague cases occurring in Vurra and Okoro counties within the Arua and Zombo districts , respectively . This included the previously sequenced UG05-0454 strain , which was isolated from a patient in the West Nile region in 2004 [14] . For isolation of cultures , patient specimens ( blood , bubo aspirate or sputum ) were plated on 6% sheep blood agar ( SBA ) or cefsulodin-irgasan-novobiocin ( CIN ) agar and incubated at 35°C or 25°C , respectively . Plates were checked daily for growth and Y . pestis isolates identified by direct fluorescence antibody ( DFA ) , polymerase chain reaction ( PCR ) , and bacteriophage lysis [23] . For isolation of DNA , isolates were grown on 6% SBA at 35°C for 24 hrs and DNA isolated using the QIAmp DNA Mini Kit ( Qiagen , Hilden , Germany ) . Clinical and demographic data on source patients was collected as part of ongoing surveillance and diagnosis of plague in the region . Extracted data included patient age , sex , illness onset date , form of clinical disease , and illness outcome . GPS coordinates for each patient’s residence were recorded as the likely exposure site . A total of 9 SNPs , previously identified from whole genome sequencing of the Y . pestis strain UG05-0454 , were selected for genotyping [14] . Melt mismatch amplification mutation assays ( Melt-MAMA ) were developed for each SNP [24–26] . Ancestral , derived , and consensus primer sequences for each assay are listed in S1 Table . All Melt-MAMA reactions were performed in a 10 μl final volume and contained SYBR green PCR Master Mix ( Applied Biosystems , Foster City , CA ) at a 1X final concentration , a common reverse primer ( 200 nM ) , derived and ancestral allele-specific MAMA primers ( 200 nM ) , water , and 1 μl of diluted template ( approximately 2 ng DNA ) . Melt-MAMA assays were performed on an Applied Biosystems 7500 Fast Dx real-time PCR system using SDS software v1 . 4 and the following cycling conditions: 50°C for 2 min , 95°C for 10 min , then 95°C for 15 s and 55°C ( s447: CO92 gene ypo0064 ) or 60°C ( all others ) for 1 min for 33 cycles followed by melt curve analysis using a dissociation protocol of 95°C for 15 s , then temperature ramping in 0 . 2°C/min increments from 60°C to 95°C . PFGE was performed with AscI enzyme using the standardized PulseNet protocol ( http://www . pulsenetinternational . org/protocols ) . Isolates were grown on 6% SBA at 35°C for 24 hrs . Cells were resuspended in buffer ( 100 mM Tris:100 mM EDTA ) at an optical cell density of 0 . 45–0 . 50 and agarose plugs prepared . To lyse cells , plugs were incubated in 50 mM Tris:50 mM EDTA , 0 . 1mg/ml Proteinase K , 1% sarcosyl for 2 hrs at 55°C , then washed in sterile water followed by TE buffer . DNA in embedded plugs was digested for 5 hrs with 40 U of AscI enzyme ( New England Biolabs , Ipswich , MA ) at 37°C and PFGE performed for 17 . 5 hrs at 14°C . Salmonella enterica serotype Braenderup ( H9812 ) cut with 50 U XbaI enzyme ( Roche Diagnostics , Indianapolis , IN ) was used as a reference standard . PFGE parameters were 6 V/cm , 120°C , linear ramping , and switch times of 1 . 79 s-18 . 66 s . Gels were stained with ethidium bromide and imaged using a Gel Doc 1000 ( BioRad , Hercules , CA ) . PFGE patterns were analyzed using BioNumerics software Version 6 . 64 ( Applied Maths , Inc . , Sint-Martens-Latem , Belgium ) and patterns normalized to the reference standard . Cluster analysis of PFGE patterns was performed using Dice similarity coefficients ( 1 . 0% optimization and 1 . 5% tolerance ) and unweighted pair group method with averages ( UPGMA ) . Mutation rate estimates for Y . pestis AscI PFGE loci were determined from a parallel , single passage experiment [25 , 26] . Briefly , a single colony of UG09-2299 was used to generate an in vitro population representing ~10 , 000 generations . The generation rate was determined by preparing 10 fold serial dilutions from a single pgm+ colony resuspended in 1 ml of PBS , plating dilutions to 6% SBA plates and counting colonies after growth for 48 hrs at 35°C . After growth for 48 hrs at 35°C , 100 colonies were selected and streaked to 100 SBA plates to represent passage 1 . This process was repeated 3 more times and PFGE performed on the 100 clones from the final passage . Congo Red plates were included at all steps to verify the presence of the pigmentation ( pgm ) locus . The mutation rate was determined using the following: number of clonal lineages x number of passages x generations/colony . Eighteen VNTR markers located across the Y . pestis genome , with previously determined in vitro mutation rates , were chosen [15 , 27 , 28] . PCR reactions were performed using PuReTaq Ready-To-Go PCR beads ( GE Healthcare Life Sciences , Pittsburgh , PA ) and the following cycling conditions: 94°C for 5 min , 35 cycles of 94°C for 20 s , 65°C for 20 s , and 72°C for 30 s , then 72°C for 5 min . PCR products were visualized on a QIAxcel system ( Qiagen , Hilden , Germany ) using the QIAxcel DNA High Resolution cartridge . VNTR amplicon sizes were calculated using BioCalculator 3 . 2 software ( Qiagen , Hilden , Germany ) and the QX reference alignment marker 15-500bp and QX DNA size marker 25-500bp ( Qiagen , Hilden , Germany ) . Amplicon size and repeat length was used to determine the number of motifs in each allele . A correction factor was calculated for each marker using the product size estimated for the Y . pestis reference control strain CO92 ( 1 . ORI; North America ) and the known size for the loci in the CO92 genome . Repeat copy numbers for each loci in each strain were determined by applying the correction factor to the product size determined by the QIAxcel . The resulting data was imported into BioNumerics software Version 6 . 64 and cluster analysis performed using categorical coefficients and UPGMA . Epidemiologic analyses were performed with SAS , version 9 . 3 ( SAS Institute , Cary , NC ) . χ2 tests were used for categorical data; Wilcoxon rank-sum tests were used to compare continuous data . Analyses were considered to be statistically significant when P < . 05 . Maps were generated with the use of ArcMap , version 10 . 3 ( ESRI , Redlands , CA ) by plotting isolates based on global positioning systems ( GPS ) coordinates of patient residence .
During January 2004—December 2012 , a total of 1 , 092 suspect human plague cases were recorded in the West Nile region of Uganda , with a high in 2007 of 321 cases and a low in 2011 of 14 cases ( Fig 1 ) . Overall , 61 cases ( <10% of suspect cases ) were culture confirmed during this time period; 5 from 2004 ( 8% ) , 45 from 2008 ( 74% ) , 1 from 2009 ( 2% ) , 2 from 2011 ( 3% ) and 8 from 2012 ( 13% ) ( Fig 1 ) . Based on unique SNPs previously identified in the UG05-0454 strain [14] , nine melt-MAMA assays were developed and used to screen the 61 Y . pestis isolates ( Table 1 and S2 Table ) . Across all isolates , four SNPs ( s376 , s1427 , s357 and s1333 ) were the same as UG05-0454 , indicating all 61 Y . pestis isolates fall into the 1 . ANT lineage . One of these SNPs , s1333 , was conserved across 58 isolates , but for 3 isolates this SNP could not be determined due to multiple failed amplification attempts ( S2 Table ) . Three of the SNPs , s949 , s860 , and s968 , separated the 61 Y . pestis isolates into 2 genetic groups , termed Group 1 and Group 2 , with 22 ( 36% ) and 39 ( 64% ) strains comprising Group 1 and 2 , respectively . The remaining two SNPs , s272 and s447 , were unique to UG05-0454 . PFGE typing ( using AscI ) of the 61 Y . pestis isolates yielded multiple PFGE patterns ( Fig 2 ) . The majority of isolates ( 84% ) clustered into two PFGE groups , termed Group A and Group B , which were differentiated from each other by at least 5 band differences ( Fig 3 ) . The remaining ten isolates did not cluster together and most displayed PFGE patterns with a higher molecular weight AscI fragment as compared to PFGE patterns for isolates in Group A and Group B . Among isolates in Groups A and B , differences in PFGE patterns were also observed , although these differences were primarily limited to a single band . Comparison of groups assigned by SNPs and PFGE indicated that all isolates assigned to SNP Group 1 were classified as PFGE Group A . Isolates assigned to SNP Group 2 included the 29 isolates from PFGE Group B as well as the 10 additional isolates that did not fall into PFGE Group A or B . The diversity of PFGE patterns observed among the Ugandan Y . pestis strains prompted an in vitro parallel passage experiment to determine if AscI loci were rapidly changing . Comparison of PFGE patterns for 100 clones , representing 10 , 000 generations of the isolate UG09-2299 , demonstrated no differences between the PFGE patterns , indicating no chromosomal changes affected these loci over the time frame analyzed . The mutation rate frequency of AscI loci was calculated to be >10−4 ( 100 clonal lineages x 4 passages x 25 . 26 generations/colony ) . Eighteen VNTR loci were chosen to encompass the entire chromosome and to include a mixture of slow , intermediate , and rapidly changing loci . Among the 61 isolates , 16 of the VNTR loci ( M12 , M18 , M19 , M21 , M22 , M23 , M25 , M27 , M28 , M29 , M31 , M34 , M58 , M59 , M68 , M79 ) were found to be polymorphic while 2 markers ( M24 and M33 ) showed no changes in repeat number ( S2 Table ) . The number of repeats varied among the VNTR loci , ranging from 0 to 17 repeats as compared to the CO92 reference strain . Cluster analysis based on the 18 VNTRs yielded two distinct groups , denoted Group I and Group II ( Fig 4 ) . Isolates assigned to MLVA Group I and II showed 100% correlation with those isolates in SNP Groups 1 and 2 . As expected , resolution of individual isolates was achieved by MLVA ( Fig 4 ) . Overall , most culture-confirmed plague cases ( 88% ) occurred during the months of October through December . The median age of source patients was 13 years ( range: 3 years-65 years ) ; most patients ( 61% ) were female . The majority of infections ( 84% ) were bubonic; overall plague mortality was 34% . The month or year of illness onset did not differ among infections caused by strains within SNP Groups 1 and 2 . Strains within SNP Group I and II were isolated throughout the entire 9 year period of the study , specifically in three of the five years in which cases were cultured confirmed ( 2004 , 2008 and 2012 ) . Likewise , age , sex , clinical form of disease , and illness outcome did not differ between infections caused by SNP Group 1 or 2 . GPS coordinates corresponding to the source patients’ residence were available for 56 isolates ( 92% ) . Source patient residences’ mapped to an overall area of approximately 330 km2 ( 12 km across and 44 km long ) . Geographic distribution differed between infections caused by the two SNP groups . Group 1 strains were almost entirely within Vurra county , whereas Group 2 strains predominated in Okoro county ( Fig 5 ) . Furthermore , SNP Group 1 samples were from more northerly ( median latitude: 2 . 8811° , range: 2 . 5955°-2 . 9064° ) locations relative to SNP Group 2 ( median latitude: 2 . 7197° , range: 2 . 5105°-2 . 8398°; Wilcoxon rank sums test , P<0 . 0001 ) . Geographic separation of the two subpopulations was not absolute; overlap was observed in several areas ( Fig 5 ) . The observed geographic difference for the two genetic groups was paralleled by a significant difference in the elevation at which the source patients resided . The median elevation for Group 1 was 1330 meters ( range: 1272–1496 ) as compared to 1442 meters ( range: 1342–1608 ) for Group 2 ( Wilcoxon rank sum test , P <0 . 0001 ) .
Here , we investigated the diversity and molecular epidemiology of Y . pestis strains causing human infection in the West Nile region of Uganda . All 61 isolates recovered over the 9 year time period ( 2004 through 2012 ) fell into the 1 . ANT lineage . Strains belonging to this lineage were previously shown to be geographically restricted to East and Central Africa and evolution of the 1 . ANT lineage estimated to have occurred 628–6 , 914 years ago [14] . Three independent typing methods , SNPs , MLVA and PFGE , revealed that human plague in the West Nile region of Uganda is caused by at least two distinct 1 . ANT genetic subpopulations , which are largely segregated by elevation . The two Y . pestis subpopulations persisted through the 9 year period consistent with their maintenance in highly localized enzootic cycles . Source patients whose infections were due to SNP Group 1 strains lived at significantly lower elevations and further north as compared to those patients whose infections were due to SNP Group 2 strains , consistent with the relative independence of the two Y . pestis subpopulations , despite being localized to an overall area of only 330 km2 . Of note , plague cases in Uganda were recorded only from Okoro county in Nebbi district for much of the 20th century , until 1989 , when cases were first noted further north in the Arua district [29] . It’s therefore possible the more northern Group 1 subpopulation represents a recent emergence . We hypothesize the genetic differences between the two Y . pestis subpopulations may be associated with maintenance of the two subpopulations in different foci involving enzootic cycles with differing vector-host community composition . Subtle differences in vector community composition were observed previously between higher elevation sites in Zombo compared with lower elevation sites in Arua [4] . Xenopsylla cheopis was commonly encountered in the lower elevation sites , but rare or absent at higher elevations where it was replaced by X . brasiliensis , a finding that was consistent across studies [4 , 10] . Typing of Y . pestis strains isolated from small mammals and fleas across an elevation gradient will be important for determining what host and vector species are involved in enzootic maintenance of the two subpopulations . Additionally , as small mammals and fleas have considerably smaller home-ranges compared with humans , Y . pestis strains from these hosts may better delineate the geographic boundaries of the two genetic subpopulations . Notably , the three typing methods used here , SNPs , PFGE and MLVA , all identified two genetic subpopulations , even though each method recognizes different mutation types in the Y . pestis genome ( i . e . single nucleotide variation , tandem repeats , insertion sequence ( IS ) mediated rearrangements ) . Although previous studies have shown correlation between strain groupings by SNPs and MLVA [24 , 30] , the links detected here between PFGE and SNPs and PFGE and MLVA were not expected . Genome sequencing of Y . pestis isolates with distinct FseI PFGE patterns previously demonstrated that PFGE band differences were due to large-scale IS element mediated genome rearrangements ( different order and orientation of genome segments ) [20] . With the rapid rise in the number of assembled Y . pestis genomes , large-scale rearrangements mediated by IS elements has become an emerging theme in this pathogen [18 , 20 , 31 , 32] . It is therefore likely that two Y . pestis subpopulations identified by AscI PFGE have rearranged genome orders with respect to one another ( e . g . , 5 band difference ) and that other genomic rearrangements , as evidenced by AscI PFGE band differences , are present within each of the two subpopulations . A correlation between large-scale genome rearrangement and increased nucleotide variation localized to rearrangement breakpoints , has recently been shown in several other prokaryotes , including Burkholderia , Pseudomonas and Shigella [33] . Genome sequencing and assembly of Y . pestis strains encompassing the two West Nile subpopulations will be important for determining if increased nucleotide variation may also be observed near genome rearrangements in Y . pestis , thus explaining the correlation between typing methods that detect differences in nucleotides or genome structure . Previous analyses of Y . pestis and other bacterial vector-borne zoonotic pathogens have demonstrated the presence of multiple strains and subpopulations within a defined , localized geographical region [20 , 34] . The finding that human plague occurring in a limited geographic area of Uganda is also caused by distinct genetic subpopulations of Y . pestis , is consistent with genomic diversification of Y . pestis at small geographic scales . The identification and characterization of Y . pestis subpopulations causing human plague in the West Nile region will be useful for ongoing efforts to identify ecologic and environmental factors associated with elevated plague risk . | Plague , a severe and often fatal zoonotic disease , is caused by the bacterium Yersinia pestis . Currently , the majority of human cases have been reported from resource limited areas of Africa , where the proximity to commensal rats and other small mammals increases the likelihood for human contact with infected animals or their fleas . Over a 9 year time period , >1000 suspect cases were recorded in the West Nile region of Uganda within the districts of Arua and Zombo . Culture-confirmed cases were shown by three independent typing methods to be due to two distinct 1 . ANT genetic subpopulations of Y . pestis . The two genetic subpopulations persisted throughout the 9 year time period , consistent with their ongoing maintenance in local enzootic cycles . Additionally , the two subpopulations were found to differ with respect to geographic location and elevation , with SNP Group 1 strains being found further north and at lower elevations as compared to SNP Group 2 . The relative independence of the two Y . pestis subpopulations is suggestive of their maintenance in distinct foci involving enzootic cycles with differing vector-host community composition . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"plagues",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"pathogens",
"geographical",
"locations",
"microbiology",
"uganda",
"vertebrates",
"animals",
"pulmonology",
"mammals",
"bacterial",
"diseases",
"yersinia",
"fleas",
"molecular",
"biology",
"techniques",
"bacteria",
"africa",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"yersinia",
"pestis",
"artificial",
"gene",
"amplification",
"and",
"extension",
"medical",
"microbiology",
"microbial",
"pathogens",
"molecular",
"biology",
"insects",
"arthropoda",
"people",
"and",
"places",
"polymerase",
"chain",
"reaction",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"organisms"
] | 2016 | Two Distinct Yersinia pestis Populations Causing Plague among Humans in the West Nile Region of Uganda |
Within the liver a single Plasmodium parasite transforms into thousands of blood-infective forms to cause malaria . Here , we use RNA-sequencing to identify host genes that are upregulated upon Plasmodium berghei infection of hepatocytes with the hypothesis that host pathways are hijacked to benefit parasite development . We found that expression of aquaporin-3 ( AQP3 ) , a water and glycerol channel , is significantly induced in Plasmodium-infected hepatocytes compared to uninfected cells . This aquaglyceroporin localizes to the parasitophorous vacuole membrane , the compartmental interface between the host and pathogen , with a temporal pattern that correlates with the parasite’s expansion in the liver . Depletion or elimination of host AQP3 expression significantly reduces P . berghei parasite burden during the liver stage and chemical disruption by a known AQP3 inhibitor , auphen , reduces P . falciparum asexual blood stage and P . berghei liver stage parasite load . Further use of this inhibitor as a chemical probe suggests that AQP3-mediated nutrient transport is an important function for parasite development . This study reveals a previously unknown potential route for host-dependent nutrient acquisition by Plasmodium which was discovered by mapping the transcriptional changes that occur in hepatocytes throughout P . berghei infection . The dataset reported may be leveraged to identify additional host factors that are essential for Plasmodium liver stage infection and highlights Plasmodium’s dependence on host factors within hepatocytes .
Malaria remains one of the greatest burdens to global health with an estimated 440 , 000 deaths in 2015 [1] . While fatalities attributed to malaria have significantly declined over the past decade as a result of increased interventions , the spread of resistance to current antimalarial treatments and insecticides threatens the progress that has been made in eliminating this infectious disease . The causative agent of malaria is the Plasmodium parasite , an obligate intracellular pathogen that strategically exploits host processes to ensure its survival . Many of these interactions between the host and parasite remain unknown , particularly during the parasite’s elusive liver stage . The Plasmodium parasite is transferred to the human host via the bite of an Anopheles mosquito in the form of a sporozoite . Before the Plasmodium parasite can invade erythrocytes and cause disease , it must first pass through an obligatory liver stage . Sporozoites travel through the bloodstream to the liver where they pass through multiple hepatocytes before establishing an infection in a final cell [2–5] . As the sporozoite invades the hepatocyte , invagination of the host cell membrane forms the parasitophorous vacuole membrane ( PVM ) around the parasite , protecting it from clearance [6 , 7] . It is within this PV that the sporozoite undergoes dramatic morphological changes to become the exo-erythrocytic form ( EEF ) [8] , which subsequently divides into tens of thousands of merozoites within a single hepatocyte [9] . After maturation , the blood infective merozoites are released into the blood stream where they cause the symptoms of malaria . The liver stage of the Plasmodium life cycle is an attractive target for drug and vaccine development as it represents a bottleneck in the parasite population . Furthermore , inhibition of liver-stage parasites prevents disease manifestation that emerges in the subsequent blood stage . During this stage , the parasite relies on host cell factors for successful invasion , development , and nutrient acquisition to support its massive replication [10] . However , few host proteins that are involved in the maturation of these parasites have been identified . The host CD81 receptor [11] , scavenger receptor type B class I ( SR-BI ) [12 , 13] , and the surface protein EphA2 [14] have all been identified as important invasion factors , but among these only SR-BI has been implicated in the maturation of parasites after invasion . Few other host proteins are known to be critical for the developing parasites . Thus , while significant advances have been made to our current understanding of hepatocyte invasion , host-pathogen interactions that are beneficial for the parasite’s development within the hepatocyte remain poorly understood . Host-based strategies represent an attractive , yet underexplored route to malaria control as they reduce the probability that parasites will develop drug resistance when compared to parasite-directed therapeutics [15] . However , a greater understanding of host-parasite interactions during the liver stage is critical to develop these measures . A comprehensive analysis of the host response to infection during the liver stage has been challenging because of the low infection rate of hepatocytes by sporozoites [16] and the technical difficulty of obtaining sporozoites in large quantities from the salivary glands of live mosquitoes . Previous work aiming to elucidate important hepatocyte factors assessed the host transcriptome in response to P . berghei infection by microarray analysis . Researchers found an initial host response to stress followed by maintenance of cell viability in response to infection [17] . This important work provided a global analysis of the host response to Plasmodium infection , but it remains unclear whether any of the differentially expressed genes ( DEGs ) are regulated in a manner that is advantageous to the parasite . In addition to the previously reported sequential regulation of genes in hepatocytes , we hypothesized that conserved host-parasite interactions during the liver stage have evolved that are essential for the proper maturation and development of EEFs . To search for essential host factors , we used RNA-sequencing ( RNA-seq ) to identify genes that are induced in HepG2 hepatocytes upon P . berghei infection . Among genes that were differentially expressed , aquaporin-3 ( AQP3 ) was chosen for functional analysis because it was one of the most highly induced genes upon infection and because of previous literature related to aquaporins and parasite development . Aquaporins ( AQPs ) are a family of highly conserved membrane proteins that transport water and other small solutes [18] . There are 13 human AQP isoforms that belong to one of two groups: orthodox aquaporins that strictly transport water and aquaglyceroporins that are capable of transporting small solutes , such as glycerol , in addition to water molecules across membranes [19 , 20] . The role of the aquaporin family of proteins has been previously studied in the context of Plasmodium infections , but primarily focusing of the parasite aquaglyceroporin [21] . The P . falciparum aquaporin , PfAQP , localizes to the parasite cell membrane where it is thought to be critical for glycerol transport . The role of host aquaporins during Plasmodium infection is understood to a lesser extent , but have been implicated in blood stage infections [21] where one report demonstrated that AQP3 localizes to the parasitophorous vacuole of P . falciparum infected red blood cells [22] . AQP3 is an aquaglyceroporin that transports both water and glycerol in mammalian cells [20 , 23] . We found that AQP3 expression was induced in human hepatocytes in response to P . berghei infection at times that implicated a role in the rapid expansion and replication of the parasite . Our subsequent molecular studies revealed that this host protein is actively trafficked to the PVM after invasion . Genetic disruption or depletion of AQP3 leads to a significant decrease in P . berghei liver stage parasite load . Furthermore , treatment with auphen , a known AQP3 inhibitor , reduces both P . berghei parasite load in hepatocytes and P . falciparum parasite load in erythrocytes , suggesting the host protein has an essential role during various parasite life stages . Importantly , deletion of AQP3 in mice is not lethal [24] , suggesting the host protein has potential as a possible therapeutic target . This work enhances our current understanding of host liver processes influenced by Plasmodium infection and highlights the potential of targeting these processes for future drug and vaccine development against malaria . Further , the provided datasets of DEGs during Plasmodium infection can serve as the foundation for future studies aimed at understanding the host response .
To evaluate potential host responses to Plasmodium infection , we performed RNA-seq on HepG2 human hepatocytes infected with GFP-expressing P . berghei . Because of the inherently low infection rate of sporozoites in the best in vitro model systems [16] , we isolated infected cells by fluorescence-activated cell sorting ( FACS ) ( S1 Fig ) . Approximately 3 , 000 Plasmodium infected-cells were sorted at 4 , 24 , and 48 hpi as well as uninfected cells in 2–4 biological replicates . RNA isolation and library preparation followed by sequencing was achieved using a protocol designed for low cell numbers . From an evaluation of each candidate gene , AQP3 was chosen for functional studies because it was one of the most significantly induced genes and evidence from previous studies that aquaporins are involved in Plasmodium infection [21 , 22] . RNA-seq analysis shows that AQP3 is one of the most differentially expressed genes in HepG2 cells infected with P . berghei . Human AQP3 , which is widely expressed in the kidney , skin , and erythrocytes [25–27] , can transport both water and glycerol . The human genome encodes thirteen AQP isoforms , and based on literature , AQP3 , AQP6 , AQP7 , AQP8 , AQP9 , and AQP11 mRNAs are expressed in uninfected HepG2 cells [28] . Upon infection , AQP3 expression is induced and it is the only aquaporin isoform to be significantly ( p = 1 . 55 x 10−7 ) differentially expressed ( Fig 1A ) . We further confirmed the induction of AQP3 expression in P . berghei-infected hepatocytes by qRT-PCR of RNA extracted from both HepG2 and HuH7 infected cells ( Fig 1B ) . At 48 hpi , HepG2 cells exhibited a 15-fold increase in AQP3 mRNA expression ( p = 0 . 001 , two-tailed Student’s t-test; n = 3 independent experiments ) and HuH7 cells , an alternate human hepatoma cell line , had a 4-fold increase in AQP3 expression compared to uninfected HuH7 cells ( p = 0 . 0075 , two-tailed Student’s t-test; n = 4 independent experiments ) . AQP3 is a membrane protein that generally localizes to the plasma membrane in human cells where it facilitates the transport of water and glycerol [18] . However , upon hepatocyte infection by Plasmodium a second membrane is present , the PVM , which is derived from the host cell plasma membrane when the parasite first penetrates the hepatocyte . Because Plasmodium infection leads to the upregulation of AQP3 gene expression , we hypothesized that the protein localizes to the PVM of Plasmodium-infected hepatocytes . Evidence of human AQP3 localization to the PVM in P . falciparum-infected erythrocytes also supported this hypothesis [22] . To test if AQP3 localizes to the PVM during the liver stage , we used immunofluorescent ( IF ) microscopy with antibodies targeting human AQP3 and monitored the intracellular localization of the protein throughout infection . Visualization of AQP3 in P . berghei-infected hepatocytes demonstrated that the protein localizes to the PVM in both HepG2 and HuH7 cells ( Fig 2 , S1 Movie ) . We utilized confocal microscopy to show colocalization of AQP3 with UIS4 , a parasite protein that is known to be integrated into the PVM [29] , in HepG2 cells ( Fig 2A ) . A Pearson colocalization coefficient of 0 . 56 ± 0 . 014 ( mean ± SD ) was calculated from z-stacked images of three individual cells with staining for AQP3 and UIS4 . We also show that AQP3 localizes to the PVM of P . berghei infected HuH7 cells as early as 28 hpi , after the parasite initiates nuclear division ( Fig 2B ) . Some infected cells showed partial localization of AQP3 to the host cell membrane at 28 hpi , but localization thereafter is restricted to the PVM . In this in vitro infection model system protein quantification by Western blot analysis is difficult due to the limited number of Plasmodium-infected cells that can be feasibly collected; however , antibody staining shows that protein levels are elevated in P . berghei infected hepatocytes when compared to uninfected cells ( S2 Fig ) . Taken together , this analysis indicates that human AQP3 protein levels in hepatocytes increase in response to P . berghei infection , in agreement with the observed increase in gene transcription , and that AQP3 is actively trafficked to the PVM rather than being incorporated from the host cell membrane during parasite penetration of the host cell . To address whether host AQP3 expression in P . berghei infected cells is necessary for the parasites to undergo morphological changes and replication within the PV , AQP3 mRNA was depleted by RNA interference ( RNAi ) and mutant AQP3 cell lines were generated . First , AQP3 mRNA expression was depleted by pooling two small interfering RNAs ( siRNAs ) targeting AQP3 ( Qiagen ) . Scavenger receptor type B class I ( SR-BI ) siRNA depletion was used as a positive control as it is one of the few host genes that is known to be important for parasite development [12] . Efficiency of siRNA knockdowns was evaluated by extracting RNA from cells 48 hours after siRNA transfection and measuring mRNA expression of targeted genes by qRT-PCR . In parallel experiments , the impact of gene silencing on parasite load and cell viability was assessed in hepatocytes 48 hours after P . berghei infection . Parasite load after infection was quantified using a luciferase reporter constitutively expressed in P . berghei parasites [30] . Upon AQP3 siRNA reverse transfection of HuH7 cells , a 90% reduction in AQP3 transcripts resulted in a 60% reduction in parasite load ( p < 0 . 0001 , two-tailed Student’s t-test; n = 3 independent experiments ) ( Fig 3A and 3B ) . Reverse transfection of hepatocytes did not lead to any changes in cell viability 48 hpi ( S3A Fig ) . The depletion of SR-BI transcripts by 86% led to a 68% reduction in parasite load ( p < 0 . 0001 , two-tailed Student’s t-test; n = 3 independent experiments ) , similar to what was observed in AQP3 depleted cells . To further assess the impact of aquaporins on liver stage Plasmodium parasite development , all human aquaporin isoforms present in hepatocytes [28] were targeted by RNAi . Each gene was targeted by four individual siRNAs ( S1 Table ) . Efficiency of siRNA knockdowns , parasite load , and cell viability were assessed ( S2 Table , S3B and S3C Fig ) and Fig 3 shows representative data that include the two siRNAs for each gene that correspond to the greatest mRNA transcript depletion . The only aquaporin genes with a significant decrease in expression and simultaneous reduction in parasite load , with a minimum of two separate siRNAs , were AQP3 , AQP7 and AQP9 , all of which are aquaglyceroporins . AQP3 mRNA depletion had the greatest influence on parasite load with an average of 69% reduction ( Fig 3C ) . AQP7 and AQP9 depletion also significantly reduced parasite load , resulting in a 65% and 63% reduction on average , respectively ( Fig 3C ) . One siRNA targeting the orthodox aquaporin , AQP11 ( AQP11-1 in Fig 3C ) , reduced parasite load by 52% ( p = 0 . 0343 , One-way ANOVA , Dunnnet’s multiple comparison test ) . However , all four siRNAs reduced AQP11 transcripts ( >85% ) with only AQP11-1 significantly altering parasite load , suggesting its activity is due to off-target effects . Importantly , liver cell viability was not altered upon siRNA transfection of the selected aquaglyceroporins , indicating that parasite load reduction was not a byproduct of hepatocyte toxicity ( p > 0 . 5 , One-way ANOVA , Dunnett’s multiple comparison test; n = 3 ) . To validate the role of AQP3 on parasite load by an alternate means , the AQP3 gene was genetically mutated in HuH7 cells using CRISPR/Cas9 [31] . Multiple guide RNAs ( gRNAs ) were used for the generation of AQP3 mutant cell lines , targeting either exon 1 or 2 of the AQP3 gene ( S3 Table ) . After single cell sorting , PCR amplification of the AQP3 gene in clonal populations was used to screen for mutations in the genomic DNA of the cells , resulting in four AQP3 mutant cell lines ( AQP3mut1-4 ) . All four AQP3mut cell lines had confirmed mutations in the AQP3 gene and reduced P . berghei parasite load by approximately 80% compared to wild type HuH7 cells at 48 hpi ( S4A Fig ) . Fig 4 shows the data for two of these AQP3mut cell lines . Two gRNAs targeting exon 2 of the gene were used to generate AQP3mut1 and resulted in a 39 bp deletion ( S4B and S4C Fig ) . Because the mutation did not result in a frame-shift , mRNA was still transcribed , however , protein was not detected by IF staining ( Fig 4B ) , likely because of the inability of the protein to fold properly . Protein structure modeling of AQP3mut1 ( SWISS-MODEL ) suggests that the 39 bp deletion results in a nonfunctional channel that cannot transport nutrients ( S4D Fig ) . Separately , AQP3mut2 was generated with two gRNAs targeting exon 1 of AQP3 . mRNA transcripts and protein expression were not detected in this mutant cell line ( Fig 4A and 4B ) . AQP3mut1 and AQP3mut2 cells had significantly lower parasite load 48 hours after P . berghei infection , with 77 and 82% reduction in parasite load , respectively , compared to wild type HuH7 cells ( p < 0 . 001 , One-way ANOVA , Dunnet’s multiple comparison test; n = 3 independent experiments ) ( Fig 4C ) . While genetic disruption of AQP3 significantly decreased parasite load in the mutant hepatocytes , it did not influence hepatocyte viability or cell growth ( S4E Fig ) . The EEFs after genetic disruption of AQP3 in liver cells were analyzed by microscopy for possible defects in development . EEF size calculations were compeleted by widefield fluorescent microscopy of HuH7 infected cells . The size of the EEFs in wild type HuH7 cells and AQP3mut1 cells were measured at 24 , 36 , and 48 hpi . Significant changes in size were not detected at 24 hpi with both cell lines having roughly 40 μm2 EEFs on average ( Fig 4D ) ( p = 0 . 81 , unpaired Student’s t-test; n > 70 ) . However , by 36 hpi EEFs in wild type HuH7 cells were 2 . 9x larger than in AQP3mut1 cells , with EEFs averaging 112 and 40 μm2 , respectively ( p < 0 . 0001 , unpaired Student’s t-test; n > 85 ) . At 48 hpi EEFs in wild type HuH7 cells were 123 μm2 while EEFs in AQP3mut1 were 48 μm2 ( p < 0 . 0001 , unpaired Student’s t-test; n > 100 ) . Interestingly , EEFs in AQP3mut1 cells had no significant increase in size between 24 and 36 hpi and only had an 18% increase in size between 24 and 48 hpi . For comparison , EEFs in wild type HuH7 had a 218% increase in size between 24 and 48 hpi . Thus , parasite growth is stunted during the late liver stage development of P . berghei . While the size of EEFs was significantly reduced in AQP3mut1 cells , the infection rate , measured by EEFs/well , was not altered ( p = 0 . 165 , two-tailed Student’s t-test ) ( Fig 4E ) . Three independent experiments were performed and representative data is shown from a single experiment . Significant changes were not observed in any experiments . To probe if AQP3 is necessary for water permeability or glycerol transport we employed the gold-based compound [AuCl2 ( phen ) ]Cl ( auphen ) ( Fig 5A ) that selectively and irreversibly inhibits glycerol transport [32] , but only minimally affects water permeability of AQP3 at concentrations above 100 μM [32 , 33] . This functional selectivity is thought to be due to the interaction of Au ( III ) with a conserved cysteine residue at the selectivity filter of AQP3 that is not present in orthodox aquaporins [33] . Auphen was synthesized ( S5 Fig ) and found to inhibit liver stage parasite load with an EC50 = 0 . 77 ± 0 . 08 μM ( mean ± SEM; n = 3 independent experiments ) ( Fig 5B ) , when P . berghei infected HuH7 hepatocytes were treated at 0 hpi . Because of the abundance of AQP3 in erythrocytes [27] , we also tested auphen inhibition of P . falciparum Dd2-infected erythrocytes . Auphen treatment resulted in inhibition of P . falciparum with an EC50 of 0 . 81 ± 0 . 10 μM ( mean ± SEM; n = 2 ) ( Fig 5C ) , indicating that auphen is an effective inhibitor of multiple species and stages of the Plasmodium life cycle . We hypothesized that auphen inhibits parasite load during the liver stage by inhibiting the glycerol permeability of AQP3 . Most parasite functions that require glycerol , such as membrane and organelle synthesis , occur after 24 hpi , so we compared auphen potency with treatment at 0 and 24 hpi . Auphen treatment at 24 hpi yielded nearly identical EC50 values in both HepG2 and HuH7 cells when compared to cells treated at 0 hpi ( Figs 5B and S6A ) . Importantly , hepatocyte cell viability was not affected by auphen treatment up to 20 μM ( S6B Fig ) ( p = 0 . 165 , One-Way ANOVA; n = 3 independent experiments ) . We also wanted to examine the effects of auphen when only treating cells during the early liver stage . HuH7 cells were treated with auphen ( 0 . 05–20 μM ) at the time of infection and parasite load was quantified at 11 hpi , before the parasites begin to replicate , as well as after replication starts at 24 and 44 hpi ( S6C Fig ) . Parasite load was not found to be significantly inhibited when assessed at 11 hpi ( One-Way ANOVA , Tukey’s post hoc analysis , p > 0 . 05 ) even with 20 μM auphen treatment–the highest concentration tested . This observation indicates that auphen does not have inhibitory effects during the early stages of liver stage development . Additionally , by comparing the parasite load from early intrahepatic development ( 11 hpi ) to late development ( 44 hpi ) we observe that with high concentrations of auphen , parasite load is greater at early times when compared to later time points . This observation suggests that parasites that are unable to develop upon auphen treatment are cleared by the host cell . When the parasite load of auphen-treated cells was assessed at 24 hpi , inhibition was observed at 10 and 20 μM auphen , but this inhibition was substantially less than that observed at 44 hpi ( S6C Fig ) . Taken together , this time course study indicates auphen acts between 12 and 48 hpi to inhibit Plasmodium . The auphen mode of action was further assessed through a phenotypic study . Similar to AQP3mut cells , auphen significantly reduced the size of EEFs in hepatocytes compared to control DMSO-treated hepatocytes ( Fig 5D ) . HuH7 cells treated with 1 . 2 μM auphen developed EEFs 56 ± 0 . 08% smaller than DMSO-treated cells ( p < 0 . 0001 , unpaired Student’s t-test; n>100 ) . Furthermore , the number of EEFs in cells treated with 1 . 2 μM auphen was not significantly different from that of wild type cells ( Fig 5E ) ( p = 0 . 198 , two-tailed Student’s t-test; n = 9 ) . Additionally , to probe whether auphen acts on the host AQP3 and/or the parasite aquaporin , we pre-treated sporozoites with auphen prior to infection . Freshly dissected sporozoites were incubated with various concentrations of auphen or DMSO for 30 minutes . After the incubation , the sporozoites were pelleted and resuspended in media before being used to infect HuH7 cells . Treatment of sporozoites with 2–10 μM auphen did not significantly reduce parasite load in HuH7 cells when assessed at 48 hpi ( p = 0 . 926 , One-Way ANOVA; n = 2–3 independent experiments ) while treating cells with the corresponding concentration of auphen immediately after infection reduced parasite load by 83 . 9–99 . 5% ( p < 0 . 0001 , One-Way ANOVA; n = 2–3 independent experiments ) ( Fig 5F ) . Pretreatment of hepatocytes with auphen also does not result in parasite inhibition ( S6D Fig ) , which was expected as host AQP3 is not significantly expressed until after P . berghei infection . The effects of auphen on blood-stage Plasmodium parasites was further evaluated . Immunofluorescent staining of AQP3 in P . falciparum-infected erythrocytes was completed to confirm the localization of the protein to the PVM as previously described [22] . We found that AQP3 localizes to the erythrocyte membrane in uninfected cells , but is translocated to the PVM of P . falciparum-infected cells ( Fig 6A ) . Infected erythrocytes were visualized after treatment with 2 μM auphen or DMSO for 40 hours ( Fig 6B ) . Treatment began during the early schizont stage to assess the ability of the parasites to complete maturation and invade new erythrocytes . Across three independent experiments , parasitemia significantly decreased by 75 ± 12% when treated with 2 μM auphen ( Fig 6C ) compared to DMSO treated erythrocytes ( p = 0 . 0257 , unpaired Student’s t-test ) . Erythrocyte toxicity was not observed microscopically in response to auphen treatment . For one biological replicate of P . falciparum auphen treatment , the parasite stages ( ring , trophozoite , schizont ) were scored to assess their development in the newly invaded erythrocytes ( Fig 6D ) . Interestingly , the parasites that were able to infect erythrocytes after auphen treatment exhibited a developmental delay compared to the DMSO control as indicated by an increase in ring stage parasites and a decrease in trophozoites . Our initial RNA-seq screen enabled the identification and validation of AQP3 as an essential host protein for Plasmodium , highlighting the potential of our transcriptomic approach to discover new biology . To more broadly explore host pathways that may be critical to Plasmodium , we performed RNA-seq on additional time points throughout P . berghei infection of HepG2 and HuH7 hepatocytes . Infected cells were collected at early ( 2–12 hpi ) , mid ( 18–24 hpi ) , and late ( 36–48 hpi ) time points . Uninfected controls were also collected for both cell lines . In total , we performed RNA sequencing on 40 samples ( S4 Table ) . We observed a steady decrease in the proportion of reads mapping to the human transcriptome as a function of time after infection in both HuH7 and HepG2 cells ( Fig 7A ) . This decrease was expected due to the rising parasite occupancy of the host cell at later time points and reads that did not map to the human genome largely mapped to P . berghei . Principal component analysis ( PCA ) on the gene expression data showed HepG2 and HuH7 clustered separately , but such discreet clustering was not observed for individual time points within a cell line ( Fig 7B ) . We performed a differential gene expression analysis for the infected hepatocytes collected at early , mid , and late P . berghei infection compared to uninfected controls of the respective cell line ( S5 Table ) . DEGs ( adjusted p < 0 . 05 ) are shown for P . berghei-infected HuH7 cells ( Fig 7C ) . The majority of host genes that are differentially expressed were found to be associated with the early and mid-time points . Genes involved in cell cycle ( ARG1 ) , immune processes ( IL8 ) and the chemokine responses ( CXCL family of genes ) were among the most highly upregulated , while genes involved in transcriptional regulation and apoptosis ( MYCN ) were down-regulated . Other genes of interest to be upregulated included MTRNR2L1 and PHLDA1 , two anti-apoptotic factors . We then performed a gene set enrichment analysis using GSEA [34] to identify gene sets and pathways associated with P . berghei infection at various stages of infection ( S6 Table , S7 Fig ) . Interestingly , cholesterol biosynthesis was the most highly downregulated pathway during the early- and mid-time points following P . berghei infection ( Fig 7D ) . Conversely , the fatty acid , triacylglycerol , and ketone body metabolism pathway was significantly upregulated during the late stage of P . berghei hepatocyte infection ( Fig 7E ) . These data suggest that genes in the host hepatocyte are temporally regulated to first avoid clearance and prevent cell death after invasion , and then to ensure a favorable environment for growth and replication .
Elucidating the complex interplay between the Plasmodium parasite and its obligatory host liver cell is critical to understanding how a single sporozoite transforms into thousands of merozoites . Here we use the transcriptional changes that occur in HepG2 hepatocytes upon P . berghei infection to identify host factors that may be important for Plasmodium growth and development . We were particularly interested in genes highly induced upon infection , as the protein products of these genes could be utilized by the developing parasite . To test this possibility , we selected AQP3 , one of the most significantly differentially expressed genes in infected HepG2 cells ( p = 1 . 55 x 10−7 ) , for further study . Through an independent secondary study with qRT-PCR analysis , we confirmed that AQP3 mRNA levels increase in P . berghei-infected HepG2 and HuH7 hepatoma lines . Aquaporins are a family of water channel proteins that are found in most living organisms and are essential for the rapid movement of water to maintain cellular homeostasis [35] . AQP3 is an aquaglyceroporin , a unique aquaporin subtype that is highly permeable to both glycerol and water [36] , but is not generally expressed in hepatocytes at the protein level ( proteinatlas . org ) . Due to the selective induction of AQP3 , but not other proteins within the aquaporin family in response to Plasmodium infection , we predicted a critical role during the parasite’s liver stage as a nutrient source for the developing parasite . Using immunofluorescent microscopy , we show that host AQP3 protein is more abundant in Plasmodium infected hepatocytes compared to uninfected cells and that the protein localizes to the PVM . The PVM , which is derived from the host cell membrane , is modified by the loss of host proteins and the integration of secreted parasite proteins [10 , 37–39]; however , few host proteins are known to be trafficked to the PVM after Plasmodium infection of hepatocytes . Host autophagy markers , such as LC3 , surround the PV of infected hepatocytes where they can integrate into the PVM as a pathogen clearance mechanism by the host cell . In roughly half of infections , LC3 integration into the PV successfully recruits lysosomes that clear the parasitic infection [40] . Plasmodium infections that are not successfully cleared by the host cell are able to fully mature and eventually lose the LC3 localization around the PV . Unlike the recruitment of LC3 and other autophagy markers to the PVM , here we demonstrate the incorporation of host AQP3 into the PVM in a manner that is beneficial for the growth and development of the intrahepatic parasite . We also observe the localization of AQP3 to the PVM of P . falciparum-infected erythrocytes , as previously reported [22] . These data suggest a conserved role of human AQP3 in Plasmodium development in multiple stage of the parasitic life cycle as well as multiple species of Plasmodium . Furthermore , transcriptomic studies of mice chronically infected with T . gondii show a 17-fold increase in host AQP3 mRNA expression in brain cells [41] , suggesting the role of AQP3 could be conserved in other apicomplexans . The observations that AQP3 expression is induced in response to infection and that the protein localizes to the PVM , led us to hypothesize that AQP3 is essential for Plasmodium development during the liver stage . The importance of AQP3 for intrahepatic Plasmodium development is demonstrated by the significant decrease in P . berghei parasite load upon AQP3 gene depletion ( 4 targeting siRNA ) and genetic disruption ( 4 different CRISPR cell lines ) . Gene depletion of all other aquaporins expressed in hepatocytes indicate that AQP7 and AQP9 also contribute to Plasmodium development , although to a lesser extent than AQP3 . AQP7 and AQP9 belong to the aquaglyceroporin family of proteins , but their expression is not induced in response to Plasmodium infection . In contrast , gene depletion of orthodox aquaporins did not influence parasite load . The redundant function of AQP7 and AQP9 may be responsible for the 20% parasite load observed in the AQP3mut hepatocytes , where EEFs were significantly reduced in size . Taken together with our data that show AQP3 localizes to the PVM , our work suggests that AQP3 expression is co-opted by the Plasmodium parasite and trafficked to the PVM for survival . We predict that AQP3 is recruited and integrated into the PVM to enable nutrient transport . Plasmodium is an obligate intracellular pathogen that requires the host environment for development . In particular , the rapid replication of the parasite within the PV necessitates a high demand of nutrients for parasite membrane formation and organelle synthesis . All newly synthesized membranes of the developing parasite rely on glycerol as a precursor as it is the backbone of lipid and phospholipid synthesis [42 , 43] . Plasmodium can scavenge host lipids for the rapid assembly of membranes and organelles [44 , 45] . The host lipids are tethered to glycerol and subsequently incorporated into cell membranes as phospholipids . The parasite is also able to synthesize some nutrients de novo through the FASII pathway , a fatty acid synthesis pathway that is distinct from mammalian biosynthetic pathways [46] . The FASII pathway has been found to be especially important during the liver stage of the Plasmodium life cycle [47] and several components of this pathway require glycerol precursors that could be obtained from the host cell . The fatty acid products of the FASII pathway are then incorporated into essential lipids , such as glycerophospholipids and acylglycerols [48 , 49] . Thus , glycerol is an important component in both the de novo synthesis of fatty acids as well as the synthesis of plasma membranes from host-scavenged fatty acids . The importance of glycerol in parasite development has been demonstrated in P . falciparum glycerol kinase knockout parasites ( ΔPfGK ) [50] . Glycerol kinase is an enzyme that mediates the phosphorylation of glycerol to form glycerol-3-phosphate ( G3P ) , an essential component of glycerophospholipids . Upon PfGK deletion , parasite proliferation was reduced by approximately 45% , which corresponded to a 50% reduction in the incorporation of 14C-glycerol . This incorporation of 14C-glycerol from the host environment suggests unknown mechanisms of the parasite to utilize carbon sources other than glycerol kinase . Taken together , these studies show that nutrient requirements of the Plasmodium parasite are complex , requiring abundant resources from the host cell and that aquaporins are a protein family that may facilitate this process . The possible requirement of AQP3 for nutrient acquisition is particularly intriguing in light of a previous study that suggested the PVM is a molecular sieve for small molecule movement [51] . Using dyes of varying sizes , the selectivity filter of the PVM was proposed to be linked to molecular weight , where small molecules like glycerol are predicted to diffuse freely . Our observations suggest a more complex filtering of solutes occurs , which would enable greater control of the chemical components within the PV . The role of aquaporins in Plasmodium infections has been studied for many years , but the research has primarily focused on the parasite aquaporin . Plasmodium encodes an aquaglyceroporin , AQP [42] , which has been shown to localize to the parasite plasma membrane where it transports glycerol to the parasite [21] . Knockout of the parasite AQP significantly reduces the parasite growth rate but does not affect viability during the blood stage [21] . Based on these observations , it has been proposed that the parasite aquaporin is a potential drug target [52] . Our data suggest that human AQP3 is recruited to the PVM in hepatocytes to facilitate liver stage infection . We predict that AQP3 is important for the delivery of nutrients , possibly glycerol , through the PV , which enables the parasite AQP to then transport the nutrients into the parasitic cytoplasm to facilitate rapid growth . Depletion of AQP3 transcripts or genetic disruption of the gene in hepatocytes did not influence liver cell viability , which was expected as the protein is not generally expressed in hepatocytes and AQP3 knockout in mice has been previously shown to be non-lethal [24] . These observations suggest that a therapeutic window for host AQP3 inhibition may be found that reduces parasite load without inhibiting the host cell . To this end , we identified the AQP3-targeting auphen [32] as a compound that reduces parasite burden during the liver ( P . berghei ) and asexual blood stages ( P . falciparum ) of the parasite’s life cycle . The efficacy of auphen in inhibiting parasite growth in both hepatocytes and erythrocytes can likely be attributed to the conserved localization of AQP3 to the PVM in both the liver and blood stages . Interestingly , AQP3 is highly abundant in human erythrocytes where it is considered the main channel for glycerol transport [27] . Our data showing liver stage P . berghei and blood stage P . falciparum inhibition of parasite load by auphen suggests that AQP3 is a host factor that two Plasmodium species ( mouse- and human-infective ) rely on in different parasite stages . Inhibition of P . falciparum infected erythrocytes by auphen had similar phenotypic characteristics to liver stage parasite inhibition . When P . falciparum was treated with auphen for 40 hours starting at the early schizont stage , a significant reduction in parasite load was observed . We predict auphen’s inhibitory effects do not target parasite invasion , but likely prevents schizonts from properly developing mature merozoites capable of invading new erythrocytes . Though confirmation studies are necessary , this hypothesis is supported by the observation that parasites that have invaded new erythrocytes during auphen treatment exhibit delayed development into trophozoites . Auphen is a gold-based compound that inhibits glycerol transport of human AQP3 , but only minimally effects water and urea transport at high concentrations ( >100 μM ) [33] . It was first described for the inhibition of AQP3 , but has since been shown to inhibit AQP7 glycerol transport [53] . However , it does not inhibit other orthodox aquaporins , such as AQP1 [32] . The potency of auphen against liver stage parasites was the same when the drug was administered at 0 or 24 hpi ( EC50 ~ 0 . 8 μM ) , indicating the target was not important for invasion or early stage parasite development . Auphen has been tested in the context of cancers and shown to inhibit glycerol transport in erythrocytes with an EC50 = 0 . 80 ± 0 . 08 μM [32] , similar to our observed P . berghei inhibition in hepatocytes by auphen . During the first 12 hours of P . berghei infection of hepatocytes , the parasite undergoes morphological changes into the EEF and by 24 hours the volume of the EEF grows four-fold before it begins to divide [8] . The parasite then divides rapidly 24–55 hpi which requires large amounts of nutrients for the synthesis of membranes and organelles . Because auphen inhibits parasite development when administered at 24 hpi , it suggests that AQP3 is needed during the late stage of Plasmodium hepatocyte infection . This is also supported by our results showing that parasite load is not inhibited when it is assessed at 11 hpi after treatment with auphen at the time of infection . Additionally , incubating parasites with auphen prior to infection did not prevent invasion and replication . This lack of inhibition upon parasite pre-incubation suggests that auphen does not inhibit the parasite AQP to influence invasion under our experimental conditions , as the compound is known to be an irreversible inhibitor of aquaglyceroporins active site cysteine residues [33] . Unfortunately , the complementary host preincubation study to determine auphen specificity to AQP3 is complicated by the fact that the protein is only upregulated many hours after infection . Indeed , host cell pretreatment with auphen did not influence parasite load in liver cells , suggesting that the target is translated after infection . But we cannot rule out other potential parasite or host proteins that may be a target for auphen . However , the similarities in infection phenotypes between AQPmut cells and auphen treated cells support the targeting of host AQP3 by auphen . In order to obtain a more robust understanding of the host response to Plasmodium infection during the liver stage , we performed RNA-seq of an additional 28 samples of an alternate hepatocyte cell line , HuH7 , infected with P . berghei . Principle component analysis indicated that 91% of the variation observed within the 40 samples we collected could be attributed to different expression profiles of the two cell lines , HuH7 and HepG2 . In agreement with a past microarray study that reported changes to host hepatocyte transcriptome expression in response to Plasmodium infection [17] , we find that early time points ( 2–12 hours ) after P . berghei infection accounted for the majority of differentially expressed genes . Some of the most highly upregulated genes included anti-apoptotic MTRNR2L1 and PHLDA1 –an exciting finding as several studies have shown that apoptosis is inhibited during Plasmodium invasion of hepatocytes [54 , 55] . Gene set enrichment analysis also indicated that cholesterol biosynthesis genes are downregulated at early and mid-time points after infection , an interesting observation as it has been shown that Plasmodium relies on cholesterol acquisition from the host hepatocyte [56] . However , similar to studies that have shown an increase in cell metabolism during the blood stage [43 , 57] , we find that genes that are involved in metabolic pathways , such as fatty acid , triacylglycerol , and ketone body metabolism , are enriched among the genes upregulated upon P . berghei infection . The activation of this metabolic pathway during the late stage could suggest a breakdown of stored nutrients within the host liver cell that can then be utilized by the developing parasite . Future work dissecting the genes in these pathways will be critical to better understanding the host response to infection . Together our dataset of two hepatocyte cell lines , HepG2 and HuH7 , is a rich resource for identifying future areas of functional research in host-pathogen interactions during the liver stage of malaria . In this work , we have uncovered a strategy by which Plasmodium may secure small molecules in hepatocytes while remaining protected within the PV . Our study suggests that the parasites have a route to induce expression of host AQP3 , which is incorporated in to the PVM to ensure development . Further studies investigating metabolism within the PV will be critical for testing if AQP3 is importing glycerol as a nutrient source for the parasites , as it remains possible that the parasites rely on AQP3 to maintain homeostasis or remove toxic waste . However , no strategies currently exist to separate viable parasites in the PV from the host liver cell and the low infection rate of hepatocytes hinders metabolic labeling studies . The identification of AQP3 and other potential host factors that are differentially regulated throughout the liver stage of malaria has implications for understanding host-parasite interactions . Understanding these interactions may in turn enable the development of prophylactic measures to prevent malaria . Thus , elucidating how Plasmodium alters host gene expression can further enhance our understanding of the host-pathogen processes that enable the parasites to cause disease . To that end , we have generated a comprehensive data set of DEGs at early , mid and late P . berghei infection of hepatocytes . We believe this data will be a valuable resource for further uncovering interactions between the Plasmodium parasite and its host liver cell .
HepG2 were purchased from ATCC and HuH7 cells were a kind gift from Dr . Peter Sorger ( Harvard Medical School ) . Hepatocytes used for P . berghei infections were maintained in Dulbecco’s Modified Eagle Medium ( DMEM ) with L-glutamine ( Gibco ) supplemented with 10% heat-inactivated fetal bovine serum ( HI-FBS ) ( v/v ) ( Sigma-Aldrich ) and 1% antibiotic-antimycotic ( Thermo Fisher Scientific ) in a standard tissue culture incubator ( 37°C , 5% CO2 ) . Anopheles stephensi mosquitoes infected with luciferase- or GFP-expressing P . berghei ANKA were purchased from the NYU Langone Medical Center Insectary Core Facility . Sporozoites used for liver stage experiments were harvested from freshly dissected salivary glands of mosquitoes . P . falciparum parasites of the 3D7 strain ( ATCC ) were maintained in red blood cells ( Golf Coast Regional Blood Center ) cultured in RPMI 1640 medium supplemented with 0 . 5% ( m/v ) AlbuMAX II , 25 mM HEPES , 25 ug/mL gentamycin , 24 mM sodium bicarbonate and 50 μg/mL hypoxanthine at a pH of 7 . 2 and maintained at 37°C under 92% N2 , 5% CO2 , and 3% O2 . Synchronization was performed using 5% sorbitol as previously described [58] . T25 flasks were seeded with 3x105 HepG2 or 8x104 HuH7 cells . Cells were infected with 1x105 GFP-expressing P . berghei-ANKA sporozoites 24 hours after seeding . Infected cells and uninfected controls were sorted at various times post-infection directly into lysis buffer using the BD FACSAria II cell sorter ( BD Biosciences ) at the Duke Human Vaccine Institute . Sytox blue was used as a live/dead cell indicator ( Thermo Fisher Scientific ) . RNA was extracted using SMART-seq v4 Ultra Low Input RNA Kit for Sequencing ( Clonetech ) and libraries were prepared at the Duke Next Generation Sequencing Core Facility and sequenced on the Illumina HiSeq 4000 as 50 base pair single-end reads . FASTQ files comprising reads from RNA-seq were tested for quality using FastQC v . 0 . 11 . 5 [59] . Adaptor sequences and low quality reads were trimmed using Trimmomatic v . 0 . 36 [60] . Next , alignment was performed using the STAR v . 2 . 5 . 1a aligner [61] to map reads to the reference Homo sapiens GRCh37 . 82 genome . PCR duplicates were marked using Picard v . 2 . 8 . 2 ( http://broadinstitute . github . io/picard/ ) . The final output was a matrix of read counts per transcript . We used DESeq2 [62] to normalize the read count matrix , and to perform differential analysis . Geneset enrichment analysis was performed using GSEA [34 , 63] . All statistical analyses and plots were generated using R v3 . 4 . 3 . HepG2 or HuH7 cells infected with GFP-expressing P . berghei ANKA were sorted along with uninfected controls at 48 hpi directly into lysis buffer . RNA from samples were extracted using the Quick-RNA microprep ( Zymo ) according to manufacturer’s protocol . cDNA was synthesized using GoScript reverse transcriptase ( Promega ) according to manufacturer’s protocol . Relative abundance of H . sapiens AQP3 was quantified using a SYBR Green Light Cycler 480 ( Roche ) according to manufacturer’s protocol in a 96-well plate . Relative abundance of AQP3 expression in P . berghei-infected cells was compared to uninfected cells and normalized to the H . sapiens 18S housekeeping gene ( primers listed in S1 Table ) . qRT-PCR was run in technical triplicates . RNA was isolated for 3 biological replicates for HepG2 cells and 4 biological replicates for HuH7 cells . HepG2 and HuH7 cells ( 2 . 5x105 ) were seeded on cover slips in 24-wells plates and infected with P . berghei with a multiplicity of infection ( MOI ) of 1:4 . Cells were fixed at various times post infection in 4% paraformaldehyde for 20 minutes at room temperature . Cells were washed three times in PBS and were subsequently blocked with 3% BSA for 45 minutes at room temperature . Cells were incubated in primary antibodies , rabbit-anti Pb heat-shock protein 70 [64] , rabbit anti-HsAQP3 ( Rockland ) , and goat anti-UIS4 ( LSBio ) that were diluted 1:100 , 1:300 , and 1:1 , 000 , respectively , for 1 . 5 hours at room temperature or 4°C overnight . Cells were washed three times with PBS and incubated in secondary antibodies , AlexaFluor 488 goat anti-mouse and AlexaFluor 568 donkey anti-rabbit diluted 1:400 ( Thermo Fisher Scientific ) , for 1 hour at room temperature . For UIS4 , staining AlexaFluor 488 donkey anti-goat was used at 1:5 , 000 . Cells were washed with PBS and incubated with 0 . 5 μg/mL DAPI for 7 minutes . Cells were washed once more with PBS and slides were mounted with ProLong Gold Antifade ( Thermo Fisher Scientific ) . iRBCs samples were fixed , permeabilized and blocked as described previously [65] . Incubations with anti-HsAQP3 primary antibody and secondary antibody were completed as described for P . berghei . After nuclear staining in 1 μg/mL Hoechst 33342 , cells were imaged on a widefield fluorescent microscope . Images were taken using the Zeiss Axio Observer wide field fluorescence microscope or the Zeiss 880 Airyscan inverted confocal . 3D stacks were acquired on a Zeiss LSM 510 inverted confocal microscope with a 100x 1 . 4 NA Plan-Apochromat oil objective . 3D images were constructed and colocalization parameters were determined using Imaris software ( Bitplane ) . A minimum of three independent experiments were completed for all localization studies . EEF size quantification was completed using ImageJ . Number of EEFs for AQP3mut cells and auphen treated cells were quantified/well in 384-well plates . Three independent experiments were performed with 5–10 wells/condition in each experiment . Parasites isolated from salivary glands of infected A . stephensi were used to infect HuH7 and HepG2 hepatocytes as previously described [66] . Cell viability and parasite load were assessed using CellTiter-Fluor ( Promega ) and Bright-Glo ( Promega ) reagents , respectively , according to manufacturer’s protocols . Fluorescence and luminescence readouts were collected using the EnVision plate reader ( Perkin Elmer ) . For inhibition assays , auphen was synthesized [67] and 20–0 . 01 μM of compound was added to P . berghei ANKA-infected HuH7/HepG2 cells or P . falciparum Dd2-infected erythrocytes 10 minutes post infection in a dose-dependent manner with DMSO normalized to 1% in all wells . P . berghei parasite load upon auphen treatment was quantified using the luciferase reporter as described above in biological triplicate . P . falciparum Dd2 parasite load was assessed with DAPI nuclear staining as previously described in biological duplicate [66] . For auphen pre-incubation studies , freshly harvested sporozoites from mosquito salivary glands were incubated with 2–10 μM auphen for 30 minutes at room temperature . As a control , sporozoites were also incubated with DMSO and all treatments had 1% DMSO . After incubation , the compound was removed by centrifugation and resuspension of sporozoites in untreated media . Cells were pretreatment with the same concentrations of auphen . After a 30 minute incubation with auphen or DMSO control ( 1% ) , media was gently aspirated and fresh media was added to the wells . For auphen treatment of P . falciparum , synchronized iRBCs in the schizont stage were treated with 2 μM auphen for 40 hours . After 40 hours , Geimsa ( Sigma-Aldrich ) stained blood smears were imaged on a brightfield microscope . Parasitemia was calculated by counting the number of P . falciparum infected red blood cells out of a total of 1 , 200 cells . HuH7 cells were reverse transfected with 50 nM siRNAs ( S1 Table ) at 2x103 cells/well in 384-well plates . siRNAs ( Qiagen ) were transfected using Lipofectamine 2000 ( Thermo Fisher Scientific ) according to manufacturer’s protocol . Cells were reverse transfected in 6 technical replicates and in duplicate plates . One plate was used to assess RNA knockdown efficiency 48 hours post transfection using Quick-RNA MicroPrep ( Zymo ) . cDNA was generated using GoScript ( Promega ) and qRT-PCR was performed as previously described to assess the knockdown of genes of interest compared to mock-transfected cells . Primers that were used for the qRT-PCR are listed in S1 Table . 48 hours post transfection the duplicate plate was infected with 4x103 luciferase-expressing P . berghei ANKA sporozoites per well . 48 hpi cell viability and parasite load were assessed as described above . Each experiment was completed with 6 technical replicates and three independent experiments were performed . Four separate AQP3mut cell lines were generated using different guide RNAs ( gRNA ) . The gRNA sequences were determined ( crispr . mit . edu ) for recruitment of Cas9 to either the first or second exon of AQP3 and are listed in S3 Table . Tails were added to gRNA sequence for introduction into px33034 ( Addgene plasmid #42230 ) , a plasmid containing a human codon-optimized SpCas9 . px330 was digested using BbsI ( NEB ) for one hour according to manufacturer’s protocol and gel purified . Reverse complements of gRNA were annealed using T4 ligation buffer ( NEB ) . Linearized px330 and gRNA were ligated using T4 ligase ( NEB ) according to manufacturer’s protocol . Ligated plasmid was transformed into XL-Gold competent cells and plasmids were isolated using a standard mini-prep kit ( Qiagen ) . Sequencing was verified using Eton BioScience Inc . HuH7 cells ( 2x105/per well ) were seeded in a 6-well plate in DMEM . Cells were co-transfected with pX330 , containing the gRNA sequence , and pCDNA3 . 0 containing a blasticidin resistance marker using Lipofectamine 2000 ( Thermo Fisher Scientific ) according to manufacturer’s instructions . Cells that were transfected with gRNAs targeting exon 2 were transfected with two separate gRNA-containing px330 plasmids , homologous to the second exon of AQP3 and 6 nucleotides away from one another . Mock cells containing no plasmids were included . HuH7 cells containing plasmids were then trypsinized and replated into 10 cm tissue culture plates in cDMEM 24 hours after transfection . Cells were washed and then cDMEM+5 μg/mL blasticidin was added 48 hours after transfection . Mock cells were monitored for cell death daily until no surviving cells remained . Transfected HuH7 cells were then single-cell sorted into 96-well plates and maintained in standard culture medium as described above . Populations were grown from single cells and expression of AQP3 and Actin RNA was assessed by RT-PCR ( AQP3 F: ATGGGTCGACAGAAGGAGCT , R:TCAGATCTGCTCCTTGTGCTT; ACTB F: GCCTCGCCTTTGCCGA , R: GTTGAAGGTCTCAAACATGATCTGG ) . RNA was extracted and cDNA synthesized as previously described and genomic DNA ( gDNA ) was purified using the Quick-DNA microprep ( Zymo ) . Primers to amplify both exons of the AQP3 gene were designed as well as for the full-length AQP3 mRNA ( S1 Table ) . Clonal populations showing no expression of the full-length transcript were selected for further evaluation and designated as AQP3mut cell lines . Parasite load in the AQP3mut cell lines infected with luciferase- or GFP-expressing P . berghei were compared to wildtype HuH7 cells as previously described . Reactions were carried out under a nitrogen atmosphere with dry solvents and oven-dried glassware under anhydrous conditions unless specified otherwise . Reagents were purchased at the highest commercial quality and used without further purification , unless otherwise stated . Yields refer to chromatographically and spectroscopically ( 1H and 13C NMR ) homogeneous materials , unless otherwise stated . NMR spectra were recorded on a 400 MHz Varian Inova spectrometer instrument , and were calibrated using residual undeuterated solvents as internal reference ( dimethylsulfoxide , δ = 2 . 50 ppm , 1H NMR; 39 . 52 ppm , 13C NMR ) . Chemical shifts ( δ ) are reported in parts per million ( ppm ) ; NMR peak multiplicities are denoted by the following abbreviations: s = singlet , d = doublet , t = triplet , q = quartet , p = pentet , dd = doublet of doublets , dt = doublet of triplets , m = multiplet , br = broad; coupling constants ( J ) are reported in Hertz ( Hz ) . [AuCl2 ( phen ) ]Cl ( Auphen ) was prepared as previously described [67] . Briefly , to a solution of HAuCl4●3H2O ( 100 mg , 0 . 254 mmol ) in EtOH ( 1 . 0 mL ) a solution of 1 , 10 phenanthroline ( 151 mg , 0 . 838 mmol ) in EtOH ( 1 . 0 mL ) was added slowly . The reaction was stirred at reflux for 4 hours . The reaction was cooled to room temperature , filtered and washed with cold EtOH ( 3 x 5 mL ) , affording pure auphen as orange crystals ( 116 mg , 0 . 240 mmol , 95% ) . 1H NMR ( 400 MHz , DMSO-d6 ) : δ 9 . 31 ( dd , J = 4 . 9 , 1 . 6 Hz , 2H ) , 9 . 08 ( dd , J = 8 . 2 , 1 . 5 Hz , 2H ) , 8 . 37 ( s , 2H ) , 8 . 23 ( dd , J = 8 . 2 , 4 . 9 Hz , 2H ) ppm; 13C NMR ( 100 MHz , DMSO-d6 ) : δ 147 . 60 , 141 . 94 , 137 . 37 , 129 . 55 , 127 . 53 , 125 . 73 ppm . | Plasmodium parasites undergo an obligatory morphogenesis and replication within the liver before they invade red blood cells and cause malaria . The liver stage is clinically silent but essential for the Plasmodium parasite to complete its life cycle . During this time , the parasite relies on the host cell to support a massive replication event , yet host factors that are critical to this expansion are largely unknown . We identify human aquaporin-3 ( AQP3 ) , a water and glycerol channel , as essential for the proper development of the parasite within the liver cell . AQP3 localizes to the parasitophorous vacuole membrane , the interface between the host cytoplasm and the parasite , possibly aiding in the nutritional uptake for the parasite . Genetic disruption or treatment with the AQP3 inhibitor auphen , reduces parasite load in liver and blood cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"liver",
"plasmodium",
"gene",
"regulation",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"parasitology",
"developmental",
"biology",
"apicomplexa",
"protozoans",
"red",
"blood",
"cells",
"monomers",
"(chemistry)",
"small",
"interfering",
"rnas",
"polymer",
"chemistry",
"malarial",
"parasites",
"animal",
"cells",
"gene",
"expression",
"hepatocytes",
"life",
"cycles",
"chemistry",
"biochemistry",
"rna",
"eukaryota",
"cell",
"biology",
"anatomy",
"nucleic",
"acids",
"glycerol",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"non-coding",
"rna",
"organisms",
"parasitic",
"life",
"cycles"
] | 2018 | Plasmodium parasite exploits host aquaporin-3 during liver stage malaria infection |
The ability to generate new meaning by rearranging combinations of meaningless sounds is a fundamental component of language . Although animal vocalizations often comprise combinations of meaningless acoustic elements , evidence that rearranging such combinations generates functionally distinct meaning is lacking . Here , we provide evidence for this basic ability in calls of the chestnut-crowned babbler ( Pomatostomus ruficeps ) , a highly cooperative bird of the Australian arid zone . Using acoustic analyses , natural observations , and a series of controlled playback experiments , we demonstrate that this species uses the same acoustic elements ( A and B ) in different arrangements ( AB or BAB ) to create two functionally distinct vocalizations . Specifically , the addition or omission of a contextually meaningless acoustic element at a single position generates a phoneme-like contrast that is sufficient to distinguish the meaning between the two calls . Our results indicate that the capacity to rearrange meaningless sounds in order to create new signals occurs outside of humans . We suggest that phonemic contrasts represent a rudimentary form of phoneme structure and a potential early step towards the generative phonemic system of human language .
The vast lexicons that characterise human languages are the product of physical and cognitive processes that guide the combination of a limited number of meaningless sounds ( phonemes ) in a variety of ways to generate new meaning [1 , 2] . In a simple example , the phonemes /k/ , /æ/ and /t/ can be rearranged in different ways to create the words cat [kæt] , act [ækt] or tack [tæk] [1] . Alternatively , the phoneme /k/ from the word cat can be eliminated to create the word at [æt] , with the first position ( i . e . , presence or absence of the phoneme /k/ ) representing a phonemic contrast that generates the differentiation in meaning [3] . In all four arrangements , the meaningless phonemes maintain their acoustic identity across words , and this , paired with the arbitrary relationship between phoneme structure and word meaning , results in words with shared phonemes having distinct semantic content [4] . Such phoneme structure is a basic ingredient of word generation in human language , and when combined with the rules governing assemblages of meaningful words ( a syntactic layer ) , provides much of language’s generative power [5–7] . Despite the crucial role that phoneme structure plays in language , little is known about how such a capacity might have evolved [8–11] . Whilst comparative data from animal communication systems can elucidate early forms of language components , data demonstrating the critical rudiments of phoneme structures outside of humans is lacking . Evidence that animals can employ a basic syntactical layer of language in their communication system has been provided in nonhuman primates . For example , Campbell’s monkeys ( Cercopithicus cambelli ) produce two predator-specific alarm calls that are each modified in a predictable way into more general disturbance calls upon addition of the same suffix [12 , 13] . However , because the constituent calls are themselves meaningful ( with the suffix carrying an abstract meaning in this case [14] ) , this , and equivalent findings [15 , 16] , do not exemplify phoneme structure . Several candidates of phoneme-like structures in nonhuman animals have been proposed , but defining features are either lacking or have yet to be demonstrated [8 , 11 , 17] . One set of contenders comes from the songs of birds and mammals , in which meaningless elements are combined to create complex , higher-order structures [11 , 18 , 19] . However , experiments investigating behavioural responses to element reorganisation within songs are either lacking [18–21] or have not shown that such reorganisation confers a qualitative change in contextual meaning [22–24] . Another set includes calls produced in movement and alarm contexts . For example , parid birds can produce variable vocal sequences of apparently meaningless acoustic elements . However , in these cases , although call elements are commonly repeated or omitted , the required association between sequence structure and qualitative changes in informational content has not been demonstrated [25–29] . Using acoustic analysis , natural observations , and controlled playback experiments we provide evidence for rudimentary phoneme structure in the calls of chestnut-crowned babblers ( Pomatostomus ruficeps ) ( see Materials and Methods ) , a 50 g , highly social , cooperatively breeding bird [30 , 31] . Observations over the past 10 years suggest that the repertoire of adult chestnut-crowned babblers consists of at least 15 discrete , context-specific vocalizations , of which three pairs appear to share sound elements , with the reused elements in each case being restricted to a specific pair of calls [32] . Here , we specifically focused on a single pair: a double-element call produced during flight ( flight calls , elements F1 and F2 ) and a triple-element call produced during nestling provisioning [33] ( prompt calls , elements P1 , P2 , and P3 ) ( Fig 1A ) . Importantly , the constituent elements within these calls appear to be contextually meaningless . For example , none of the elements is used as an individual call in isolation , suggesting that none can function to confer contextual information . Additionally , because none is used in combination with other call types , they cannot clearly operate to modify calls in a predictable way , as would be required of affixes [13] . First , we establish , using acoustic analyses , that the two calls comprise statistically equivalent acoustic elements . Second , we present natural observations showing that the two calls are context-specific , a prerequisite of reliable information transfer in animals . Finally , playbacks of natural , switched-element , and artificial calls in a standardised aviary environment confirm that the call elements are perceptibly equivalent and that element addition/elimination at one position creates a phoneme-like contrast , yielding the functional changes in meaning .
Acoustic analyses were conducted to test whether prompt and flight calls are composed of statistically indistinguishable acoustic elements . To avoid problems of pseudo-replication arising from using calls of genetic relatives within groups [34] , we analysed a single flight call and a single prompt call per group recorded ( n = 23 flight , 11 prompt calls ) . Five parameters were extracted from the fundamental frequency of the resulting 79 elements: start and end frequency , frequency range , time to peak frequency , and element duration ( S1 Text , S1 Table , and S2 Table ) . A Discriminant Function Analysis ( DFA ) demonstrated that the five elements across the two calls comprised just two independent acoustic structures ( Fig 1B ) . Mahalanobis distances generated from the DFA revealed that F1 and P2 could not be reliably distinguished and neither could F2 , P1 , and P3 ( all p values > 0 . 2 ) , but that F1 and P2 could be distinguished easily from F2 , P1 , and P3 ( all p values < 0 . 001 ) ( Fig 1B and S3 Table ) . Thus , the two calls appear to comprise the same two distinct elements , with flight and prompt calls displaying AB and BAB patterns , respectively . Natural observations were conducted to quantify the context in which flight and prompt calls are produced . Natural flights were accompanied by flight calls in 274 of 450 observations ( 61%; n = 6 groups , 1 h/group ) , with all flights being short , low , and easily quantified . Similarly , hand-held releases following capture induced flight calls in 58 of 90 occasions ( 64% , n = 24 groups ) . No prompt calls were recorded in either set of observations , and flights/releases lacking flight calls were either silent or associated with alarm calls in response to observer presence . Finally , recordings from within nests in conjunction with automated nest entry-exit recorders revealed that 62% flights to/from nests were accompanied by flight calls ( n = 140 visits , 7 groups ) but rarely prompt calls ( 0 . 08% of nest visits ) , while 70% of nestling provisioning events were associated with prompt calls ( n = 140 visits , 7 groups ) and rarely flight calls ( 0 . 03% of nest visits ) . Additionally , in 97% of nest visits in which both flight and prompt calls were recorded , individuals used only flight calls travelling to/from the nest and only prompt calls within nests ( n = 60 visits , 7 groups ) . Thus , flight and prompt calls are highly context-specific , with the former maintaining group cohesion during movement [35] and the latter increasing the efficiency of food transfer to offspring by stimulating begging [33] . To verify experimentally that flight and prompt calls are context specific and are generated from rearrangement of the same acoustic elements , we performed playback experiments on 16 birds captured from 7 groups during periods of breeding . Each of the 16 birds received six playback trial-sets presented in a randomised order . Behavioural responses to two natural , two switched-element and two artificial calls were recorded in aviary compartments ( 2 x 2 . 5 x 2 m l x b x h ) containing natural perches , foraging substrate , a view to the outside , and a recently used babbler nest ( 30x45 cm dome-shape , 6 cm diameter entrance ) ( Fig 2 ) . The playback speaker was positioned out of view in a neighbouring compartment; birds had to look perpendicular to the speaker to look outside the aviary and in the opposite direction to look at the nest ( S2 Text ) . Given our natural observations , we predicted flight calls would elicit increased observations to the outside and increased movement in anticipation of an incoming bird , while prompt calls would provoke greater nest attentiveness . Combined , these three behaviours comprised 61% of the activity budget in each trial ( SD [standard deviation] = 23%; correlation coefficients among these behaviours ranged from +0 . 1 to -0 . 3 , indicating that time spent in one activity did not preclude time available for another ) . Compared with natural prompt calls , natural flight call playbacks were associated with a 49% increase in the proportion of time spent looking outside ( Generalized Linear Mixed Model [GLMM]: χ21 = 11 . 8 , p < 0 . 001 ) and a 36% increase in time spent hopping/flying between perches ( χ21 = 6 . 5 , p = 0 . 02 ) . By contrast , during natural flight call playbacks , individuals spent 81% less time looking at the nest ( 2% of monitoring time ) than during prompt call playbacks ( 15% of time ) ( χ21 = 11 . 6 , p < 0 . 001 ) ( Fig 3 ) . Together , these results confirm the two calls are distinct and encode perceptible , context-specific information . To test whether unmeasured acoustic variation dissociates the two calls [15] , we played back switched-element versions of both calls to all 16 birds by generating flight calls from prompt elements P2 P3 and prompt calls from elements P1 F1 F2 . The proportion of time birds spent engaged in the three behaviours of functional relevance were statistically equivalent between natural and switched-element flight calls ( GLMM: all p values > 0 . 6; Fig 4A ) as well as between natural and switched-element prompt calls ( all p values > 0 . 3; Fig 4B ) . Additionally , there were no significant interactions between call type ( flight versus prompt ) and whether or not calls were natural or switched-element on behavioural responses ( GLMM: all p values > 0 . 4 ) . The absence of such interactions generated differences in behavioural responses to switched-element flight versus switched-element prompt calls of similar magnitude to those found in comparisons of natural calls ( see Fig 4C versus Fig 3 ) . Compared with switched-element prompt calls , switched-element flight calls were associated with 33% more time looking out , 33% more time in-movement , and 80% less time looking at the nest . Accordingly , it is improbable that any unmeasured acoustic differences between the elements of flight and prompt calls are responsible for the distinct responses , reinforcing our acoustic analyses that the calls comprise the same sound elements . The results above suggest that the meaning-differentiating element between the two calls is P1 . Before a phonemic-like system can be supported , two other interpretations require testing . First , element P1 might , by itself , be responsible for generating the contextual information carried by the prompt call , in which case , our results could be more akin to a syntactic , rather than phonemic , communicative system [12 , 13] . Second , the differences in response to flight calls versus prompt calls might arise from their differences in element number [36] . In this case , our results could represent stimulus intensity effects ( triple-element prompt versus double-element flight call ) or priming effects [12] ( any acoustic element preceding a flight call results in a prompt-type response ) . To test these alternative interpretations , we presented two artificial stimuli to the 16 birds: element P1 alone and CAB , with the latter representing call elements P2 P3 ( i . e . , AB ) preceded by an element ( C ) from chatter calls , a common call naturally repeated in mixed-element bouts and associated with excitement [32] . These two artificial stimuli elicited similar behavioural responses ( all p values > 0 . 2; Fig 5A ) . First , they both generated relatively high look out and movement responses . One explanation lies with the fact that each is unnatural: impossible vocal scenarios have been shown to increase attentiveness behaviour in other contexts [37 , 38] . In support , separate analysis of the proportion of time spent looking around the aviary showed that general attentiveness behaviour during natural flight playbacks ( mean ± SE = 16% ± 4% ) was 36% , 47% , and 48% lower than during playbacks of CAB , P1 , and natural prompts , respectively ( GLMM: χ23 = 10 . 6 , p = 0 . 01 ) . Second , and more crucially , neither the P1- nor the CAB-stimulus elicited a hint of an elevated response in nest-attentiveness ( Fig 5B ) . Like the flight call , P1 element and CAB playbacks were both associated with ca . 80% reductions in nest-attentiveness behaviour over natural prompt calls ( Fig 5B ) . That neither the P1 element alone nor CAB elicits any increase in nest-attentiveness confirms that ( a ) P1 does not carry any nest-associated information in isolation and ( b ) differential nest-attentiveness responses to flight and prompt calls are not derived from either stimulus intensity or priming effects . Thus , it is the presence or absence of element P1 from the P2 P3 ( or F1 F2 ) call sequence that appears integral to generating the qualitatively distinct meaning carried by the two calls .
Phoneme structure represents a critical component of the vast lexicons in human languages , but a lack of suitably comparable evidence in animals has hindered our understanding of candidate selection pressures on , and early forms of , phoneme structure . Two related hypotheses have been proposed to explain the emergence of phonemic systems; both advocate a role of selection acting on increasing the capacity of vocal communication beyond that currently possible under an existing vocal repertoire . The here-named “enhanced-perception hypothesis” proposes that stringing together existing sounds in new ways reduces perception errors over the generation of new , but similar , sounds [39–41] . By contrast , the “vocal-constraints hypothesis” proposes that when the generation of new sounds is constrained [42] , reusing pre-existing sounds in new combinatorial forms can provide an alternative solution to increasing communicative output [15 , 16] . Testing the predictions arising from these hypotheses represents a major challenge because human languages are generally too derived to address the pressures selecting for their emergence . Additionally , testing whether animals make perceptual mistakes for sounds that do not exist or are vocally constrained will be rarely feasible . A necessary first step in elucidating the pressures selecting for , and early forms of , phonemic structure is to address whether animals possess the capacity for generating functionally distinct vocalizations by rearranging contextually meaningless elements , and how such rearrangements are manifest . Here , using acoustic analyses , natural observations , and playback experiments , we reveal that chestnut-crowned babblers use two acoustic elements ( A and B ) in different arrangements to create two functionally distinct vocalizations: flight calls ( F1 F2 , or AB ) and prompt calls ( P1 P2 P3 , or BAB ) . The meaning differentiation between the two calls is not a result of the different number of elements or priming effects , but specifically the presence or absence of P1 ( element B ) at the head of the same call sequence . The fact that element P1 is both contextually meaningless on its own and meaning differentiating when used in combination with elements P2 ( F1 ) and P3 ( F2 ) signifies a phoneme-like contrast , with element B used in this position likely representing a phoneme-equivalent . To our knowledge , this is the first demonstration that animals have the basic capacity to use phoneme-like contrasts to derive qualitatively new meaning , a basic component of phoneme structuring . However , whether or not our results can also be interpreted as providing evidence for more advanced forms of phoneme structuring in an animal depends on two critical features . First , in human languages , phoneme structure has potentially boundless generative power: the sum of derivable information is substantially greater than the number of its phonemic parts [1] . In contrast , the babbler vocal system that we describe is strictly bounded in its generative nature ( i . e . , two elements generate only two distinct calls ) . Part of the difference in human versus any nonhuman phonemic system will inevitably arise from vast differences in cognitive capacity [9] . Notwithstanding , cognitive capacity alone does not appear to be sufficient to explain differences in phonemic complexity and boundedness . For example , the sign language of the Al Sayyid Bedouin , an emerging language shared by deaf and hearing people of a small Israeli village , has been shown to have a fully functional and productive syntactic layer , but is so far characterized by only one phonological form [43 , 44] . Thus , when a phonemic layer emerges , even in human language , it appears initially to be finite and strongly bounded . This evidence suggests that the use of phonemic structure in communication should not be defined a priori by its complexity or boundedness , for it is likely that all phonemic systems evolve from simple beginnings like the one we describe here . Second , the level of phonemic complexity used by babblers depends on the number of phoneme-equivalent entities in use . For example , whilst babblers generate a phonemic contrast by inserting the phoneme-like entity P1 before P2 ( F1 ) and P3 ( F2 ) , whether or not P2 and F1 or P3 and F2 also represent phoneme-equivalent entities in the linguistic sense is equivocal . Unlike combinatoriality based on affixation rules or the generation of idioms , in which constituent parts have meaning [12 , 16] , definitively testing whether all sound elements within call sequences of animals are contextually meaningless , and yet individually perceptible and meaning-differentiating , will be a major challenge . This is because any sound uttered by a conspecific can lead to a behavioural response irrespective of any perception of contextual meaning [38] , and their limited vocal repertoires preclude investigation of whether distinct functional meaning is derived from the same meaningless elements in multiple different arrangements . A key component in discerning whether F1/P2 and F2/P3 are also phonemic depends on whether they represent a compound of two discrete elements , perceptible independently ( i . e . , A and B ) , or a holistic unit ( i . e . , AB ) . That the B element is phonemic in position P1 hints that AB is reducible , and hence F1/P2 and F2/P3 are probably also phoneme-like . However , this is an untested hypothesis at this stage , and we do not wish to speculate on whether chestnut-crowned babblers use more advanced forms of phoneme structure , beyond the identified use of a simple contrast , as part of their communication system . Either way , we propose that the bounded use of phoneme-like contrasts in the vocal system of chestnut-crowned babblers represents a simple precursor of phoneme structuring that can elucidate how early forms of phonemic systems might emerge . For example , our results lead to the hypothesis that the addition or elimination of elements , i . e . , basic phonemic contrasts ( e . g . , /kæt/ versus /æt/ ) , might represent a simpler evolutionary step than complete element rearrangement ( e . g . /kæt/ versus /tæk/ ) , due to its reduced structural complexity . However , generating distinct contextual meaning through the former rather than latter process is likley to be more prone to perception errors , because it results in higher acoustic similarity . That babblers have opted for the more error-prone means of generating functionally distinct vocalizations , and done so by adding or eliminating a common element , is more supportive of a vocal-constraints hypothesis [15 , 16] than an enhanced-perception hypothesis [39–41] . Limiting the use of phonemic contrasts to short-range calls used in low-urgency , social contexts might be one way of reducing perception errors and mitigating associated costs when vocal constraints are operating . In conclusion , the salient message here is that the basic capacity to generate qualitatively new meaning from rearranging contextually meaningless elements appears to exist outside of humans . One explanation is that for vocally constrained , highly social species , such as chestnut-crowned babblers , evolving new meaning by rearranging existing sounds offers a faster route to increasing communicative output than evolving new sounds . We hypothesise that reusing acoustic elements has facilitated the emergence of phoneme-like contrasts , which potentially drove sensitivity to phoneme structure or “phonemic awareness” in receivers [45 , 46] . The capacity to recognise vocalizations as sound constructs composed of smaller , meaningless elements , instead of a holistic unit , may have been the first step in the emergence of the elaborate phonemic systems seen in human languages . Further experiments are now required to determine exactly how babblers compute and perceive the elements from the two calls . More generally , further evidence for the use and manifestation of phonemic systems in animals is required; we propose that such systems will be most operant in the short-range communication of vocally constrained , social animals .
Ethics approval was provided by Macquarie University , Sydney , Australia ( Number ARA 2013/025 ) . The study was conducted on a population of wild , unhabituated chestnut-crowned babblers at the Fowlers Gap Arid Zone Research Station in far western New South Wales , Australia ( 141°42´E , 31°06´S ) . The population has been studied intensively since 2004 . The habitat is characterised by low , open , chenopod shrubland , with trees largely confined to short , linear stands in drainage zones . Chestnut-crowned babblers ( ~50 g ) are ground-foraging , weak-flying , and highly cooperative . During non-breeding they live in groups of 3–23 ( mean ≈ 10 ) individuals , which then partially fragment into 1–4 units of 2–15 individuals ( mean ≈ 6 ) for breeding . Non-breeders associate with those breeders to which they are most related and have substantial effects on their breeding success , primarily by reducing nestling starvation and facilitating additional reproductive attempts by the breeders . Further details on habitat and babbler socio-ecology are provided elsewhere [30 , 31 , 33 , 47–49] . All statistical analyses were performed in Genstat Release 17 ( VSN International Ltd , Hemel Hempstead , UK , 2014 ) . Data used in analyses and figure generation can be found in Dryad: http://dx . doi . org/10 . 5061/dryad . 082v2 [50] . We quantify the use of flight and prompt calls in three different contexts . First , in 2010 , we used focal observations ( 1 h each on six groups from a distance of ~25 m ) to determine the frequency with which the two calls are uttered during natural flights ( n = 450 flights ) . Second , in 2011 , we used a Fostex FR2-LE and wind-shielded Sennheiser ME67 shotgun microphone to record the vocalizations uttered during manual releases from cloth bags following capture ( observers under a bedsheet; n = 90 releases from 25 groups ) . Third , in 2012 , we fitted Yoga EM-400 mini tie-clip microphones to the wall of nests during nestling provisioning and recorded vocalizations using an Olympus LS-10 PCM or Fostex FR2-LE . To quantify the use of flight and prompt calls during flights to and from the nest , as well as during provisioning within the nest , we coupled the above nest-recording system with a transponder system , allowing the timing of bird entrances and exits to and from the nest to be determined [30 , 33 , 47] . Briefly , by inserting transponder tags ( 2 x 12 mm ) into the flanks of the birds and fitting an antenna around the nest entrance linked to a TROVAN decoder , we were able to determine the use of the two calls within 5 s of entering and exiting the nest . Nest recordings were made 7 A . M . –4 P . M . , in August–October , when broods were 1–12 days old . The first 20 nest visits within recording periods were used to quantify call-use at seven nests ( time taken for 20 visits: 68–401 min . ; n = 140 visits ) . To quantify the resemblance among the five elements within and between double-element flight calls and triple-element prompt calls , we selected a single flight and prompt call recorded from each group during releases and nest recordings ( sampling frequency of 44 . 1 kHz , 16 bits ) . Calls were selected randomly from those exhibiting no obscuring vocalizations , high signal-to-noise ratio and low background noise , and blindly with respect to the analyses . The elements of such calls ( n = 23 double-element flight calls and 11 triple-element prompt calls ) were then extracted using Raven Pro , version 1 . 4 ( Bioacoustics Research Program , Cornell Lab of Ornithology , Ithaca , NY , 2011 ) . Five parameters were extracted from the fundamental frequencies of the five elements in the two call types ( start and end frequency , time to peak frequency , frequency range , and element duration ) . All parameters were normalised when necessary and then centred to the mean and standardised by dividing the centralised mean values by 2-fold their standard deviation , allowing direct comparison of the importance of each parameters within and between models [51] . Two of the ten possible correlation coefficients among the five parameters were significantly positive: element length and frequency range ( rp = 0 . 65 , p < 0 . 001 ) ; and start and end frequency ( rp = 0 . 38 , p = 0 . 002 ) ( S1 Table ) . Preliminary Analyses of Variance ( ANOVA ) ( S1 Text ) showed that time to peak frequency was statistically invariant across the five elements , and this was also the case for element durations after controlling for its correlation with frequency range . By contrast , start and end frequency , as well as frequency range , all varied between the elements , and for start and end frequency , this was the case after controlling for their correlation with each other . The three element parameters found to have significant independent effects on element structure ( frequency range , start frequency , end frequency ) were then used in a Discriminant Function Analysis ( DFA ) to determine acoustic similarity ( S2 Table; S3 Table ) . All playback experiments were conducted on wild birds captured in mist-nets on their territories during periods of breeding . Test subjects were chosen randomly from all adult birds captured ( >6 months old ) , excluding the breeding female , without regard for sex and age . Depending on a test subject’s group size , 1–3 individuals were removed ( <30% of group members ) ; resulting in 16 individuals from 7 different groups being tested . Removed birds were transported the 1–5 km immediately by car to aviaries on site at Fowlers Gap and released into separate aviary compartments ( Fig 2 ) . The aviaries consisted of six single compartments each of 2 m long , 2 . 5 m deep and 2 m high . Birds were housed singly , and each fed 20 mealworms every 2–3 h of daylight , delivered through a tube into each aviary compartment , of which 8–15 were typically consumed per bout . Birds gained a mean of 0 . 65 g ( range = -3 . 1 to +4 . 8 g ) during their time in the aviary; all birds were released near their original group less than 48 h after initial capture , and were accepted back into their group without any signs of aggression [49] . Our primary objective in this study was to test whether babblers used a phonemic contrast to generate qualitatively new information . For purposes of experimental rigour and analytical clarity , we chose a fully balanced design , with each bird being presented with the full set of selected playback stimuli . The drawback of presenting multiple stimuli to the same birds lies in the risk of habituation , leading to the generation of ambiguous results . For this reason , we decided to limit the number of playback trials to the absolute minimum number required to test for a phonemic contrast ( i . e . , six ) . Our rationale for the six playback stimuli chosen was as follows . First , given the primary focus , the critical experiments needed to include natural and switched-element versions of both calls ( i . e . , amounting to four playback conditions ) . Second , because the acoustic analyses suggested that the only difference between the two calls derives from P1 in prompt calls , we deemed it key to test whether this element alone partially contributes to the overall meaning of the prompt call by eliciting an increased nest-attentiveness response compared with the flight call . If this were the case , we would have evidence of something more akin to a syntactic than phonemic system . Finally , because flight and prompt calls comprise two and three elements , respectively , we thought it essential to test for an influence of this difference in generating variation in nest attentiveness . We chose a stimulus including C1 P2 P3 because , again , we deemed it most informative for the key aim to manipulate the one element that differs between the two calls ( i . e . , P1 ) . The C1 element was taken from chatter calls: a common multi-element call uttered in bouts of several seconds in contexts of excitement or alarm [32] . The single C1 element was of comparable duration to the replaced P1 element . The calls used in the playback experiments were obtained from natural recordings at the nest of six groups . In each case , a Sennheiser directional microphone ( ME66/K6 ) connected to a Marantz solid-state recorder ( PMD660 , sampling frequency 48 KHz , 24 bits ) was positioned within 1 m of a nest . Playbacks , including the construction of artificial calls ( see below ) , were created with Adobe Audition CC ( Version 6 Build 732 , Adobe Systems ) , selecting high-quality calls ( as above ) . Of the high-quality calls obtained , a single double-element flight call , triple-element prompt call , and a single element of the mixed-element chatter call were selected from each of the six groups ( n = 18 calls ) . For each of the six groups from which recordings were obtained , the set of six playback stimuli were created , with each set including a natural flight call ( F1 F2 ) , a natural prompt call ( P1 P2 P3 ) , a switched-element flight call ( P2 P3 ) , a switched-element prompt call ( P1 F1 F2 ) , a P1 element stimulus ( P1 ) , and a triple-element stimulus ( C1 P2 P3 ) . In all cases , except for one , birds were tested with a new call-set played in randomized order , and birds never received a call-set from their own group . When elements for the generation of artificial and control calls were added and/or replaced , it was ensured that inter-element distance and amplitude matched the original call ( Fig 1A ) . During each playback , a stimulus was repeated six times randomly distributed over 3 . 2–3 . 6 s; a break of at least 10 min was given for focal individuals to resume pre-stimuli behaviour before the initiation of another stimulus . Playback experiments were conducted on the day following capture . Individuals of the same group were tested simultaneously with the same playback-set , but they were always housed separately and could not see each other ( Fig 2 and S2 Text ) . Nevertheless , birds tested simultaneously could influence each other’s behaviour if they reinforced ( or countered ) the playback experiment with their own vocalizations . This was not the case . In the 420 seconds of the playback experiment , not a single prompt call was uttered , and only 24 flight calls were given by the 14 individuals tested simultaneously , leading to a flight call rate of 0 . 28 per bird per 10 s trial . Additionally , of these 24 , only ten were produced during natural or artificial flight call playbacks , all by two of the five groups . Finally , adding whether or not a flight call vocalization was uttered during the playbacks never impacted the explanatory power of the models ( all p values > 0 . 8 ) . During testing , individuals were recorded using digital Sony handycams ( HDR-CX220 and HDR-CX160 ) through a viewing hole to increase image clarity . Visual recordings of 10 s from playback onset were analysed frame by frame using Adobe Audition CC ( Version 6 Build 732 , Adobe Systems ) , with time ( s ) spent in camera view ( mean = 9 . 4 s , range = 10–6 s ) , looking at the nest , looking outside ( i . e . , towards mesh wall ) , and in movement ( hopping or flying ) representing the primary parameters of interest , although general looking around behaviour was also recorded . Marker lists created in Adobe Audition were extracted into txt-files by using CueListTool ( Version 1 . 7 ) , and rates were calculated . Analyses of behavioural data arising from the playback experiments were conducted using Generalized Linear Mixed Models ( GLMM ) , in which the time spent engaged in a given behaviour was fitted as the response term and the total amount of time spent in camera view was fitted as the binomial denominator . Explanatory terms included natural flight and natural prompt calls only ( Fig 3 ) ; call type ( flight or prompt ) , trial type ( natural or switched-element ) , and their interaction ( Fig 4A and 4B ) ; switched-element flight and switched-element prompts calls only ( Fig 4C ) ; and element P1 and CAB stimuli only ( Fig 5A ) or as a four-level factor with natural flight and natural prompt calls ( Fig 5B ) . Additionally , the time spent in view was fitted as a covariate in a single movement analysis ( Fig 3 ) . In all GLMM analyses , individual identity nested within group identity were fitted as random terms . Doing so served two purposes: ( 1 ) it blocked the analyses by individual , effectively generating a more powerful repeated measures statistical design , and ( 2 ) it accounted for any lack of independence arising from testing birds from the same group simultaneously with the same playback stimuli . Regarding this potentially important latter issue: in all analyses , group identity was non-significant ( all p values = 0 . 4–0 . 9 ) , indicating that there was statistically equivalent variation in individual responses from the same group to the same playback stimuli as there was in individual responses from different groups to different playback stimuli . | A major question in language evolution is how its generative power emerged . This power , which allows the communication of limitless thoughts and ideas , is a result of the combinatorial nature of human language: meaningless phonemes can be combined to form meaningful words ( phonology ) , and words can be combined to form higher-order , meaningful structures ( syntax ) . While previous work has indicated the potential for animals to form syntax-like constructions , there exists little convincing evidence for a basic phonemic capacity in animals . Here , we demonstrate , using analyses combined with natural observations and playback experiments , that the cooperatively breeding chestnut-crowned babbler reuses two meaningless acoustic elements to create two functionally distinct vocalizations . This result suggests the basic ability for phoneme structuring occurs outside of humans and provides insights into potential early evolutionary steps preceding the generative phonemic system of human language . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Experimental Evidence for Phonemic Contrasts in a Nonhuman Vocal System |
The intersection of genome-wide association analyses with physiological and functional data indicates that variants regulating islet gene transcription influence type 2 diabetes ( T2D ) predisposition and glucose homeostasis . However , the specific genes through which these regulatory variants act remain poorly characterized . We generated expression quantitative trait locus ( eQTL ) data in 118 human islet samples using RNA-sequencing and high-density genotyping . We identified fourteen loci at which cis-exon-eQTL signals overlapped active islet chromatin signatures and were coincident with established T2D and/or glycemic trait associations . At some , these data provide an experimental link between GWAS signals and biological candidates , such as DGKB and ADCY5 . At others , the cis-signals implicate genes with no prior connection to islet biology , including WARS and ZMIZ1 . At the ZMIZ1 locus , we show that perturbation of ZMIZ1 expression in human islets and beta-cells influences exocytosis and insulin secretion , highlighting a novel role for ZMIZ1 in the maintenance of glucose homeostasis . Together , these findings provide a significant advance in the mechanistic insights of T2D and glycemic trait association loci .
Genome-wide association studies ( GWAS ) have identified approximately 80 loci robustly associated with predisposition to type 2 diabetes ( T2D ) [1–3] and a further 70 influencing a range of continuous glycemic traits [4–10] in non-diabetic subjects . There is substantial , though far from complete , overlap between these two sets of loci . Physiological studies in non-diabetic individuals indicate that most of these loci primarily influence insulin secretion rather than insulin sensitivity , highlighting a key role for the pancreatic islets of Langerhans in the mechanistic underpinnings of these association signals [11 , 12] . These findings have motivated efforts to catalogue the epigenomic and transcriptional landscape of human islets and to apply these findings to deliver biological insights into disease pathogenesis . Recently , it has been shown , for example , that GWAS signals for T2D and fasting glucose show significant co-localization with islet enhancers [13 , 14] . The identification of variant associations mapping to islet regulatory elements raises the question of which downstream ( or “effector” ) transcripts are responsible for mediating those regulatory effects . Relatively few of the T2D GWAS regions feature compelling biological candidates . The identification of cis-eQTL ( expression quantitative trait locus ) signals , especially in disease-relevant conditions and tissues , has , in other contexts , proven a powerful approach for connecting regulatory association signals to their effector transcripts [15–17] . Another major advantage of cis-eQTL data is that , by providing a direction of effect at the transcript level , they can help clarify whether genetic associations affect their phenotype through gain or loss of function–crucial information for translating the genetic findings into therapeutic options . Until now , difficulties in amassing adequate numbers of purified human islet samples have been a barrier to applying this approach at scale in this key tissue . Human islet material is not , for example , available through resources such as the Genotype-Tissue Expression ( GTEx ) project [18] . In this study , we set out to generate eQTL data from human islet samples , and to establish the extent to which this allowed us to identify candidate effector transcripts at GWAS loci for T2D and glycemic traits .
We performed eQTL mapping in islet preparations from 118 human cadaveric donors of Northern European descent ( isolated in Oxford , UK [n = 40] , and Edmonton , Canada [n = 78] ) to elucidate molecular mechanisms underlying both physiological and pathological variation in glucose homeostasis . Expression levels were profiled using RNA sequencing with 100 nucleotide paired-end reads on the Illumina HiSeq2000 platform . This generated an average of 72 million reads per sample uniquely mapping to exons ( range 29–165 million ) . These were aligned to the GENCODE [19] v18 transcriptome reference . Genotypes were obtained using the Illumina HumanOmni2 . 5-Exome array ( 2 , 567 , 513 genotyped SNPs ) with imputation from the 1000 Genomes Phase 1v3 cosmopolitan panel [20] providing data on up to 38 , 089 , 605 autosomal variants . The islet consists of multiple cell types of which the insulin-secreting beta-cells are the most abundant . In line with this , the beta-cell secreted hormone insulin ( INS ) had , on average , 5-fold higher expression across all samples ( an average RPKM [reads per kilobase of transcript per million reads mapped] of 58846 ) than the next most abundantly expressed protein-coding gene ( S1 Fig ) . There was also high RNA expression of other canonical islet cell hormones including glucagon ( GCG; average RPKM 4030 ) , somatostatin ( SST; average RPKM 1708 ) and pancreatic polypeptide ( PPY; average RPKM 1452 ) ( S1 Fig ) . Islet eQTL analysis was performed using an additive linear model implemented in the R package MatrixEQTL [21] . For known common T2D and glycemic trait association loci , these data were integrated with genetic information ( that is , patterns of association seen in large GWAS meta-analysis for T2D and continuous glycemic traits ) and islet regulatory state maps [13 , 14] . We chose to focus on eQTL analyses at the level of the exon ( as opposed to overall gene-level eQTLs ) , given that the former additionally captures variants that influence exon splicing . To account for variance attributable to factors such as donor characteristics , islet isolation center , purity , and storage ( e . g . 55% of the samples had been cryopreserved for an extended period [22] , see Methods ) , exon counts were normalized using gender and 15 PEER [23] factors derived from the normalized expression profile ( these capture hidden covariates present in the data using Bayesian factor analysis methods ) . This normalization procedure successfully eliminated much of the structure observed in the raw data , most of which we attribute to experimental and technical factors For each transcript , all variants within 1Mb flanking regions of the transcriptional start site ( TSS ) were tested for association . To correct for multiple testing ( i . e . the many different cis-variants considered for each exon expression value ) , an empirical p-value was calculated from the most significant eQTL p-value per exon by permuting expression values between 1 , 000 and 10 , 000 times , while retaining the relation between expression value and covariates ( see Methods ) . From this empirical p-value distribution , we calculated a false discovery rate ( q-value ) for each exon using the Storey method [24] , imposing a study-wide false-discovery rate threshold of q<0 . 05 . Across the 27 , 772 protein-coding and long non-coding ( lncRNA ) transcripts expressed in the human islet samples ( expression was taken to be non-zero exon counts in at least 10% of individuals ) , we identified 2 , 341 genes that included at least one exon meeting this criterion ( S1 Table ) . The majority ( 90% ) of significant islet exon-eQTLs was located within 250kb of the transcriptional start site , in line with observations in other tissues [17] . Even considering only the index variant for each of the significant islet exon-eQTLs , there is clear consistency with published islet chromatin maps: 735/2 , 341 ( 31% ) variants overlapped enhancer or promoter signatures in at least one of the datasets [13 , 14] ( S1 Table ) . When we discarded variants that had no chromatin annotation in either published map [13 , 14] , the overlap with enhancers and promoters was even greater ( 59%; 735/1 , 252 ) . The overlap of the 2 , 341 significant islet exon-eQTL variants with active islet chromatin signatures is significantly higher than that observed with 10 , 000 random samplings of 2 , 341 variants with no significant eQTL ( 2-fold enrichment , Fisher’s p = 1 . 7x10-23 with all variants; 1 . 7-fold enrichment , Fisher’s p = 5 . 7x10-9 when excluding non-overlapping variants ) . We could also compare islet expression with RNA-Seq data for nine additional tissues analyzed , in approximately the same numbers of samples , as part of the GTEx project pilot study [18] . Since GTEx eQTLs are generated at the gene level , we reprocessed the data to generate exon-eQTLs . There was substantial sharing of islet exon-eQTLs across the full range of GTEx p-values with a mean estimated replication rate ( π1[25] ) of 70% ( ranging from 66% [heart–left ventricle] to 73% [tibial artery] ) . There were , however , a total of 309 exons with an islet exon-eQTL that were expressed in at least one of the GTEx tissues ( out of 1 , 659 such exons; 19% ) , but showed no association ( p> = 0 . 05 ) in the GTEx data . These are likely to represent islet-specific regulatory regions . Next , we focused on further analysis of the subset of cis-exon-eQTLs that mapped to the 82 known common variant T2D loci [1–3] and 49 loci for glycemic traits for which altered beta-cell function has been shown to be the main driver [4–10] . The latter included fasting glucose , fasting proinsulin , 2-hour glucose , HOMA-B , insulinogenic index , disposition index , corrected insulin response ( insulin response to glucose after the first 30 minutes ) and AUCInsulin/AUCGlucose [4–10] . Seventeen of the glycemic trait loci overlap with T2D signals , whereas the other thirty-two are independent . To identify putative cis-effector transcripts for lead regulatory variants in these regions , we considered , for each of the regions , all genes with transcriptional start sites within 1Mb of any reported genome-wide significant lead variant ( n = 218 variants ) . We adapted the genome-wide eQTL detection strategy describe above to identify , for each cis-region of interest , the single exon with the strongest cis-eQTL association . To minimize the possibility that co-localizing cis-eQTL and GWAS variants were tagging different functional variants ( incidental overlaps are frequent given the abundance of cis-eQTLs in the genome ) , we required that the exon-eQTL index variant was in strong LD ( 1000 Genomes project CEU r2>0 . 8 ) with the lead T2D or glycemic trait variant . We further verified the co-incidence of eQTL and GWAS variants by performing conditional analyses: specifically , we confirmed whether regressing out the variance explained by the T2D or glycemic trait lead GWAS variant eliminated , or at least , seriously depleted the cis-eQTL association signal . Within the GWAS regions , there were a total of 232 transcripts that met the study-wide significance criteria ( i . e . q<0 . 05 ) . Over 90% of the exon-eQTLs for these genes were statistically independent of the GWAS signal , but nine ( marked by eleven GWAS index variants ) met the LD criterion of r2>0 . 8 and evidence for co-localization from the conditional analysis ( S2 Table ) . Since GWAS regions have a higher biological prior expectation of harboring an islet regulatory eQTL [13 , 14] , we also considered an additional ten cis-eQTLs at which the statistical evidence did not reach study-wide significance , but which nonetheless displayed nominal significance ( permuted p<0 . 05 , corresponding to q<0 . 44; S2 Table ) as well as meeting the other criteria related to GWAS signal overlap and conditional analysis . The combined set of twenty one variants was distributed over sixteen loci . With the exception of AP3S2 , all showed a consistent direction of effect across the other exons of the implicated transcript ( S2 Table ) . At two loci ( ABO and ZFAND6 ) , none of the variants in the set in strong LD ( r2>0 . 8 ) with the GWAS and exon-eQTL lead variants overlapped an islet-active regulatory state annotation in published datasets [13 , 14] . Whilst this does not necessarily exclude an effect on islet gene expression or relevance to the maintenance of glucose homeostasis , we did not consider these loci further . We compared the islet eQTL data generated by the present study to that from a recent analysis of an entirely independent set of 89 human islets by colleagues in Sweden [26] . Though there were substantial experimental and processing differences between the two analyses , the present study replicated overlap of islet eQTL and GWAS signals at 80% ( 4/5 ) of the GWAS-related islet eQTLs reported in that study ( ABO , AP3S2 , ERAP2 , and MTNR1B ) . Only two of these make it into our final list: at ABO there was no overlap with active islet chromatin , whilst at ERAP2 conditional analysis could not confirm co-localization of eQTL and GWAS signal . There is also substantial replication of the genome-wide set of 616 eQTL signals described by Fadista et al . Of these 616 , 503 had gene identifiers that could be mapped to the data described in this manuscript , with 43% ( 216/503 ) also having a significant ( q<0 . 05 ) islet exon-eQTL ( S3 Table ) . The observed gene-level replication rate is substantially higher than , for example , the 32% reported in a recent study [27] comparing two independent cis-eQTL mapping experiments in blood . The data reported by Fadista and colleagues uses gene-level rather than exon-level analyses . Nonetheless , we found that , amongst the 216 genes that had a cis-eQTL in both datasets , the same variant was associated in the majority of instances ( 56%—S3 Table ) . The vast majority ( 94% ) of the 122 shared cis-eQTL signals are directionally consistent ( S3 Table ) . This overlap provides reassurance that , despite technical and other challenges , and modest sample size , a high proportion of the cis-eQTL signals detected in these studies are robust . The various filters described above left us with a set of nineteen variants , at fourteen loci , where multiple lines of evidence supported the candidacy of the exon-eQTL transcript as the effector for the relevant GWAS signal ( Table 1; S2 Table ) . At four of these loci , the islet exon-eQTL overlapped GWAS variants that are genome-wide significant for both T2D and glycemic trait variation ( ADCY5 , ARAP1 , DGKB , MTNR1B ) . At four others ( AP3S2 , CDC123/CAMK1D , TMEM163 , ZMIZ1 ) the GWAS signal was for T2D alone . For the remaining six ( AMT , ANK1 , FADS1 , MADD , PCSK1 , WARS ) , the co-incident GWAS data implicated a range of continuous glycemic phenotypes ( Table 1; S2 Table ) . At three of the loci ( ADCY5 , DGKB , FADS1 ) , the exon-eQTL data provide an independent empirical link between the GWAS signals and transcripts that already have strong biological candidacy with respect to glucose homeostasis . At ADCY5 , where the GWAS variant influences T2D [3 , 4] , fasting glucose [4] , 2-hour glucose [10] , HOMA-B [4] and birth weight [28] , the rs11708067 A T2D-risk allele was associated with lower transcript expression levels ( exon permuted p = 8 . 4x10-3 , q = 0 . 183 , ß = -0 . 44 ) . This is consistent with a previous report , from a small candidate gene study [29] , of a negative correlation between risk allele count and ADCY5 expression levels . In human islets , ADCY5 , a member of the adenylate cyclase family , is thought to couple glucose stimulation to insulin secretion , and this coupling is disrupted upon gene knockdown [29] . There are two independent T2D GWAS signals at the DGKB locus ( lead variants rs2191349 and rs17168486 ) [3 , 4] , separated by about 160 kilobases . At both , the T2D-risk allele is also associated with raised fasting glucose and reduced HOMA-B in non-diabetic individuals [3 , 4] . In the exon-eQTL data , both T2D-risk alleles independently drove higher expression levels of DGKB ( rs2191349 signal , exon permuted p = 1 . 0x10-3 , q = 0 . 040 , ß = 0 . 41; rs17168486 signal , exon permuted p = 9 . 3x10-3 , q = 0 . 194 , ß = 0 . 52 ) . Variant sets for both the 5’ of DGKB ( rs17168486 ) and the more distal signal at rs2191349 overlapped islet chromatin signatures denoting either active promoters or enhancers [13 , 14] . DGKB is a subunit of diacylglycerol kinase , a regulator of the glucose-responsive secondary messenger diacylglycerol [30] . At FADS1 , the GWAS allele associated with raised fasting glucose ( in non-diabetic individuals ) was implicated in increased islet expression of FADS1 ( exon permuted p = 1 . 6x10-2 , q = 0 . 262 , ß = 0 . 31 ) . FADS1 encodes the delta-5 fatty acid desaturase , which plays a role in the biosynthesis of highly unsaturated fatty acids . Variants in the same LD block as the fasting glucose GWAS variant are associated with altered blood levels of the substrate/product pair for the enzyme [31] . The lipid-related function of FADS1 might appear , at first thought , to connect this locus to insulin sensitivity: however , the fasting glucose-raising allele [4] at this locus has also been associated with a lower HOMA-B [4] and insulinogenic index [12] , consistent with an islet-mediated effect . The hypothesis that FADS1 might modulate insulin secretion through altered insulin sensitivity in the islet itself is supported by studies demonstrating the effects of fatty acid composition on insulin secretion both in vitro [32] and in vivo [33] . At two further cis-eQTL loci , our findings replicate previous studies . At the MTNR1B locus , the T2D-risk allele [1 , 3] also has a substantial impact on continuous glycemic traits ( higher fasting glucose [4] , lower HOMA-B [4] and corrected insulin response [8] ) . In the present study , as in two previous analyses of human islet expression [26 , 34] , the same allele was associated with increased expression of the melatonin receptor 1B ( exon permuted p = 1 . 5x10-2 , q = 0 . 252 , ß = 0 . 40 ) . At the T2D-associated CDC123/CAMK1D locus [1 , 3] , the islet cis-eQTL for CAMK1D ( calcium/calmodulin-dependent protein kinase ID; exon permuted p = 2 . 0x10-4 , q = 0 . 011 , ß = 0 . 61 ) endorsed the designation of CAMK1D as the likely effector emanating from previous studies conducted in other tissues [18 , 35] . Recent work has demonstrated that the T2D-risk allele is associated with increased transcriptional activity in a luciferase reporter system [36] , again consistent with the islet eQTL data . Whilst a single effector transcript was involved in the examples above , at certain other loci , the expression data are less conclusive . At the ARAP1 locus , the islet exon-eQTL data link the T2D-risk allele [3] ( also fasting glucose-raising [5 , 9] , and fasting proinsulin-reducing [6] ) to lower expression of STARD10 ( exon permuted p = 4 . 0x10-4 , q = 0 . 019 , ß = -0 . 39 ) . This exon-eQTL is one of the 309 potentially islet-specific eQTLs based on comparison with data from nine GTEx tissues ( see above ) . STARD10 , which encodes StAR-related lipid transfer ( START ) domain containing 10 , is thought to be involved in the regulation of bile acid metabolism [37] , and has no reported role in the islets . At this locus , there have been reports , from human islet studies , of allele-specific expression of an alternative regional gene , ARAP1 , encoding Arf-GAP with Rho-GAP domain , ANK repeat and PH domain-containing protein 1 [38] . The variants found to exhibit allele-specific expression were shown to affect promoter activity of the ARAP1 P1 promoter in a dual luciferase system [38] . However , the published data on allelic imbalance in ARAP1 are inconsistent [38 , 39] , and we found no evidence of allelic imbalance for the relevant variant ( rs11603334; Wilcoxon signed rank test p>0 . 1; S2A Fig ) in our data . Neither was there any significant islet cis-eQTL signal for ARAP1 . Therefore the data from this much larger islet cohort suggest STARD10 rather than ARAP1 as the likely effector transcript . Additional studies ( e . g . conformational capture , CRISPR–Cas9 genome editing ) will be instrumental in definitively assigning this locus to its effector transcript . At the AP3S2 locus , the T2D GWAS signal[3] coincided with an islet eQTL for AP3S2 , encoding adaptor-related protein complex 3 , sigma 2 subunit ( exon permuted p = 1 . 0x10-4 , q = 0 . 006 , ß = -0 . 55 ) . The identical signal was also detected in the recent report from an independent islet eQTL analysis [26] . However , in non-islet tissues , variants in strong LD with the T2D index variant have been reported as significant eQTLs for both AP3S2 and ANPEP , a second regional gene which encodes alanyl ( membrane ) aminopeptidase [35 , 40] . Variants in ANPEP , although not in strong LD with the T2D signal , also showed allelic imbalance in human islets in both our data ( S2B and S2C Fig ) and a previous study by Locke et al [39] . Islet expression data for this locus , therefore , implicates both genes . Variants at the MADD locus are associated with fasting glucose [4] and insulin processing defects [6] . At this locus , the islet exon-eQTL data implicated two regional transcripts: MADD , encoding MAP-kinase activating death domain ( exon permuted p = 1 . 0x10-4 , q = 0 . 006 , ß = 0 . 25 ) ; and ACP2 , encoding lysosomal acid phosphatase 2 ( exon permuted p = 1 . 0x10-4 , q = 0 . 006 , ß = 0 . 31 ) . Analysis of a beta-cell specific knockout mouse recently demonstrated that Madd plays a key role in glucose-stimulated insulin secretion , but the marked abnormalities of insulin processing that characterize the human GWAS signal were not observed [41] , indicating that MADD might not mediate all the phenotypes associated with this signal . ACP2 is a lysosomal enzyme: disruption of the homolog in mice impacts lysosome function and causes cerebellar and skin abnormalities [42] . The known role of lysosomes in the degradation of aging insulin granules [43] provides a potential link between this gene and altered composition of the insulin secretory pool , which might explain the observed effects of the human association signal on fasting glucose and proinsulin levels . These examples act as reminders of the importance of the independent validation of expression findings . They also highlight the potential for non-coding variants of interest to influence multiple transcripts , although this does not necessarily mean that all affected transcripts are involved in T2D pathogenesis . The mechanisms through which the other six implicated transcripts ( CTD-2260A17 . 2 , MGAT5 , NKX6-3 , RBM6 , WARS and ZMIZ1 at the PCSK1 , TMEM163 , ANK1 , AMT , WARS and ZMIZ1 loci , respectively ) influence islet physiology are less clear . The fasting glucose-raising allele at the PCSK1 locus [5 , 9] was associated with increased expression of the uncharacterized protein CTD-2260A17 . 2 ( exon permuted p = 2 . 6x10-2 , q = 0 . 331 , ß = 0 . 58 ) . However , at this locus there is strong biological candidacy of PCSK1 [44] , with coding variants in this gene thought to be causal for the association signal [9 , 45] . Loci where the underlying molecular mechanism affects protein function rather than regulation of transcript levels ( also for example SLC30A8 ) are unlikely to be detected in eQTL studies . Therefore this raises doubts about the biological relevance of the association with CTD-2260A17 . 2 expression at the PCSK1 locus . The gene implicated at the TMEM163 locus was MGAT5 , for which the T2D risk-increasing allele was associated with higher islet expression of the gene ( exon permuted p = 2 . 4x10-2 , q = 0 . 320 , ß = 0 . 26 ) . MGAT5 encodes the protein N-glycosylation enzyme mannosyl ( alpha-1 , 6- ) -glycoprotein beta-1 , 6-N-acetyl-glucosaminyltransferase . The properties of cell surface receptors and transporters can be modulated through N-glycosylation; in beta-cells expression of the glucose transporter GLUT2 [46] and the incretin receptors [47] at the cell surface is , for example , altered by this process . Whole-body Mgat5 knockout mice had improved insulin sensitivity and decreased gluconeogenesis [48] , although effects on the beta-cell have not been studied . This direction of effect would be consistent with higher expression levels of MGAT5 increasing risk of developing T2D . At the AMT fasting glucose locus [5] , the islet exon-eQTL implicated RBM6 ( exon permuted p = 5 . 9x10-3 , q = 0 . 147 , ß = -0 . 23 ) . RBM6 encodes RNA Binding Motif Protein 6 , but neither the gene nor the protein has any defined phenotypic links . NKX6-3 , which encodes NK6 homeobox 3 , was implicated as the effector transcript for the ANK1 locus variants influencing insulin secretion [8] ( exon permuted p = 1 . 0x10-3 , q = 0 . 040 , ß = -0 . 36 ) . The same region is also associated with T2D [3] . However , the T2D-risk variants are in comparatively low LD ( r2 = 0 . 14 ) with the corrected insulin secretion association signal , and no exon-eQTL signal was observed for these . NKX6 . 3 has a known role in the development of the gastrin-producing ( G ) and somatostatin producing ( D ) cells of the gastric endocrine system [49] . It is also active in the developing central nervous system [50] . There is no literature on the role of NKX6 . 3 in the islet , but , given the key role of other NKX6 transcription factors in the development of the endocrine pancreas [51] , further follow-up of the islet consequences of altered NKX6 . 3 expression is clearly warranted . The fasting glucose-raising allele [5] at the WARS locus was associated with markedly reduced WARS expression in human islets ( exon permuted p = 1 . 0x10-4 , q = 0 . 006 , ß = -1 . 58 ) . WARS encodes a tryptophanyl-tRNA synthetase involved in protein synthesis , regulated by cytokines and involved in cellular growth pathways such as angiogenesis [52] . It has , until now , not been allocated a role in the regulation of pancreatic islet function . The final gene implicated by our data was ZMIZ1 , encoding zinc finger , MIZ-type containing 1 . ZMIZ1 maps to a locus implicated in T2D-risk [3] . The ZMIZ1 islet eQTL ( exon permuted p = 3 . 8x10-2 , q = 0 . 392 , ß = 0 . 13 ) showed a consistent direction of effect across 23/24 ZMIZ1 exons . The same cis-eQTL had a directionally consistent , although not significant , signal in the recently published independent islet expression [26] . It has not been detected in any other available cis-eQTL dataset , suggesting an islet-specific effect . To establish whether the putative effector transcripts identified by the exon-eQTL data provide novel biological inference , functional validation is essential . We used ZMIZ1 as our exemplar for this purpose . At the ZMIZ1 locus , the exon-eQTL index variant was in near complete linkage disequilibrium ( r2 = 0 . 98 ) with the T2D GWAS variant rs12571751 , and overlapped an extended region of active islet enhancer chromatin ( Fig 1A ) . Stretch enhancers such as this have been linked to cell-specific gene regulation [13] and , in human islets , to T2D [14] . Current understanding of ZMIZ1 function is limited , but it has been shown to act as a transcriptional co-regulator , playing a regulatory role in the p53 [53] , Notch [54] and Smad [55] signaling cascades , and as a PIAS-like E3 SUMO-ligase [56] . Several variants in the wider region , independent of the T2D and islet eQTL signal ( r2<0 . 04 ) , have been associated with a variety of autoimmune and inflammatory disorders ( including inflammatory bowel disease and multiple sclerosis ) [57 , 58] , in addition to ZMIZ1 expression in immune-relevant monocytes [15] . Our exon-eQTL approach has therefore highlighted a previously-unsuspected role for ZMIZ1 in pancreatic islet function , independent of the regional association to immune phenotypes . Within human pancreas sections , ZMIZ1 was preferentially expressed in the islet and co-localized with both insulin and glucagon ( n = 4 individuals; Fig 1B ) . Since ZMIZ1 expression is higher in carriers of the T2D-associated rs12571751 A allele , we first determined the effects of ZMIZ1 over-expression in dispersed human islet cells . We infected dispersed human beta-cells ( n = 5 donors , 8 replicates for each condition in each donor ) with a control adenovirus ( Ad-GFP ) or adenovirus expressing ZMIZ1 ( Ad-ZMIZ1 ) . Increasing ZMIZ1 ( to 4520% of control expression levels , as confirmed by qPCR ) impaired both glucose- and KCl-induced insulin secretion ( 20 . 5% and 25 . 8% reduction in stimulation index , p<0 . 01 and <0 . 001 , respectively; Fig 1C ) . Knockdown of ZMIZ1 in dispersed human islet cells ( to 39 . 6% of control , confirmed by qPCR ) had no significant effect on glucose-stimulated insulin secretion ( also n = 5 donors , 8 replicates for each condition in each donor; Fig 1D ) , although KCl-induced insulin secretion was , paradoxically , reduced ( p<0 . 05; Fig 1D ) . To further explore the potential impact of ZMIZ1 up-regulation , we measured exocytosis in human beta-cells directly . Upon membrane depolarization , fusion of insulin granule-containing secretory vesicles with the plasma membrane results in an increase in membrane surface area that can be detected by whole cell patch clamp as an increase in membrane capacitance . Over-expression of ZMIZ1 reduced insulin exocytosis in individual human beta-cells to 29% of that in GFP-transfected controls ( 41–44 beta-cells from 6 individuals , p<0 . 001; Fig 1E and 1F ) . This represents a true impairment in exocytosis , rather than a reduction in the Ca2+ influx needed to trigger exocytosis , since voltage-dependent Ca2+ channel activity was unchanged by ZMIZ1 over-expression ( 24–27 beta-cells from 3 individuals; Fig 1G ) . Together these data indicate a novel role for ZMIZ1 in the regulation of insulin secretion in human islets .
One of the key challenges faced in the biological interpretation of common variant GWAS signals lies in establishing the functional connections between causal variants within regulatory sequence and the downstream ( or “effector” ) genes through which they mediate their phenotypic effects . This is an essential step if we are to be effective in using human genetics to define pathways and networks central to the pathogenesis of common complex disease , and in identifying targets that may lead to novel preventative and therapeutic strategies . A range of complementary , bioinformatic and experimental , approaches are available to address this challenge . These include mapping the correlations between assays of chromatin state and cis-promoter activity [59] , direct interrogation of local DNA interactions [60] , and the search for coding variants in regional genes that recapitulate the disease phenotype [61] . In the present study , we demonstrate , through integration of human genetic disease association signals with information on patterns of exon-eQTLs and chromatin state in human islets , the potential for studies of human islet mRNA expression to implicate genes that play a previously unsuspected role in the maintenance of normal glucose homeostasis and the development of T2D . The focus on human islets was motivated by compelling evidence , from a variety of sources [1 , 11 , 13 , 14] , which places islet dysfunction center-stage with respect to T2D pathogenesis . Despite this , and for understandable reasons to do with tissue accessibility and purity , human islets are largely absent from major eQTL and transcriptome cataloguing efforts such as GTEx [18] , necessitating parallel efforts to define the interplay between DNA sequence variation and transcript expression in this key tissue . As expected [17 , 62] , the cis-exon-eQTL signals we detected in islets were a mixture of those shared across multiple tissues , and those that are islet specific . For example , 20% of the islet exon-eQTLs were not significant in any of the tissues studied in the GTEx pilot ( though this may change as the GTEx sample size increases ) . Of the cis-eQTLs identified at GWAS loci for T2D and/or glycemic traits , only those involving AP3S2 and CAMK1D had been identified as significant eQTLs in other tissues [18 , 35 , 40] . The STARD10 islet exon-eQTL , for example , was not even nominally significant in any of nine GTEx tissues . These data emphasize the importance of extending such expression studies to the tissues most directly implicated in disease pathogenesis . The identification of candidate effector transcripts through this and other routes motivates efforts to characterize the functional role of these genes in relevant cellular and animal systems . In the present study , we focused on one such gene , ZMIZ1 , on the basis that the strength of the evidence for the cis-exon-eQTL was intermediate ( it did not attain study-wide significance ) , and because it had no previous documented relationship to islet biology , other than localization within a T2D GWAS signal . We were able to show that ZMIZ1 expression is localized to the endocrine pancreas ( ruling out the possibility that the eQTL signal emanated from contaminating exocrine tissue ) , and that perturbation of ZMIZ1 within the islet has a marked effect on exocytosis and insulin secretion , data that are clearly consistent with the designation of this gene as the likely mediator of the T2D association signal at this locus . Having said that , further work is required to fully enumerate the role of ZMIZ1 in the islet , to explain , for example , the apparently paradoxical reduction in KCl-stimulated insulin secretion observed in the knockdown experiment . This observation may be a consequence of the exaggerated attenuation of ZMIZ1 expression in these experiments , when compared to the more subtle perturbation associated with the cis-eQTL . As well as providing insights into transcript candidacy , these human eQTL studies are also informative with respect to the question of the directional impact of T2D-risk alleles on those genes . Recent studies of protein-truncating variants in SLC30A8 [63] have demonstrated how crucial such information can be for guiding the design of potential pharmacological agents . Two examples are worth highlighting . The islet exon-eQTL data presented here indicates that the T2D-risk allele at the ADCY5 locus is associated with reduced expression of ADCY5 and that reduced ADCY5 activity contributes to T2D pathogenesis . However , rare coding variants in ADCY5 have been shown to be causal for a Mendelian disease phenotype characterized by neuromuscular features [64] . These rare Mendelian alleles act through gain of ADCY5 function , and this is presumably why the phenotype of this condition ( familial dyskinesia with facial myokymia ) does not feature diabetes . This pattern of directional effects also diminishes the attraction of ADCY5 as a potential drug target for T2D . In contrast , at MTNR1B the islet eQTL data presented here , along with several previous studies [26 , 34] , tie the T2D-risk allele to increased expression of the cognate transcript . This replicated observation runs counter to a combined genetic and functional analysis of rare coding variants in MTNR1B , which reported that T2D risk was conveyed by alleles that reduced MTNR1B function [65] . Though increased MTNR1B transcript levels and reduced MTNR1B function could both be implicated in T2D susceptibility if reduced MTNR1B function was accompanied by changes in MTNR1B subcellular localization or a secondary increase of protein levels , the data by Bonnefond and colleagues [65] is not consistent with this explanation . It has also been proposed that these apparently contradictory findings could be explained by an absence of a negative feedback loop on MTNR1B expression in conditions of seriously impaired melatonin receptor function [65] . However , this appears inconsistent with the observation that islet expression of MTNR1B was entirely absent ( below background , RPKM < 0 . 1 ) in 69% of individuals homozygous for the non-risk allele ( and 37% of homozygous risk-allele carriers ) . These contrasting data hint at a complexity in the relationship between genetic variation and MTNR1B function that may only be resolved by direct assessment of the effects of melatonin on glucose homeostasis in human studies . The present study represents the largest sample of human islet gene expression reported to date , but the sample size remains modest compared to those available for many other tissues . However , whereas association studies typically need effective sample sizes in the tens of thousands , the current islet eQTL study of 118 samples already identified putative effector transcripts at eight T2D loci . Physiological data had previously implicated a role for the islet at the majority of these loci , showing they affected beta-cell function [11] . This , combined with the extensive , but incomplete , overlap with the signals detected in a recent report of human islet expression [26] , indicates that there is much to be gained by combining available data sets . Such efforts will likely generate many additional signals , at GWAS loci and beyond , as well as supporting additional analyses ( e . g . of allele-specific expression ) . Similar studies in other T2D-relevant tissues will shed light on effector transcripts for loci that do not directly modulate insulin secretion–an example of this can be found at the KLF14 locus , where eQTL studies in adipose tissue uncovered a large KLF14-regulated trans-eQTL network underlying the T2D association signal [16] . Data for non-islet tissues will also help answer whether loci that have been associated with changes in beta-cell function by in vivo studies in humans act directly on the islet or affect insulin secretion indirectly by altering , for example , expression in brain or gut . As a more complete picture of the islet cis-eQTL landscape emerges , it will be highly informative to integrate these data with those obtained from the implementation of orthogonal , informatic and experimental , approaches for linking regulatory variants of interest to their transcriptional targets . Recent advances that enable scale up of conformational capture across multiple genomic regions are likely to be particularly relevant here [60] . Additionally , dense genomic annotations have become available for key T2D-relevant tissues , and similar data is being generated on islets at different developmental stages and after application of metabolic stimuli ( e . g . comparing high versus low glucose culturing ) . This provides a rich framework for deriving functional inference from human genetics , and for identifying translational opportunities with respect to target identification and biomarker discovery .
Human islets were collected in two locations . Forty samples were freshly isolated at the Oxford Centre for Islet Transplantation ( OXCIT ) in Oxford , UK , as described [66] , and processed for RNA and DNA extraction after 1–3 days in culture in CMRL media . In Edmonton , Canada , 65 samples were extracted from the long-term cryopreserved biobank and thawed as described [22] , or were freshly isolated ( n = 13 ) from donor pancreas as described previously [67] . For functional studies islets from a total of 12 donors were used ( age = 52 . 4 +/- 3 . 9 years , 50% male , BMI 27 . 8+/-1 . 7 ) . Pancreas biopsies were taken , fixed in Z-fix , and paraffin embedded prior to sectioning and immunostaining ( described below ) . Isolated or thawed islets were cultured in CMRL media for 1–3 days prior to storage for RNA extraction or in vitro experimentation . Only freshly isolated islets were used for electrophysiology and insulin secretion studies . All studies were approved by the Human Research Ethics Board at the University of Alberta ( Pro00001754 ) , the University of Oxford's Oxford Tropical Research Ethics Committee ( OxTREC Reference: 2–15 ) , or the Oxfordshire Regional Ethics Committee B ( REC reference: 09/H0605/2 ) . All organ donors provided informed consent for use of pancreatic tissue in research . RNA was extracted from human islets using Trizol ( Ambion , UK or Sigma Aldrich , Canada ) . To clean remaining media from the islets , samples were washed three times with phosphate buffered saline ( Sigma Aldrich , UK ) . After the final cleaning step 1 mL Trizol was added to the cells . The cells were lysed by pipetting immediately to ensure rapid inhibition of RNase activity and incubated at room temperature for ten minutes . Lysates were then transferred to clean 1 . 5 mL RNase-free centrifuge tubes ( Applied Biosystems , UK ) . For islet preparations isolated in Edmonton , Trizol fractions were shipped to Oxford before further processing . For the phase separation , 200μL chloroform ( Fisher Scientific , UK ) was added to each tube . Samples were vigorously shaken to begin organic and aqueous phase separation . This was followed by a 5 minute incubation room temperature and 30 minute-spin at 12 , 000 x g and 4°C to complete phase separation . The aqueous phase containing the RNA was transferred to a clean 1 . 5ml RNase-free tube by pipette , and 500μl isopropanol ( Fisher Scientific , Loughborough , UK ) was added to precipitate the RNA . The remaining organic and DNA phases were used for DNA extraction ( see below ) . The RNA solution was incubated for 5 minutes at room temperature and stored overnight at -20°C . The following day , RNA was pelleted by centrifugation at 12 , 000 x g for 50 minutes ( 4°C ) and supernatant was carefully removed . The pellet was washed twice in 1 ml 75% ethanol ( Sigma Aldrich , UK ) before centrifugation at 12 , 000 x g for 30minutes . After the final ethanol wash was removed , the RNA pellet was allowed to air-dry for 10 minutes . To re-suspend the RNA , a minimum of 20μl RNase-free water ( more as necessary for complete re-suspension ) was added to each sample . RNA quality ( RIN score ) was determined using an Agilent 2100 Bioanalyser ( Agilent , UK ) , with a RIN score > 6 deemed acceptable for inclusion in the study . Samples were stored at -80°C prior to sequencing . For the majority of samples , DNA was extracted from either spleen or the exocrine fraction of the islet isolation using the Tissue DNA Purification Kit according to manufacturer’s instructions on an automated Maxwell 16 system ( both Promega , USA ) . When no other tissue was available , DNA was extracted from human islets using the Trizol fraction remaining after extraction of RNA ( see above ) . To precipitate the DNA , 300μl 100% ethanol was added to the thawed solution . This mixture was incubated at room temperature for a minimum of 30 minutes . DNA was then pelleted by centrifugation at 4 , 000 x g for 5 minutes at 4°C . After removing the supernatant , the pellet was twice washed with 0 . 1M trisodium citrate ( Sigma Aldrich , UK ) in 10% ethanol and left at room temperature for 30 minutes , followed by another wash step with 75% ethanol . After the final wash step , pellets were air-dried for 10 minutes to remove residual ethanol and re-suspended in a minimum of 100 μL 8mM NaOH ( Sigma Aldrich ) . Extracted DNA was stored at -20°C before further use . In total , 118 samples were genotyped on the Illumina Omni2 . 5+Exome genotyping array . Samples were prepared according to the Illumina Infinium protocol and run on the Illumina iScan platform at the Oxford Genomics Centre ( Wellcome Trust Centre for Human Genetics , University of Oxford , Oxford , UK ) . Genotypes were called with Illumina GenCall software using the standard Illumina cluster file and default genotype calling cut-offs . The direct genotypes were then used for imputation . Principal component analysis was performed to confirm European ancestry of all samples ( S3 Fig ) . Variants with a call rate < 99% and minor allele frequency ( MAF ) < 0 . 01 , as well as those deviating from Hardy–Weinberg equilibrium ( p<0 . 0001 ) , were filtered out before imputation–leaving 1 , 323 , 351 variants . Haplotypes were inferred from these genotype data using SHAPEIT [68] . Genotypes were imputed into the phased haplotypes using IMPUTE2 [69] with the entire 1000 Genomes Phase 1 v3 release [20] as the reference panel . For the QTL analysis , we used 5 . 8 million imputed autosomal single nucleotide variants with an INFO score > 0 . 4 and MAF > 0 . 05 . Poly-A selected libraries were prepared from total RNA at the Oxford Genomics Centre using NEBNext ultra directional RNA library prep kit for Illumina with custom 8bp indexes [70] . Libraries were multiplexed ( 3 samples per lane ) , clustered using TruSeq PE Cluster Kit v3 , and paired-end sequenced ( 100nt ) using Illumina TruSeq v3 chemistry on the Illumina HiSeq2000 platform . Samples were mapped with TopHat2 [71] on default settings with GENCODE v18 [19] as transcriptome and GRCh37 as genome reference . Exon level reads counts for all protein-coding and long non-coding transcripts present in GENCODE v18 were quantified with RNA-SeQC [72] with the “strictMode” flag set . Transcript level counts were compiled by adding up the counts for all exons . The sequenced data was required to contain at least 10M mapped and properly paired reads after applying the quality filters . First , exons with no expression in 10 or more samples were removed . To normalize for variation in read depth across samples , exon counts were scaled to the median number of exon-mapping reads per sample . The scaled exon counts were log2-normalized followed by per exon transformation to a standard normal ( to minimize the effects of outliers in the linear regression ) . Even considering only the index variant variation in the QTL analysis , we derived 15 synthetic covariates from the normalized exon profile using PEER with default settings [23] . Since none of the 15 PEER factors were significantly correlated ( q-value < 0 . 05 ) with gender , we added this as an additional covariate . The QTL analysis was performed on all SNP-exon pairs within 1Mb flanking regions of the transcripts transcriptional start site ( TSS ) using linear regression assuming an additive model as implemented in MatrixEQTL [21] . To correct for multiple testing per gene expression phenotype , we permuted the expression labels per samples ( while maintaining the relation between PEER factors and expression labels ) and compared the minimum p-value for each permutation against the minimum observed p-value until at least 15 more extreme p-values were observed ( with a minimum of 1 , 000 and maximum 10 , 000 of permutations ) . From these data we calculated a permuted p-value for each exon . False-discovery rate across the permuted p-values for all exons estimated using the q-value method [24] , with a q<0 . 05 threshold used for identifying study-wide significant islet exon-eQTL genes . For the overlap between GWAS loci and islet eQTLs we additionally considered all exons with a permuted p<0 . 05 , with the best exon used per locus . To determine the islet exon-eQTLs sharing across tissues , we generated exon-eQTL calls for the GTEx pilot dataset [18] . We used reference files and exon count from the GTEx portal ( http://www . gtexportal . org/home/datasets2 , last accessed on 30 August 2015 ) , and genotype files available through dbGaP . Exon counts were processed as described above . We replaced the 15 GTEx-supplied gene-level PEER factors with those derived from the normalized exon counts , while retaining the other GTEx covariates . Finally , exon-eQTL mapping was performed as described above . Human pancreatic biopsies were fixed in Z-fix ( Anatech , USA ) , paraffin embedded , and sliced into 5μm sections . Sections were rehydrated and antigen unmasking performed . Immunostaining was performed for insulin ( Santa Cruz Biotechnology Inc . , USA ) , glucagon ( EMD Millipore , USA ) as previously described . The antibody targeting ZIMZ1 ( ZIMPZ10; sc-82438 Santa Cruz Biotechnology Inc . 1:50 , overnight incubation ) recognizes an N-terminal epitope . All slides were coverslipped with prolong gold antifade and visualized on a WaveFX spinning disk confocal ( Quorum Technologies , Canada ) using a 40X/1 . 3 NA lens and 405 , 491 , 561 , and 642nm excitation lasers coupled with matched filter sets . Images were captured on a Hamamatsu EMC9100-13 camera ( Hamamatsu Corp , USA ) using Volocity imaging software ( Perkin Elmer , Canada ) . Analysis of images was performed using Volocity and ImageJ ( NIH ) . Human islets were hand-picked to purity and dispersed using enzyme-free cell dissociation buffer ( Life Technologies , Canada ) . Cells were plated on 35mm dishes and transfected with control ( pEGFP-N1 , Clontech , Mountain View , CA , USA ) or ZMIZ1 over-expression ( ZMIZ1 pCMV6- AC-GFP , Origene , Rockville , MD , USA ) plasmids via lipid transfection ( Lipofectamine 2000 , Life Technologies , Canada ) . Following 48hrs post-transfection culture we used the standard whole-cell techniques with the sine+DC lockin function of an EPC10 USB amplifier and Patchmaster software ( HEKA Electronics , Germany ) to measure capacitance during a series of ten depolarizations of 500ms each from -70 to 0mV . Experiments were performed at 32–35°C . Extracellular bath solution for depolarization trains contained ( in mM ) : 118 NaCl , 20 TEA , 5 . 6 KCl , 1 . 2 MgCl2 , 2 . 6 CaCl2 , 10 glucose and 5 HEPES ( pH7 . 4 with NaOH ) . Dishes were preincubated for one hour in culture media with 1mM glucose before capacitance measurements . Pipette solution for depolarization trains contained ( in mM ) : 125 Cs-glutamate , 10 CsCl , 10 NaCl , 1 MgCl2 , 0 . 05 EGTA , 5 HEPES , 0 . 1 cAMP and 3 MgATP ( pH 7 . 15 with CsOH ) . To measure voltage-dependent Ca2+ channel activity , using Ba2+ as a charge carrier , the pipette solution contained ( in mM ) : 140 Cs-glutamate , 1 MgCl2 , 20 tetraethylammonium chloride , 5 EGTA , 20 HEPES and 3 MgATP ( pH 7 . 3 with CsOH ) . The bath contained ( in mM ) : 20 BaCl2 , 100 NaCl , 5 CsCl , 1 MgCl2 , 5 glucose , 10 HEPES , and 0 . 5 μM tetrodotoxin ( pH 7 . 35 with CsOH ) . Patch pipettes , pulled from borosilicate glass and coated with Sylgard , had resistances of 3-4megaohm ( MΩ ) when filled with pipette solution . Whole-cell capacitance responses were normalized to initial cell size and expressed as femtofarad per picofarad ( fF/pF ) or picoampere per picofarad ( pA/pF ) . Human islets were hand-picked to purity and dispersed using Accutase ( Life Technologies , Canada ) and plated in a 96 V-well plate at a density of 5000 cells/well . ZMIZ1 over-expression ( AdZMIZ1 or AdGFP , Welgen Inc . , USA ) or siRNA knockdown ( siZMIZ1 or siScrambled , Life Technologies ) was performed at the time of plating . Cells were cultured in CMRL 1066 ( Corning , USA ) supplemented with 0 . 5% bovine serum albumin ( Equitech-Bio Inc . , USA ) , 1% insulin transferrin selenium ( Corning ) , 100 U/mL penicillin/streptomycin ( Life Technologies ) and L-glutamine ( Sigma-Aldrich ) at 37°C , 5% CO2 . Insulin secretion experiments were performed after 24 hours ( over-expression ) or 48 hours ( siRNA knockdown ) culture in incubation buffer containing ( in mM ) : 115 NaCl , 5 . 0 KCl , 24 NaHCO3 , 2 . 2 CaCl2 , 1 MgCl2 , 0 . 25% BSA , 24 HEPES ( pH7 . 3 with NaOH ) . Cells were pre-incubated for 45 minutes at 1mM glucose , followed by 1hour stimulation with 1mM glucose , 16 . 7mM glucose or 16 . 7mM glucose plus 20mM KCl . Samples were collected at stored at -80°C prior to assay by electrochemiluminescence ( Meso Scale Diagnostics , USA ) . To account for the normal variation in secretory responses between donors , data was normalized to the control 1 mM glucose condition and presented as stimulation index ( SI; fold increase ) . Data were analyzed by repeated measures two-way ANOVA and Tukey post-test . Genotype and sequence data have been deposited at the European Genome-phenome Archive ( EGA; http://www . ebi . ac . uk/ega/ ) , which is hosted by the European Bioinformatics Institute ( EBI ) , under accession number EGAS00001001265 . | Genetic studies have uncovered many different parts of the genome playing a role in the risk of developing diabetes , or affecting blood sugar levels in the normal population . However , it has so far been difficult to tie these parts of the genome to genes that are responsible for the observed changes in risk and/or blood sugar levels ( “effector transcripts” ) . It is clear from the genetic data that one of the key tissues in these phenotypes is the human pancreatic islet of Langerhans , but the limited availability of this tissue has been a major hurdle in translating the genetics into biology . Here , we present a study linking genetic variation to gene expression changes in 118 islet preparations . Using these cis-eQTLs , we provide candidate effector transcripts at 14 regions of the genome previously associated with glucose phenotypes . Many of the genes implicated through this approach have no known role in the islet . By experimentally changing the expression levels of one of these novel genes , ZMIZ1 , in human islets and beta-cells , we uncovered a novel role for ZMIZ1 in exocytosis and insulin secretion . These findings therefore significantly improve the discovery of biology underlying type 2 diabetes and glucose trait association . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Transcript Expression Data from Human Islets Links Regulatory Signals from Genome-Wide Association Studies for Type 2 Diabetes and Glycemic Traits to Their Downstream Effectors |
Two surgical options are available for cystic echinococcosis ( CE ) . The two principal approaches are radical ( resection of the cyst ) and conservative ( evacuation of the cyst content and partial removal of the cyst capsule ) . Here , we describe a standardized endocystectomy technique for hepatic echinococcosis . Twenty-one patients ( male/female: 4/3; median age: 28 years ) with uncomplicated , isolated hepatic CE ( cyst stages WHO CE1 , 2 , 3a , and 3b ) that were treated with the standardized endocystectomy described in this paper . Before the operation and during the follow-up period ( mean: 33 . 8 months , median: 24 months ) , patients underwent clinical and sonographical and/or magnetic resonance imaging assessment during regular visits managed by an interdisciplinary team . Forty-seven cysts were treated with the standardized endocystectomy technique . The median number of cysts per patient was two ( range: 1–8 ) . Nine patients ( 43% ) had a single cystic lesion . The median operation time was 165 minutes and the median intraoperative bleeding volume was 200 mL . The median hospital stay was nine days ( range: 6–28 days ) . Morbidity ( Clavien-Dindo III ) occurred in four patients ( 19% ) . No mortality and no recurrence were found during the median follow-up time of 24 months . The standardized endocystectomy technique presented is a safe procedure with acceptable morbidity , no mortality , and without recurrences in our patient series . Important components of our CE management are interdisciplinary patient care , adequate diagnostic work-ups , and regular pre- and postoperative visits , including long-term follow-up for early and reliable capture of recurrences .
Cystic echinococcosis ( CE ) is a parasitic disease caused by ingestion of the larval stage of Echinococcus granulosus . The liver is the most commonly infected organ ( about 75% of cases ) , followed by the lungs and other organs . In most high-income countries , the incidence of CE is very low , diagnosed mainly in migrants originating from endemic regions . CE is largely asymptomatic , diagnosed incidentally or when complications precipitate , such as biliary obstruction due to cysto-biliary fistulas , spillage of cyst content into the biliary tree , and compression of vessels ( bile ducts , portal or hepatic veins , and inferior caval vein in hepatic CE ) . Outside endemic regions , most health care professionals have limited experience with CE patients [1–4] . Medical treatment ( benzimidazoles ) , percutaneous interventions , surgery , and the watch-and-wait strategy are the available treatment options and the treatment strategy is decided based on the WHO CE cyst classification [1 , 3–11] . Without CE cyst staging , uncomplicated CE cysts are overtreated and inappropriate use of imaging modalities causes misclassification [12–14] . The recommendations of the Informal WHO Working Group Echinococcosis ( WHO-IWGE ) of 2010 remain largely valid for the surgical treatment of CE: “Surgery should be carefully evaluated against other options before choosing this treatment . It is the first choice for complicated cysts . In liver cysts , surgery is indicated for ( a ) removal of large CE2–CE3b cysts with multiple daughter vesicles , ( b ) single liver cysts , situated superficially , that may rupture spontaneously or as a result of trauma when percutaneous treatment ( PT ) options are not available , ( c ) infected cysts when PTs are not available , ( d ) cysts communicating with the biliary tree ( as alternative to PT ) , and ( e ) cysts exerting pressure on adjacent vital organs . ” [3] Recently , CE2 and CE3b were successfully treated percutaneously with a technique that awaits further consolidation in other CE treatment centers [7] . The surgical indications and approaches need to be further elaborated and defined for the various cyst presentations and health system settings . The two principal surgical approaches are radical and conservative . In radical surgery , including liver resection and pericystectomy , the entire unopened CE cyst including the host tissue-derived capsule is removed , which means loss of liver tissue . In conservative surgery , the content of the CE cyst is carefully evacuated and the host tissue-derived CE capsule is only partially removed , which spares liver tissue . Endocystectomy is the most common conservative surgical approach , originally developed by Lindeman in 1871 . Reported complication , mortality , and recurrence rates reflect varied performances across health care settings more than characterizing the technique as such . Endocystectomy has been modified multiple times over the years , including omentoplasty , protective procedures to avoid spillage of cyst content–during the most critical step when the cyst is opened and evacuated–and partial excision of the host-derived capsule . We describe in detail a standardized endocystectomy technique for treating uncomplicated hepatic CE that is suitable for surgical residents .
All procedures were performed according to the most recent revision of the Declaration of Helsinki . All included patients aged more than 18 years old and a written informed consent for anonymous collection and analysis of clinical data was provided by all patients before surgery . The study protocol was also approved by the independent ethics committee of the university ( S-754 ) . Preoperative diagnosis is based on imaging , preferentially ultrasound and/or magnetic resonance imaging ( MRI ) [3 , 11 , 14–16] . Serology is only used to confirm the diagnosis [17 , 18] . All patients receive 400 mg albendazole 6 hours prior to surgery and this treatment is continued for one month after the operation if no intraoperative complications occur . Liver function tests and neutrophils are carefully monitored during benzimidazole treatment . All patients receive a preoperative X-ray of the chest to exclude concomitant lung CE . For endocystectomy , the patient is positioned in a supine position and laparotomy is performed using a supraumbilical midline incision with a right lateral extension if necessary . Afterwards , the liver is mobilized by dividing its ligaments ( Fig 3 ) . Intraoperative ultrasound examination is performed to confirm the number of cysts ( Fig 4A and 4B ) , to define and mark the exact anatomical cyst locations ( Fig 4C and 4D ) , and to finally check if all cysts are removed ( particularly exophytic growth ) . If a cholecystectomy is performed , the cystic duct is temporarily secured using a bulldog clamp so that any bile leakage could later be determined by retrograde instillation of fat emulsion solution into the bile system ( White test ) . To protect the surrounding tissues from spillage of cyst content , the surgical field was covered with sterile green laparotomy sponges ( moistened with 0 . 9% normal saline ) ( Fig 4C ) . Blue laparotomy sponges soaked with 20% sodium chloride were placed on top of the green sponges ( Fig 4D ) . Different colors were used to safely distinguish the two types of surgical sponges . Twenty percent sodium chloride inactivates protoscolices and the germinal layer of the endocyst , but it is cytotoxic to the peritoneum and the biliary tract . Since contact with cyst content is unavoidable during the opening of the cyst and the evacuation procedure , the instrumentation table and neighbouring surfaces were covered and only the necessary surgical instruments , including the suction tip , were exposed . These were immediately removed after use . Gloves and gowns were changed before continuing with the “clean” procedures . The insertion site for the 12 mm trocar was carefully selected at the part of the pericyst exposed to the liver surface ( Fig 4D ) . The trocar was preferentially inserted vertically and tightened with a purse string suture to control spillage of CE fluid at the trocar-tissue interface ( Fig 5A ) . Four holding threads prevent retraction of the pericyst during emptying of the cyst . Fig 5B shows removal of the cyst content , including the fragmented endocyst , protoscolices , cell detritus , and hydatid fluid using a 12 mm trocar under permanent suction . The evacuated fluid is centrifuged and protoscolices microscopically assessed for viability after eosin staining . The cyst is thoroughly washed with normal saline via a three-way valve . After the trocar is removed , the capsule is partially resected at the inner circumference of the holding threads ( Fig 6 ) . The inner surface of the cyst capsule is inspected and any residual endocyst material that may have remained after evacuation and washing of the cysts is removed ( Fig 7 ) . The residual cavity is explored carefully for cysto-biliary communications ( Fig 7 ) . Cysto-biliary fistulas are apparent if the evacuated cyst content is bile stained . In a substantial proportion of cysts , cysto-biliary fistulas can be predicted with a preoperative MRI [15] . All visible cysto-biliary communications are closed with stitches . Afterwards , a White test is performed using 10–30 mL of liposaccharide to detect further bile leakage [19] . After the biliary fistulas are closed , the cyst cavity is thoroughly cleaned with swabs soaked in normal saline ( Fig 8A ) , and then stuffed with sponges soaked in 20% sodium chloride for 30 minutes ( Fig 8B ) . The same procedure is followed if no cysto-biliary fistulas are detected . After inactivation of any remaining parasitic cells ( protoscolices and germinal membrane cells ) , the sponges are removed , and the edge of the cyst is over sewn to prevent bile leakage ( Fig 9 ) . In patients with multiple cysts , the procedure described above is repeated for each cyst . To take care of the residual cavity , if possible , an omentoplasty is performed by inserting a plug of the greater omentum fixed at the margin of the cyst wall ( Fig 10 ) . Before closing the abdominal wall , a drain is placed in the operation field . Patients are regularly visited by surgeons and tropical medicine specialists during hospitalization . Intra- and postoperative complications are recorded and classified based on the Clavien-Dindo classification [20] . After discharge from the hospital , the patients are followed up in the outpatient clinic of the surgical department and the special clinic for echinococcosis in the tropical medicine unit . The latter takes care of perioperative albendazole treatment and long-term follow-up for relapses . To detect recurrences early–depending on the location of the cyst–ultrasound or MRI investigations are annually performed; the first set of images within one month of the operation to document fluid retention , which is not related to CE recurrence ( biliomas , seromas ) . This allows us to distinguish CE-regrowth from preexisting non-CE fluid collections during later follow-up sessions . The whole abdomen , including the lesser pelvis , is observed to detect CE recurrence distant from the evacuated CE cysts . Follow-up is continued for at least 7 years to exclude late recurrences .
All twenty-one patients originated from endemic regions in Eastern Europe , the Middle East , and Asia . The median patient age was 28 years ( range: 16–76 years ) . Twelve patients ( 57 . 1% ) were male , nine female ( 42 . 9% ) . Nine patients ( 42 . 9% ) had a single cyst , the remaining 12 patients ( 57 . 1% ) had multiple cysts . A total of 47 cysts were detected in the 21 patients included in the study . The median number of cysts was two ( range: 1–8 ) . One patient had eight cysts and sequential endocystectomies successfully removed all cysts in one session . The cyst diameters ranged between 4 . 1 and 15 cm . A total of 47 endocystectomies were performed with the standardized method described above . All patients received perioperative albendazole , except one pregnant patient . She was 22 years old and was referred with hepatic CE identified during a routine gynecologic examination in the 22nd week of pregnancy . The patient was operated because of the risk of cyst rupture during childbirth . The operation was performed successfully together with a gynecologist . Six of the treated patients ( 28 . 6% ) were referred from other centers because of CE recurrence after initial treatment . Preoperative cysto-biliary fistulas were detected in two patients ( 9 . 5% ) , biliary obstruction was reported in two patients ( 9 . 5% ) , and cholangitis was diagnosed in one patient ( 4 . 8% ) . Concomitant cholecystectomy was performed in 13 patients ( 61 . 9% ) . In six of these patients , the White test was performed to exclude bile leakage . Omentoplasty was performed in 14 patients ( 66 . 7% ) . The median intraoperative blood loss was 200 mL ( range: 50–800 mL ) and the median operation time was 165 minutes ( range: 120–250 minutes ) . No CE fluid spillage occurred during the operations . The median hospitalization of the patients was 9 days ( range: 6–28 days ) and the median and mean follow-up period were 24 and 33 . 8 months , respectively ( range: 1–75 months , first quartile: 15 . 5 months , third quartile: 52 months ) . Four complications ( 19 . 0% ) occurred during the follow-up period , without any mortality . Three of these complications were surgical ( bile leakage ) and one was non-surgical ( pleural effusions/pneumonia ) . All complications were Clavien-Dindo grade III; two cases required non-surgical intervention ( IIIa ) and two cases required reoperation , over sewing , and lavage because of bile leakage ( IIIb ) . Relative to the total number of operated cysts , the rate of bile leakage as the only surgical complication was 6 . 4% . No recurrence was recorded during the study period .
Surgery is the method of choice for most complicated CE cysts ( those with cyst-biliary fistulas , ruptured cysts ) and for uncomplicated active cysts ( mainly very large cysts > 10 cm ) that cannot be treated percutaneously , with benzimidazoles , or observation ( watch-and-wait strategy ) [1 , 3–10] . The criteria for assigning patients to radical ( liver resection/pericystectomy ) or conservative parenchyma-sparing surgical treatment ( endocystectomy ) are still under debate . This is no surprise , considering the high variability of CE presentations . Development of these criteria requires well-described standardized techniques for the principle surgical approaches . For our highly standardized endocystectomy protocol , we selected well-defined , uncomplicated , staged cysts ( WHO cyst stages CE1 , 2 , 3a , 3b ) to minimize heterogeneity and interacting factors . Judging the efficacy and complications of individual surgical treatment modalities has been difficult in published studies because of limited cyst staging , heterogeneous inclusion criteria and study endpoints , and short follow-up periods . A recent systematic review revealed that 71 . 2% of the papers published on hepatic CE did not mention any classification at all and WHO classification were only mentioned in 14% of papers [21] . Our endocystectomy protocol has been developed with a good understanding of the architecture of CE cysts and we have concentrated on CE-specific steps , most importantly the inactivation and spillage control of highly active cyst content , which determines recurrence . Identification and closure of biliary fistulas , management of biliomas and seromas etc . are not CE-specific but are widely encountered in liver surgery . Interpretation of published studies is complicated by the fact that patients in need of surgery are treated in a wide range of health systems in countries with varying resources and skills of surgeons , anesthetists , and other health-service staff . This means that non-CE-specific factors also play a role in managing patients . Therefore , highly standardized surgical protocols and standardization of the practice are of high importance for improving the surgical outcomes of hepatic CE worldwide , and for determination of criteria for assigning patients to different surgical techniques . With these limitations in mind , the postoperative morbidity and mortality rates in our endocystectomy patient series are similar to those after radical surgery reported in the meta-analysis of He et al . , ( 19 . 0% vs 17 . 7% and 0% vs 0 . 8% , respectively ) [22] . The parenchyma-sparing endocystectomy is more feasible than resection in a wide range of general surgical units with minimal operative trauma . However , 20% sodium chloride is used during conservative surgical procedures , which increases the risk of chemical cholangitis if not used with greatest care . In our series , surgical complications occurred in 14 . 2% of patients . The median number of cysts per patient was two ( range: 1–8 ) and 47 endocystectomies were performed in total . The risk of surgical complications depended on the number of cysts per patient and increased with the number of endocystectomies performed per patient . Therefore , reporting the rate of surgical complications by total number of cysts operated on appears to be the more relevant figure . The total and surgical complication rates per number of endocystectomies performed were 8 . 5% and 6 . 4% , respectively . No CE recurrence was observed during the follow-up period , suggesting that this procedure is efficacious and safe from a surgical point of view . However , longer follow-ups are required to accurately determine the recurrence rates of hepatic CE stratified by the various conservative and radical surgical procedures . Bile leakage is the most prevalent complication of endocystectomies . A meta-analysis identified no significant differences in bile leakage in patients treated with radical and conservative surgery . In contrast , individual reports have demonstrated a higher risk of bile leakage after conservative surgery . Wang et al . [19] suggested on the basis of a systematic review that bile leakage tests are useful and do not have adverse effects on the patients . The White test is strongly recommended to identify cysto-biliary communications and to ensure that all leaks are closed . In addition , over sewing the cyst capsule margin prevents bile leakage . In summary , this highly standardized endocystectomy protocol ( a ) removes all biologically active CE cyst content from the residual cavity that would cause local recurrence , and ( b ) avoids intraabdominal and intravascular spillage of infectious CE material to prevent abdominal and systemic dissemination . Perioperative benzimidazole prophylaxis may be beneficial; however , the efficacy of this treatment in preventing recurrence in the event of spillage remains to be determined . In the experience of our center over the past 20 years , the close interdisciplinary collaboration between tropical medicine/infectious disease specialists , interventional radiologists , endoscopists , and surgeons has been key to the development of treatment strategies , refinement of intervention techniques , and adaptation of treatment modalities to the needs of individual patients with this highly complex disease . | Cystic echinococcosis ( CE ) is a parasitic disease caused by ingestion of the larval stage of Echinococcus granulosus . The liver is the most commonly infected organ . There are currently four treatments for CE: surgery , percutaneous treatment , medical treatment ( benzimidazoles ) , and watch-and-wait strategy . Treatment is decided based on the WHO cyst staging . The surgical techniques employed depend on the cyst location and related complications ( e . g . cyst-biliary fistulas , rupture , and secondary bacterial infection ) . The two principal surgical approaches are radical ( resection of the cyst ) and conservative ( evacuation of the cyst content and partial removal of the cyst capsule ) surgeries . In this study , we presented a conservative surgical approach , a standardized endocystectomy technique , that is suitable for surgical residents . This standardized endocystectomy technique is a safe procedure with acceptable morbidity , no mortality , and without recurrences in our patient series . Important components of CE management are interdisciplinary patient care , adequate diagnostic work-ups , and regular pre- and postoperative visits , including long-term follow-up for early and reliable capture of recurrences . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"diagnostic",
"radiology",
"pathology",
"and",
"laboratory",
"medicine",
"tropical",
"diseases",
"bile",
"parasitic",
"diseases",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"magnetic",
"resonance",
"imaging",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"research",
"and",
"analysis",
"methods",
"echinococcosis",
"imaging",
"techniques",
"radical",
"surgery",
"surgical",
"resection",
"helminth",
"infections",
"laparotomy",
"fistulas",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences"
] | 2019 | Standardized endocystectomy technique for surgical treatment of uncomplicated hepatic cystic echinococcosis |
Gene tree topologies have proven a powerful data source for various tasks , including species tree inference and species delimitation . Consequently , methods for computing probabilities of gene trees within species trees have been developed and widely used in probabilistic inference frameworks . All these methods assume an underlying multispecies coalescent model . However , when reticulate evolutionary events such as hybridization occur , these methods are inadequate , as they do not account for such events . Methods that account for both hybridization and deep coalescence in computing the probability of a gene tree topology currently exist for very limited cases . However , no such methods exist for general cases , owing primarily to the fact that it is currently unknown how to compute the probability of a gene tree topology within the branches of a phylogenetic network . Here we present a novel method for computing the probability of gene tree topologies on phylogenetic networks and demonstrate its application to the inference of hybridization in the presence of incomplete lineage sorting . We reanalyze a Saccharomyces species data set for which multiple analyses had converged on a species tree candidate . Using our method , though , we show that an evolutionary hypothesis involving hybridization in this group has better support than one of strict divergence . A similar reanalysis on a group of three Drosophila species shows that the data is consistent with hybridization . Further , using extensive simulation studies , we demonstrate the power of gene tree topologies at obtaining accurate estimates of branch lengths and hybridization probabilities of a given phylogenetic network . Finally , we discuss identifiability issues with detecting hybridization , particularly in cases that involve extinction or incomplete sampling of taxa .
A molecular systematics paradigm that views molecular sequences as the characters of gene trees , and gene trees as characters of the species tree [1] is being increasingly adopted in the post-genomic era [2] , [3] . Several models of evolution for the former type of characters have been devised [4] , while the coalescent has been the main model of the latter type of characters [5] , [6] . However , hybridization , a process that is believed to play an important role in the speciation and evolutionary innovations of several groups of plant and animal species [7] , [8] , results in reticulate ( species ) evolutionary histories that are best modeled using a phylogenetic network [9] , [10] . Further , as hybridization may occur between closely related species , incongruence among gene trees may also be partly due to deep coalescence , and distinguishing between the two factors is hard under these conditions [11] . Therefore , to enable a more general application of the new paradigm , a phylogenetic network model that allows simultaneously for deep coalescence events as well as hybridization is needed [12] . This model can be devised by extending the coalescent model to allow for computing gene tree probabilities in the presence of hybridization . In this paper we focus on gene tree topologies and analyze the signal they contain for detecting hybridization in the presence of deep coalescence . Applications of probabilities of gene tree topologies given species trees include determining statistical consistency ( or inconsistency ) of topology-based methods for inferring species trees 13–15 , testing the multispecies coalescent model [13] , [16] , determining identifiability of species trees using linear invariants of functions of gene tree topology probabilities [17] , [18] , delimiting species [19] , designing simulation studies for species tree inference methods [20]–[22] , and inferring species trees [23] , [24] . We expect that similar applications may be useful for probabilities of gene tree topologies given species networks . In particular , it will be useful to be able to evaluate the performance of methods that infer species trees in the presence of hybridization as well as the performance of methods for inferring species networks . Knowing the distribution of gene tree topologies could also be useful for estimating the probability that two gene trees have the same topology , a quantity that is used in constructing the prior which models gene tree discordance in BUCKy [25] , a program that is often used to estimate species trees or concordance trees . A method for computing the probability mass function of gene tree topologies in the absence of hybridization ( i . e . , under the multispecies coalescent model is assumed ) is given by Degnan and Salter [26] . However , to handle hybridization and deep coalescence simultaneously , this method has to be extended to allow for reticulate species evolutionary histories . Indeed , attempts have been made recently for this very task [27]–[30] , all of which have focused on very limited special cases where the phylogenetic network topology is known and contains one or two hybridization events , and a single allele sampled per species . However , a general formula for the probability of a gene tree topology given a general ( any number of taxa , hybridizations , gene trees , and/or alleles ) phylogenetic network has remained elusive . A binary phylogenetic network topology contains two types of nodes: tree nodes , each of which has exactly one parent ( except for the root , which has zero parents ) , and reticulation nodes , each of which has exactly two parents . The edge incident into a tree node is called tree edge , and the edges incident into a reticulation node are called reticulation edges . In our context , we associate with a phylogenetic network a vector of branch lengths ( in units of generations , where is the effective population size in that branch ) and a vector of hybridization probabilities ( which indicates for each allele in a hybrid population its probability of inheritance from each of the two parent populations ) ; see Text S1 for formal definition . The gene tree topology can be viewed as a random variable with probability mass function . In this paper , we solve the aforementioned open problem by reporting on a novel method for computing the probability of a gene tree topology given a phylogenetic network , . We illustrate the use of gene tree topology probabilities to estimate the values of species network parameters using the likelihood of the gene tree topologies . This application allows for disentangling hybridization and deep coalescence when analyzing a set of incongruent gene trees , as both events can give rise to similar incongruence patterns . Given a collection of gene tree topologies , one per locus , in a set of sampled loci , the likelihood function is given by ( 1 ) This formulation provides a framework for estimating the parameters and of an evolutionary history hypothesis , given a collection of gene trees . Estimates of 0 or 1 for the entries in the vector reflect the absence of evidence for hybridization based on the gene tree topology distribution . As gene tree topologies are estimated from sequence data , there is often uncertainty about them . In our method , we account for that in two ways: ( 1 ) by considering a set of gene tree topology candidates , along with their associated probabilities ( produced , for example , by a Bayesian analysis ) , and ( 2 ) by considering for each locus the strict consensus of all optimal tree topologies computed for that locus ( produced , for example , by a maximum parsimony analysis ) . Finally , to account for model complexity , we employ a simple technique based on three information criteria , AIC [31] , AICc [32] and BIC [33] . While these criteria have their shortcomings for model selection , the question of how to account for phylogenetic network complexity is still wide open and no methods exist for addressing it systematically [10] . We have implemented our method in the publicly available software package PhyloNet [34] and demonstrated its broad utilities in three domains . First , we reanalyze a Saccharomyces data set and a Drosophila data set , and find support for hybridization in both data sets . Second , we show the identifiability of the parameter values of certain reticulate evolutionary histories . Third , we highlight and discuss the lack of identifiability of the parameters in other scenarios that involve extinctions .
Degnan and Salter [26] gave the mass probability function of a gene tree topology for a given species tree with topology and vector of branch lengths as ( 2 ) which is taken over coalescent histories from the set of all coalescent histories . The product is taken over all internal branches of the species tree . The term is the probability that lineages coalesce into lineages on branch whose length is . And the terms and represents the probability that the coalescent events agree with the gene tree topology . In particular , is the number of ways that coalescent events can occur consistently with the gene tree and is the number of sequences of coalescences that give the number of coalescent events specified by . However , this equation assumes that is a tree and as such is inapplicable to reticulate evolutionary histories . Recently , this equation was adapted to very special cases of species phylogenies with hybridization [28]–[30] . However , none of these adaptations is general enough to allow for multiple hybridizations , multiple alleles per species , or arbitrary divergence patterns following hybridization . We present a novel approach for generalizing this equation to handle hybridization . Our approach is general enough in that it allows for computing gene tree probabilities on any binary phylogenetic network topology , thus overcoming limitations of recent works . Our approach for computing the probability of a gene tree given a species network has three steps . First , is converted into a multilabeled ( MUL ) tree ( a tree whose leaves are not uniquely labeled by a set of taxa; see Text S1 ) ; second , the alleles at the tips of are mapped in every valid way to the tips of ; and , finally , the probability of is computed as the sum , over all valid allele mappings , of probabilities of given ( see Figure 1 ) . Thus far , we have assumed that we have an accurate , fully resolved gene tree for each locus . However , in practice , gene tree topologies are inferred from sequence data and , as such , there is uncertainty about them . In Bayesian inference , this uncertainty is reflected by a posterior distribution of gene tree topologies . In a parsimony analysis , several equally optimal trees are computed . We propose here a way for incorporating this uncertainty into the framework above . Assume we have loci under analysis , and for each locus , a Bayesian analysis of the sequence alignment returns a set of gene trees , along with their associated posterior probabilities ( ) . Now , let be the set of all distinct tree topologies computed on all loci , and for each let be the sum of posterior probabilities associated with all gene trees computed over all loci whose topology is . Thus , and . Then , we replace Eq . ( 1 ) by ( 7 ) We note that if or for each and , then Eq . ( 7 ) is equivalent to Eq . ( 1 ) , and both are multinomial likelihoods . This multinomial approach has also been used elsewhere for both species networks under simple hybridization scenarios [28] and species trees [24] . We additionally allow the terms to be between 0 and 1 ( and therefore to be non-integer values ) in order to reflect uncertainty in the estimated gene trees . In the case where a maximum parsimony analysis is conducted to infer gene trees on the individual loci , a different treatment is necessary , since for each locus , all inferred trees are equally optimal . For locus , let be the strict consensus of all optimal gene tree topologies found . Then , Eq . ( 1 ) becomes ( 8 ) where is the set of all binary refinements of gene tree topology .
Using our method to compute the likelihood function given by Eq . ( 1 ) , we reanalyzed the yeast data set of [35] , which consists of 106 loci , each with a single allele sampled from seven Saccharomyces species S . cerevisiae ( Scer ) , S . paradoxus ( Spar ) , S . mikatae ( Smik ) , S . kudriavzevii ( Skud ) , S . bayanus ( Sbay ) , S . castellii ( Scas ) , S . kluyveri ( Sklu ) , and the outgroup fungus Candida albicans ( Calb ) . Given that there is no indication of coalescences deeper than the MRCA of Scer , Spar , Smik , Skud , and Sbay [36] , we focused only on the evolutionary history of these five species ( see Text S1 ) . We inferred gene trees using Bayesian inference in MrBayes [37] and using maximum parsimony in PAUP* [38] ( see Text S1 for settings ) . The species tree that has been reported for these five species , based on the 106 loci , is shown in Figure 2A [35] . Further , additional studies inferred the tree in Figure 2B as a very close candidate for giving rise to the 106 gene trees , under the coalescent model [36] , [39] . Notice that the difference between the two trees is the placement of Skud , which flags hybridization as a possibility . Indeed , the phylogenetic network topologies in Figure 2C and 2D have been proposed as an alternative evolutionary history , under the stochastic framework of [40] , as well as the parsimony framework of [30] . Using the 106 gene trees , we estimated the times , , , and for the six phylogenies in Figure 2 that maximize the likelihood function ( we used a grid search of values between 0 . 05 and 4 , with step length of 0 . 05 for branch lengths , and values between 0 and 1 with step length of 0 . 01 for ) . Table 1 lists the values of the parameters computed using Eq . ( 7 ) on the gene trees inferred by MrBayes and Table 2 lists the values of the parameters computed using Eq . ( 8 ) on the gene trees inferred by PAUP* , as well as the values of three information criteria , AIC [31] , AICc [32] and BIC [33] , in order to account for the number of parameters and allow for model selection . Out of the 106 gene trees ( using either of the two inference methods ) , roughly 100 trees placed Scer and Spar as sister taxa , which potentially reflects the lack of deep coalescence involving this clade ( and is reflected by the relatively large values estimated ) . Roughly 25% of the gene trees did not show monophyly of the group Scer , Spar , and Smik , thus indicating a mild level of deep coalescence involving these three species ( and reflected by the relatively small values estimated ) . However , a large proportion of the 106 gene trees indicated incongruence involving Skud; see . This pattern is reflected by the very low estimates of the time on the two phylogenetic trees in Figure 2 . On the other hand , analysis under the phylogenetic network models of Figure 2C and 2D indicates a larger divergence time , with substantial extent of hybridization . These latter hypotheses naturally result in a better likelihood score . When accounting for model complexity , all three information criteria indicated that these two phylogenetic network models with extensive hybridization and larger divergence time between Sbay and the ( Smik , ( Scer , Spar ) ) clade provide better fit for the data . Further , while both networks produced identical hybridization probabilities , the network in Figure 2D had much lower values of the information criteria than those of the network in Figure 2E . The networks in Figure 2E and 2F have lower support ( under all measures ) than the other four phylogenies . In summary , our analysis gives higher support for the hypothesis of extensive hybridization , a low degree of deep coalescence , and long branch lengths than to the hypothesis of a species tree with short branches and extensive deep coalescence . It is worth mentioning that while the three networks in Figure 2C–2E were reported as equally optimal under a parsimonious reconciliation [36] , our new framework can distinguish among the three , and identifies the network in Figure 2D as best , followed by the one in Figure 2C ( the network of Figure 2E is found to be a worse fit than either of the two species tree candidates ) . We reanalyzed the three-species Drosophila data set of [41] , which includes D . melanogaster ( Dmel ) , D . yakuba ( Dyak ) , and D . erecta ( Dere ) . The data set consisted of loci supporting the three possible gene tree topologies as follows: For a species tree with three species and one individual sampled per species , the multispecies coalescent predicts that the two gene trees with topologies different from that of the species tree each occur with probability , where is the length of the one internal branch in coalescent units [42] . Two important predictions under the coalescent are therefore that the two nonmatching gene trees are expected to be tied in frequency and that both occur less than of the time , with the matching gene tree topology occurring more than of the time . This tie in the expected frequency of nonmatching gene trees is observed in some three-taxon data sets , but not in others , including the Drosophila data set . Although this deviation from symmetry can be explained by a model of population subdivision , where the subdivision must occur in the internal branch as well as the population ancestral to all three species [43] , the asymmetry can also be explained by the simplest hybridization network on three species with just one hybridization parameter ( Figure 3 ) . We considered six candidates for the species phylogeny: three with no hybridization , and three with hybridizations involving different pairs of species ( see Figure 3 ) . For the three phylogenetic trees , we estimated the time that maximizes the probability of observing all gene trees , and for the three phylogenetic networks , we additionally estimated the hybridization probability . The results in Table 3 show that of the three phylogenetic trees , the one in Figure 3A provides the best fit of the data , which is in agreement with the analysis in [41] . In fact , the value of we estimated on the other two trees was the lowest value we used in the estimation procedure . Clearly , this value can be arbitrarily small for these two trees , since the unresolved phylogeny ( Dmel , Dere , Dyak ) fits the data better . Among the three network candidates , the one in Figure 3D has the best fit of the data . This network , with a value of , indicates that of the alleles sampled from Dere shared a common ancestor first with alleles from Dyak ( reflecting the tree in Figure 3A ) , while of the alleles from Dere shared a common ancestor first with alleles from Dmel ( reflecting the tree in Figure 3B ) . Indeed , this network is the smallest network ( in terms of the number of reticulation nodes ) that reconciles both trees . Further , the change in AIC for this network is , indicating a much better fit than the best tree ( Figure 3A ) . As noted previously [43] , a -square test will also strongly reject the hypothesis that the species relationships are tree-like with random mating . This three-taxon example can be analyzed analytically . Fitting a hybridization parameter allows a perfect fit to any observed frequencies of gene tree topologies for three species for one of the three networks in Figure 3 . We let , , and represent the probabilities of topologies ( Dmel , ( Dere , Dyak ) ) , ( ( Dmel , Dere ) , Dyak ) , and ( ( Dmel , Dyak ) , Dere ) under the network in Figure 3D . Then This system has the unique solution ( 9 ) for and ( either at least one of the gene tree probabilities is less than if since they sum to 1 . 0; or if they are all exactly 1/3 , then a star tree with and any exactly fits the data ) . Thus we can estimate and using the observed and in equation ( 9 ) , and this also maximizes the likelihood . For the simulated data , we evolved gene trees within the branches of phylogenetic networks , while varying branch lengths and hybridization probabilities , and investigated two questions: ( 1 ) how much data ( gene trees ) is needed to obtain accurate inference of the parameters ( branch lengths and/or hybridization probabilities ) ? ( 2 ) are the parameters always identifiable ? To answer these two questions , we investigated six different phylogenetic network topologies that involved single reticulation scenario , two reticulation scenarios ( dependent and independent ) , and cases with extinctions involving the species that hybridize ( see Text S1 ) . Our results show that both hybridization probabilities and branch lengths can be estimated with very high accuracy provided that no extinction events were involved in the parents of hybrid populations ( see Text S1 ) . Further , this accuracy can be achieved even when using the smallest number of gene trees we used in our study , which is 10 . Under these settings , estimates using our framework seemed to converge quickly to the true values . We also investigated the performance of the method , as well as identifiability issues when phylogenetic signal from at least one of the species involved in the hybridization is completely lost . Figure 4 shows the results for one such scenario ( see Text S1 for another scenario that involves the loss of phylogenetic signal from both species involved in the hybridization ) . Panels Figure 4B–4D show that when the true values of and are assumed to be known in the estimation procedure ( the value of is irrelevant in the case when a single allele is sampled per species ) , the estimates of the hybridization probabilities converge to the true values . However , unlike the cases that did not involved extinctions , a larger number of gene trees is now required to obtain an accurate estimate ( while there are only three possible gene tree topologies , a large number of gene trees need be sampled in order for the three topologies' frequencies to be informative ) . The time intervals of coalescent units amount to a large extent of deep coalescence events , which blurs the phylogenetic signal , and results in slight over- or under-estimation of the hybridization probabilities ( Text S1 shows the results for the time interval with ) . If the topology of the network in Figure 4A is assumed to be known , but both the branch lengths and hybridization probabilities are to be estimated , then these parameters are unidentifiable; that is , two different pairs of vectors of branch lengths and hybridization probabilities can be found to explain the observed data with exactly the same probability ( see Text S1 ) . If at least two alleles are sampled from species B , then the parameter values become identifiable; however , an extremely large , and potentially infeasible , number of gene trees need to be sampled to uniquely identify the parameter values in practice ( see Text S1 ) . Furthermore , in the special case where , a phylogenetic tree , with appropriate branch lengths can be found , to fit the data exactly with the same probability that the phylogenetic network would . Let be the branch lengths vector with , , and , and let be the hybridization probabilities vector with . Now , consider the phylogenetic tree in Figure 4E . Then , if we set as a function of , , and , using , then , for any gene tree . The values of are shown in Figure 4F–4H . These results show that as increases , the value of becomes unaffected by , and that increasing proportionally to the increase in always maintains identical probabilities of gene trees under both species phylogenies ( see Text S1 ) . Our method for computing the probability of gene trees under hybridization and deep coalescence allows for analyzing data sets with arbitrary complexity of evolutionary histories in terms of the hybridization scenarios . When parameters are identifiable , our method estimates their values with high accuracy from a relatively small number of loci . Further , our method can be used to show lack of identifiability of model parameters for other cases . Our method supports a hypothesis of larger divergence time coupled with hybridization over short divergence times ( with extensive deep coalescence ) in a yeast data set . Finally , for a large Drosophila data set , our method indicated no hybridization based on the sampled loci .
We have focused on calculating probabilities of gene tree topologies and using these probabilities to infer species networks . In addition , the joint density of the coalescence times and topology in the gene trees could be used to infer species networks . Indeed , this approach has been used for networks where reticulation nodes have one descendant which is an extant species [29] , using the density for coalescence times derived by Rannala and Yang [44] . This approach is computationally faster than computing gene tree topology probabilities because it is not necessary to sum over a large number of coalescent histories . To compute this joint density , each gene sampled can potentially have to trace through up to possible paths through the network , where is the number of hybridization events ancestral to the sampled gene from species , and the density will take the form of a sum over possible paths through the network . ( In contrast , computing the probability of a topology will require mappings of alleles to the MUL-tree , and each gene topology calculation will require summing over coalescent histories . ) This joint density for the gene trees with coalescence times could then be used in either maximum likelihood or Bayesian frameworks to infer the species network . An important advantage of using coalescence times is that certain networks might be identifiable using coalescence times when probabilities of topologies might not identify the network . In the example of Figure 3A , although the gene tree topology probabilities can be obtained by a tree , the distribution of the coalescence times between lineages sampled from B and C is a mixture of three shifted exponential distributions if , but a mixture of two shifted exponential distributions if . For example , if , and are known but and are unkown , then the likelihood of observing a coalescence between a B and C lineage for times slightly greater will be very low if , and much higher for , thus making it possible to test whether when coalescence times are used . Another identifiability issue is that both population subdivision and hybridization can lead to the asymmetry in gene tree topology probabilities in the 3-taxon case such as observed in the Drosophila example discussed earlier , where the two least frequently observed topologies are not tied in frequency . Either population subdivision , with a parameter describing the probability that the two most closely related species fail to coalesce in the ancestral population due to population structure , or hybridization can fit the data for the gene tree topologies . However , the two models could imply different distributions on coalescence times , which might therefore be useful in distinguishing the models . We note that identifiability in the case of three species with one individual per species might be especially limited due to the small number of gene tree topology probabilities that can be used to estimate parameters . In the case of identifying rooted species trees from unrooted gene trees with one lineage per species , for example , identifiability is achieved only with 5 or more species [17] . We consider it desirable to develop many methods for inferring species trees and species networks so that their properties and performances can be compared . In the case of species tree inference , there are advantages and disadvantages to using topology-based methods versus methods that include branch lengths , and in using likelihood versus Bayesian methods . We expect that many of these strengths and weaknesses may carry over to the case of inferring networks . For moderately sized data sets , Bayesian methods that model branch lengths and uncertainty in the gene trees such as BEST [45] and *BEAST [46] often have the best performance [47] . However , these methods require estimating the joint posterior distribution of the species tree and gene trees and therefore are difficult to implement for large numbers of loci . Maximizing the likelihood of the gene trees and their coalescent times ( but without accounting for uncertainty in the gene trees ) , as in STEM [48] , is fast and has very good performance on known gene trees but seems to be very sensitive to the assumption that branch lengths are estimated correctly [24] , [49] . Maximizing the likelihood of the species tree using only gene tree topologies using the program STELLS , even while not accounting for uncertainty in the gene trees , tended to have better performance than STEM for a large simulated data set ( loci on 8 taxa ) and worse performance on fewer loci [24] . Which method is optimal for inferring species trees or networks might depend on many factors such as the number of loci , the number of lineages sampled per species , the accuracy with which branch lengths can be estimated , the extent to which there are model violations , and the speciation history [49] . Two common assumptions in multispecies coalescent models are that there is no recombination within loci ( and free recombination between loci ) and that ancestral population sizes are constant . Recombination can lead to different portions of a gene alignment effectively having distinct gene tree topologies . Ideally , alignments should be chosen so that recombination within genes is unlikely . This can be achieved by testing alignments beforehand for recombination using many available methods [50]–[52] , or for whole genome data , choosing the cutoffs for loci such that they are unlikely to occur at recombination breakpoints [53] . In addition , recombination may lead to greater violations of the coalescent model for branch lengths than for topologies [53] , so that topology-based methods might be less sensitive to the assumption that there is no recombination within loci . In addition , a recent simulation study found that recombination within loci did not have much impact on species tree inference methods for a wide range of recombination rates [54] . Coalescent models often assume that ancestral populations have constant size for the duration of the population ( i . e . , a constant size for a given branch of the species tree , but not necessarily the same on different branches ) . The program *BEAST [46] allows for ancestral population sizes to change linearly with time . Nonconstant population sizes will tend to result in branch lengths that make topologies more ( or less ) star-like for populations that are increasing ( or decreasing ) in size [55] . One approach to modelling a changing population size would be to break up a branch into intervals that are relatively constant in size . Suppose , for instance that a branch consists of an interval of generations with population size , and generations with size . The total time of the branch in coalescent units is . Although unequal values of can affect the distribution of coalescence times ( for example , if but , then coalescence events might be more likely to occur in the interval with size ) , the probabilities of topologies arising in this branch are not affected and can be calculated just using the total time . In particular , for the functions , which are the terms that depend on time in the calculations for gene tree topology probabilities , we havewhich is an instance of the Chapman-Kolmogorov equations because the number of lineages is a continuous time Markov chain ( a death chain ) [56] . We expect that topology-based methods may show more robustness to recombination and changing population sizes than approaches which explicitly model coalescence times . However , for estimating species trees and networks from gene trees , as in other areas of statistical inference , there is likely to be a tradeoff between power and robustness for methods that do and do not model branch lengths of the gene trees . A current limitation to the procedure we have outlined for estimating hybridization is that we require a set of candidate networks on which to perform model selection . In some cases , such a set of candidate networks can be obtained by considering specific hypotheses related to biogeographical information . Candidate networks can also be generated using supernetworks from gene trees [57] or other network methods [9] . Often these methods will generate very complicated networks if there are many conflicts in the data , so it might be useful to choose different random subsets of well-supported ( or frequently occurring ) gene tree topologies to generate candidate species networks . In the future it will be desirable to develop algorithms that directly search the space of species networks in order to automate searching for optimal species networks . | Species trees depict how species split and diverge . Within the branches of a species tree , gene trees , which depict the evolutionary histories of different genomic regions in the species , grow . Evolutionary analyses of the genomes of closely related organisms have highlighted the phenomenon that gene trees may disagree with each other as well as with the species tree that contains them due to deep coalescence . Furthermore , for several groups of organisms , hybridization plays an important role in their evolution and diversification . This evolutionary event also results in gene tree incongruence and gives rise to a species phylogeny that is a network . Thus , inferring the evolutionary histories of groups of organisms where hybridization is known , or suspected , to play an evolutionary role requires dealing simultaneously with hybridization and other sources of gene tree incongruence . Currently , no methods exist for doing this with general scenarios of hybridization . In this paper , we propose the first method for this task and demonstrate its performance . We revisit the analysis of a set of yeast species and another of Drosophila species , and show that evolutionary histories involving hybridization have higher support than the strictly diverging evolutionary histories estimated when not incorporating hybridization in the analysis . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"genetics",
"biology",
"computational",
"biology",
"evolutionary",
"biology",
"population",
"biology",
"genetics",
"and",
"genomics"
] | 2012 | The Probability of a Gene Tree Topology within a Phylogenetic Network with Applications to Hybridization Detection |
During transcription , the nascent pre-mRNA undergoes a series of processing steps before being exported to the cytoplasm . The 3′-end processing machinery involves different proteins , this function being crucial to cell growth and viability in eukaryotes . Here , we found that the rna14-1 , rna15-1 , and hrp1-5 alleles of the cleavage factor I ( CFI ) cause sensitivity to UV-light in the absence of global genome repair in Saccharomyces cerevisiae . Unexpectedly , CFI mutants were proficient in UV-lesion repair in a transcribed gene . DNA damage checkpoint activation and RNA polymerase II ( RNAPII ) degradation in response to UV were delayed in CFI-deficient cells , indicating that CFI participates in the DNA damage response ( DDR ) . This is further sustained by the synthetic growth defects observed between rna14-1 and mutants of different repair pathways . Additionally , we found that rna14-1 suffers severe replication progression defects and that a functional G1/S checkpoint becomes essential in avoiding genetic instability in those cells . Thus , CFI function is required to maintain genome integrity and to prevent replication hindrance . These findings reveal a new function for CFI in the DDR and underscore the importance of coordinating transcription termination with replication in the maintenance of genomic stability .
All cells are continuously exposed to DNA damaging agents , which can arise from exogenous sources or from endogenous metabolic processes . The DNA damage response ( DDR ) includes the activation of checkpoints and induction of DNA repair pathways . DNA lesions can generate structural distortions that interfere with basic cellular functions like transcription and replication . Such helix-distorting DNA lesions are generally handled by nucleotide excision repair ( NER ) , which can be divided into global genome repair ( GG-NER ) and transcription-coupled repair ( TC-NER ) sub-pathways , depending on whether the DNA lesion is located anywhere in the genome or on the transcribed strand ( TS ) of an active gene , respectively . At transcribed genes , TC-NER acts when elongating RNA polymerase ( RNAP ) stalls at bulky DNA lesions such as UV-induced cyclobutane pyrimidine dimers ( CPDs ) ( reviewed in [1] , [2] ) . Transcription down-regulation and proteasome-mediated degradation of engaged RNAPII take place as part of the DDR to UV-induced damages [3] , [4] . In humans , defects in TC-NER are responsible for two severe genetic disorders called Cockayne Syndrome ( CS ) and UV Sensitivity Syndrome ( reviewed in [5] , [6] ) . In S . cerevisiae , the major TC-NER factor is Rad26 , the yeast homologue of CS protein B ( CSB ) [7] . However , residual TC-NER activity remains in the absence of Rad26 , indicating that other factors are also involved in the process [7] , [8] . Mutations in several transcription and messenger ribonucleoprotein ( mRNP ) biogenesis factors including the RNAPII subunit Rpb9 , THO , THSC/TREX-2 , Paf1 , and Ccr4-NOT are partially defective in TC-NER in yeast [9]–[12] . During the past few years it has become clear that the different mRNA processing steps ( including 5′-end capping , splicing , and 3′-end cleavage ) , mRNP export , and transcription are connected to each other ( reviewed in [13] ) and that surveillance mechanisms ensure that these processes occur in a coordinated manner ( reviewed in [14] ) . THO and THSC/TREX-2 both work at the interface between transcription elongation , mRNP biogenesis and export and defects are characterized by a strong transcription-dependent hyperrecombination phenotype ( reviewed in [15] , [16] ) . THO might also act in the process of transcription termination , as in vitro assays suggest that THO mutants lead to polyadenylation defects [17] . Interestingly , other factors required for proficient TC-NER also function during transcription termination . The Paf1 transcription elongation factor contributes to the recruitment of 3′-end processing factors necessary for accurate transcription termination ( reviewed in [18] ) . The Ccr4-NOT complex acts , among other gene expression functions , during transcription elongation and interacts with mRNP export factors ( reviewed in [19] ) . In the yeast Saccharomyces cerevisiae , the transcription termination machinery can be divided into three different sub-complexes: cleavage factor IA ( CFIA ) , cleavage factor IB ( CFIB ) , and cleavage and polyadenylation factor ( CPF ) . CFIA is comprised of the Rna14 , Rna15 , Pcf11 , and Clp1 proteins . CFIB consists of the RNA-binding protein Hrp1 , which is tightly associated with CFIA . The CPF complex is a large complex that can be further classified into the cleavage factor II ( CFII ) made out of the Cft1 , Yhh1 , Pta1 , Brr5 , Ysh1 , Cft2 , and Ydh1 proteins; the polyadenylation factor I made of Fip1 , Yth1 , and Psf1; and other proteins including the Pap1 polymerase . In vitro reconstitution of the cleavage reaction demonstrated that it requires the joint action of CFIA , CFIB , and CFII [20] , [21] , while additional proteins such as the 5′-3′-exoribonuclease Rat1 are required for termination downstream of poly ( A ) sites in vivo and dismantling of RNAPII complexes in vitro [22]–[24] . In addition to their role in cleavage , many of the components of the cleavage machinery are required for transcription termination downstream of the poly ( A ) site and polyadenylation of the transcript ( reviewed in [25] , [26] ) . Notably , the CFIA rna14-1 and rna15-1 mutants suffer from transcription elongation defects and increase in transcription-dependent hyper-recombination [27] , suggesting that the CFIA complex serves important functions in transcription beyond termination and 3′-end processing . To assess the possible function of RNA 3′-processing and transcription termination on TC-NER , we analysed the impact of a number of mutations on the DDR and the repair of UV-induced lesions . We found that CFI mutants become sensitive to UV in the absence of GG-NER , but surprisingly are proficient for CPD repair . By contrast , DDR is compromised in those cells , as seen by RNAPII degradation and checkpoint activation analyses upon UV irradiation . In addition , we show that rna14-1 cells are impaired in cell cycle progression and rely on a functional G1/S checkpoint to prevent genomic instability and cell death . Our study reveals that CFI functions in DDR and is required for genomic integrity maintenance in yeast .
We first analysed the sensitivity of several transcription termination mutants to DNA damage in the absence of Rad7 , a protein required for GG-NER in yeast . Growth of each double mutant was compared to the growth of rad7Δ after irradiation with UV light and in the presence of the UV-mimetic agent 4-nitroquinoline 1-oxide ( 4-NQO ) ( Figure 1A ) . The rna14-1 rad7Δ , rna15-1 rad7Δ , and hrp1-5 rad7Δ double mutants were significantly more affected by UV irradiation or 4-NQO than the respective single mutants , while the remaining assayed alleles ( pcf11-2 , rat1-1 , and yhh1-3 ) were not . Notably , deletion of the RAD26 gene , which encodes the main TC-NER factor , further increased the sensitivity of rna14-1 rad7Δ and hrp1-5 rad7Δ mutants , indicating that the rna14-1 and hrp1-5 alleles are not epistatic to rad26Δ ( Figure 1B ) . Because UV sensitivity in the absence of GG-NER is a phenotype mostly associated with TC-NER deficiencies , we tested whether functional CFI was required for proficient TC-NER by monitoring the repair rates of the transcribed ( TS ) and non-transcribed ( NTS ) strands of the constitutively expressed RPB2 gene in rna14-1 , rna15-1 , and hrp1-5 cells ( Figure 2 , A and B ) . With the exception of the 60 min . time-point in rna14-1 , which is seemingly lower than the wild type on the TS , no significant differences were observed between the repair rates of the mutants and the wild type in either RPB2 strand . Repair experiments were thus performed in rad7Δ and rna14-1 rad7Δ cells . As can be seen in Figure 2 ( A and B ) , both strains show a similar low repair on the NTS and are repair-proficient on the TS . Together , our results indicate that the rna14-1 , rna15-1 , and hrp1-5 mutants are repair-proficient for CPDs . Because deficiencies in NER may cause an increase in recombinational repair and rna14-1 cells show moderate hyper-recombination [27] , we assessed whether recombination increased upon UV irradiation in rna14-1 , rad7Δ , and rna14-1 rad7Δ cells . For this , we used a direct-repeat ( LY ) and an inverted-repeat ( SU ) plasmid-based system [28] . As expected , rad7Δ cells show an increase in recombination upon UV-damage in both systems ( 13- and 35-fold , Figure S1 ) . However , recombination frequencies did not increase upon UV irradiation in rna14-1 cells , suggesting that UV damage is efficiently repaired by NER . Notably , the rna14-1 rad7Δ double mutant shows UV-dependent increase in recombination frequency as compared to the rad7Δ mutants in the direct-repeat system -but not in the inverted-repeat system- suggesting that these cells suffer from increased genomic instability that is not linked to increased CPD repair deficiencies . The cellular response to UV-induced damage involves , in addition to checkpoint activation , proteosomal degradation of RNAPII [3] . To check the functionality of the DDR in rna14-1 cells , we analysed the stability of Rpb1 , the largest subunit of RNAPII , and activation of the Rad53 checkpoint protein upon UV irradiation by Western analysis ( Figure 2 , C and D ) . Interestingly , UV-induced Rpb1 degradation was less pronounced and severely delayed in rna14-1 cells as compared to the wild type . Activation of the DNA-damage checkpoint , monitored by the appearance of hyper-phosphorylated Rad53 upon UV irradiation was delayed in rna14-1 cells as compared to the wild type , in which Rad53 phosphorylation occurs immediately upon UV irradiation . In addition , the rna14-1 mutation did not increase the sensitivity to UV or 4-NQO of cells lacking either one of the DNA-damage checkpoint proteins Rad9 and Mec1 ( Figure S2 ) , suggesting that CFI might act within the canonical checkpoint pathways . To gain more insights into the function of CFI in the cellular response to UV-induced damage , Rpb1 stability and Rad53 phosphorylation were also analysed in cells bearing the rna15-1 , hrp1-5 and pcf11-2 mutations ( Figure S3 ) . Both rna15-1 and pcf11-2 cells were partially impaired in UV-induced Rpb1 degradation while hrp1-5 cells behaved similarly to the wild type . However , Rad53 phosphorylation was delayed in the rna15-1 and hrp1-5 mutants but not in pcf11-2 cells . These interesting results suggest that UV-induced Rpb1 degradation might not depend on Rad53 activation . Previously , deletion of the DEF1 gene was shown to increase the sensitivity to UV in the absence of GG-NER without affecting DNA repair at the molecular level and to impair UV-dependent Rpb1 degradation [29] . Thus , we assayed viability and sensitivity of rna14-1 def1Δ , rna15-1 def1Δ , hrp1-5 def1Δ , and rat1-1 def1Δ double mutants to assess possible genetic interactions and observed strong synthetic sickness even in the absence of exogenous damage in all strains except hrp1-5 def1Δ ( Figures 2E and S4 ) . These interesting genetic interactions suggest that Def1 and CFI might have complementary functions for cell growth , which eventually rely on alternative ways to regulate RNAPII turnover . Although the penetrance of the different alleles differs depending on the analysed phenotype , our data indicate that CFI is required for the cellular response to UV-induced damage . Sensitivity analysis of different termination mutants to distinct DNA damaging agents revealed that the rna14-1 , rna15-1 , and hrp1-5 mutants were sensitive to Phleomycin and to methyl methansulfonate ( MMS ) in contrast to the pcf11-2 , rat1-1 , and yhh1-3 cells , which were either slightly or not sensitive to those genotoxic agents ( Figure 3A ) . Interestingly , the three alleles conferring significant sensitivity were those that increase the UV-sensitivity of rad7Δ mutants . To assess whether this phenotype was general rather than specific to GG-NER , we generated double mutants of rna14-1 with mutations in representative genes with known functions in the different DNA repair pathways , including homologous recombination ( HR ) , non-homologous end joining ( NHEJ ) , post-replicative repair ( PRR ) , mismatch repair ( MMR ) , base excision repair ( BER ) and NER ( Figure 3B ) . Interestingly , the rna14-1 mutant showed synthetic growth defects even in the absence of exogenous damage with several repair mutants , including rad52Δ , ku70Δ , lig4Δ , and rad1Δ . These growth defects are further sustained by DNA content profiling FACS analysis ( Figure S5 ) . In addition , synthetic UV/4-NQO sensitivity was observed in all double mutants but rna14-1 ogg1Δ ntg1Δ ntg2Δ . Thus , our results indicate that Rna14 dysfunction makes cells unable to cope with high levels of DNA damage and rely on functional repair pathways even in the absence of exogenous damage . To check whether these genetic interactions might arise from expression defects of DNA repair genes , mRNA expression was analysed by microarrays in rna14-1 and rna15-1 cells ( Table S1 ) . The results obtained with the two mutants were highly similar ( Figure S6 ) . Analysis of gene ontology terms of genes with higher ( > 2-fold ) and lower ( < 2-fold ) expression as compared to wild-type levels revealed that many genes involved in the DNA damage and/or stress response are induced in these mutants ( Table S2 ) , including genes such as OGG2 , PRX1 , DNL4 , LIF1 , RAD2 or MAG1 . In addition , we found out that in rna14-1 or rna15-1 cells , the down-regulated genes were on the average longer than those of the entire genome , while the up-regulated genes were shorter ( Figure S6 ) , but DNA repair genes were not specifically down regulated . Thus the results rule out that the reduced capability of CFI mutants to withstand DNA damage is due to reduced transcription of repair protein encoding genes . On the contrary , the elevated expression of DNA damage and/or stress response transcripts suggests that CFI mutants may accumulate DNA damage or structures that impose a steric hindrance to DNA metabolic processes . Transcription and replication need to occur in a coordinated manner in order to avoid conflicts that can result in genetic instability ( reviewed in [30] , [31] ) . To assess whether the CFI dysfunction affects replication , we first analysed sensitivity of several mutants to hydroxyurea ( HU ) , a drug that slows replication down by reducing the pool of available deoxyribonucleotides ( Figure 4A ) . Notably , the alleles that conferred sensitivity to HU were rna14-1 , rna15-1 , and hrp1-5 , while the others did not at concentrations assayed . Since the expression of genes encoding ribonucleotide reductase components were not affected in rna14-1 and rna15-1 ( Table S1 ) , the observed HU sensitivity might reflect DNA replication impairment . Next we analysed plasmid loss in rna14-1 cells as a way to measure replication efficiency genetically ( Figure 4B ) . Our results show that less than 5% rna14-1 cells maintained the pRS315 centromeric plasmid after about 10 divisions in non-selective medium as compared to the 50% value of wild-type cells . FACS analysis of cell cycle progression upon release from α-factor-mediated G1-arrest revealed that rna14-1 mutants remain trapped in G1 and suffer from a delay in S-phase entry as compared to the wild type ( Figure 4C ) . For a specific analysis of initiation and progression of replication , we monitored BrdU incorporation upon release from α-factor-mediated G1-arrest at three different early origins ( Figure 4D ) . DNA was immunoprecipitated with anti-BrdU antibody and BrdU enrichment at each locus was analysed by real-time qPCR with specific primers . Importantly , strong defects in replication were observed in rna14-1 mutants , as ARS activation was significantly reduced and occurred at later time points than in wild-type cells . Thus , cell-cycle progression is severely compromised in rna14-1 cells . Because G1 to S-phase progression was markedly delayed in rna14-1 cells , we asked whether persistent G1/S checkpoint activation might be responsible for the observed cell-cycle delay . Deprivation of Sic1 , a protein that is required for the G1/S checkpoint , suppressed the S-phase entry defects in the rna14-1 mutant upon release from α-factor-mediated G1-arrest as seen by FACS analysis ( Figure S7 ) . To evaluate the consequences of forcing S-phase entry in rna14-1 mutants by SIC1 deletion , we analysed phosphorylated H2A ( H2A-P ) levels by Western analysis ( Figure 5A ) . Our results indicate that the rna14-1 sic1Δ mutant accumulates DNA damage , as seen by the large amount of H2A-P . We then analysed recombination and Rad52-foci accumulation to gain insight into the impact of G1/S-checkpoint bypass in rna14-1 cells . As rna14-1 sic1Δ shows severe growth defects at 30°C ( Figure S8 ) , recombination was scored at 26°C , a semi-permissive temperature for the rna14-1 mutant , in a direct-repeat ( LYΔNS ) as well as an inverted-repeat ( TINV ) plasmid-based system [28] ( Figure 5B ) . A significant increase in recombination frequency was observed in the double rna14-1 sic1Δ mutants with respect to the frequencies of either single mutant in both systems . Rad52-foci accumulation was monitored in cells transformed with plasmid pWJ1344 expressing a Rad52-YFP fusion protein using fluorescence microscopy . As can be seen in Figure 5C , the percentage of S/G2 cells with Rad52-foci was significantly higher in the rna14-1 sic1Δ double mutant ( ≈35% ) than in the single mutants ( <20% ) . Altogether , these results indicate that a functional G1/S checkpoint is essential to avoid genomic instability and/or cell death in rna14-1 cells .
In this study , we asked whether transcription termination might contribute to DNA repair by TC-NER in S . cerevisiae . We found that the rna14-1 , rna15-1 , and hrp1-5 alleles of CFI confer increased UV and 4-NQO sensitivities in the absence of GG-NER , but surprisingly do not affect CPD repair in a transcribed gene . Importantly , we show that both checkpoint activation and RNAPII degradation are delayed in UV-irradiated CFI-deficient cells and that the rna14-1 mutation interacts genetically with mutations affecting several DNA repair pathway , including HR , NHEJ , MMR , PPR , and NER , in some cases even in the absence of exogenous DNA damage . Our data indicate that CFI participates in DDR in yeast and that this function is needed to cope with high amount of DNA damage . Additionally , we demonstrate that the rna14-1 mutation leads to severe cell cycle progression hindrance and that a functional G1/S checkpoint becomes essential in restraining genomic instability when CFI function is impaired . Although the precise mechanisms underlying termination downstream of poly ( A ) sites and 3′-end processing of RNAPII-transcribed genes remains unresolved , it certainly requires cooperation among several factors , including CFI , CPF , Pap1 , Rat1 and even the RNAPII holoenzyme ( reviewed in [32] , [33] ) . CFIA is progressively recruited to RNAPII during elongation and peaks at poly ( A ) sites [34] , [35] . Its role in transcription termination and 3′-end processing is recapitulated by ongoing transcription past poly ( A ) sites and in vitro cleavage and polyadenylation defects in CFI mutants [36]–[38] . The CFIB factor Hrp1 binds throughout transcribed genes [39] and displays in vitro cleavage and polyadenylation defects when mutated [40] , [41] . We found that CFIA rna14-1 and rna15-1 as well as the CFIB hrp1-5 alleles increased the UV and 4-NQO sensitivities of cells deficient in GG-NER and led to Phlemomycin and MMS sensitivities while the CFIA pcf11-2 , CPF yhh1-3 , and the rat1-1 alleles did not ( see Figures 1 and 3A ) . On the other hand , UV-induced Rpb1 degradation is impaired in rna14-1 , rna15-1 and pcf11-2 but not in hrp1-5 while Rad53-phosphorylation upon UV irradiation is delayed in rna14-1 , rna15-1 and hrp1-5 but not in pcf11-2 cells ( Figures 2C , 2D and S3 ) . Thus it appears that the penetrance of each particular mutation depends on the assayed phenotype . Indeed , different pcf11 alleles differ in phenotype strength as seen by RNAPII chromatin immunoprecipitation ( ChIP ) on the ADH1 and PMA1 genes [42] . However , transcriptional read-through or 3′-end processing defects alone might not be sufficient to impair the DDR as ongoing transcription past poly ( A ) sites are also observed in yhh1-3 and rat1-1 mutants , and yhh1-3 is deficient in 3′-end cleavage and polyadenylation as well [22] , [36] , [43] . One possibility could be that the requirement of CFI function for the DDR could rely on intrinsic sensing activity or specific interaction with DDR factors , thus enabling CFI to coordinate transcription termination and DDR . UV irradiation was shown to lead to 3′-end processing inhibition along with targeted RNAPII degradation in human cells , these responses seemingly being mediated by direct interaction between CstF , the functional homologue of yeast CFI , and BRCA1/BARD1 [44] , [45] . The link between DDR and 3′-end processing is further supported by the observations that partial depletion of the CstF-50 subunit leads to increased UV sensitivity , reduced ability to ubiquitinate RNAPII in response to UV and defects in CPD repair in human cells [46] . Our results show a notable divergence with respect to the human system though , as no CPD repair defects were observed in yeast CFI mutants ( see Figure 2A and 2B ) . Another difference between yeast and human is the observation that poly-adenylated mRNAs get stabilized upon UV irradiation in yeast [47] , while transcript deadenylation takes place under damaging conditions in humans , mediated by DNA damage-dependent physical interaction between CstF and the PARN deadenylase [48] . In addition , it has recently been shown that targeted variation of poly ( A ) site usage occurs in response to 4-NQO treatment in yeast , possibly as a consequence of transient depletion of CPF subunits [49] . Altogether , these findings suggest that transcription termination factors participate in DDR , a multiple-sided system fundamental for cell survival under genotoxic stress conditions . The cellular response to UV damage involves global down-regulation of transcriptional activity concomitantly with high expression of a subset of stress-induced genes and proteosomal-mediated degradation of RNAPII major subunit Rpb1 . Notably , UV-induced Rpb1 degradation is delayed in CFI-deficient cells ( see Figures 2C , 2D and S3 ) , RNAPII turnover being thus impaired . Interestingly , transcription termination factors - including CFI - interact with the transcription initiation factor TFIIB and this interaction is required for the formation of gene loops both in yeast and humans [50]–[53] . Gene looping has been proposed to enable the efficient recycling of RNAPII and to contribute to transcription regulation by acting on promoter directionality and transcriptional memory ( reviewed in [54] , [55] ) . It is thus conceivable that gene looping may also function to control transcription and RNAPII turnover under DNA damaging conditions . This idea is supported by recent work showing that TFIIB may function as a general transcriptional switch in humans , as it is dephosphorylated during genotoxic stress thus losing its interaction with CstF , while direct interaction between CstF and the p53 tumor suppressor ensures the recruitment of termination factors to the promoter of stress-induced genes [56] . The persistence of stalled RNAPII on transcribed genes is known to impede the progression of the replication machinery and to be one of the causes underlying transcription-associated recombination ( TAR ) ( reviewed in [30] , [31] ) . Recently , inhibition of Rho-dependent transcription termination in bacteria has been shown to induce double-strand breaks depending on replication , suggesting that Rho might function in the release of obstructing RNAP during replication [57] . It is possible that CFI might act on paused RNAP , whether or not stalled at a DNA damage , contributing to its displacement and thus allowing progression of an oncoming replication fork . Over the last few years , growing evidence supports a role for co-transcriptionally formed RNA-DNA hybrids ( R-loops ) as a source of TAR ( reviewed in [58] ) . Noteworthy , several transcription termination and 3′-end processing mutants have been shown to accumulate R-loops in yeast ( including pcf11-2 and rna15-58 ) [59] . It is thus possible that stalled RNAPIIs accumulate at DNA damages or other structures such as R-loop in CFI mutants , leading to steric hindrances to the replication machinery that would account for the observed cell cycle progression defects ( see Figure 4 ) . The mechanisms by which stalled RNAPIIs or structures presenting steric hindrance to replication are sensed to activate the G1/S cell cycle checkpoint , which is required to restrain genetic instability in rna14-1 cells ( see Figure 5 ) , are currently unknown . Interestingly , the Sen1/SETX helicase - a component of the NRD transcription termination complex - prevents R-loop accumulation at transcription termination sites both in yeast and humans [60] , [61] . In addition to its association with transcribed units , yeast Sen1 is also found at replication forks , contributing to prevent deleterious outcomes of the putative collisions between the transcription and replication machineries [62] . Noteworthy , Sen1 interacts physically with the NER repair protein Rad2 and the sen1-1 mutation increases the UV sensitivity of cells lacking RAD2 [63] , suggesting further connections between transcription termination , replication , and DNA repair . Altogether , our results support a model in which CFI dysfunction impairs DDR , probably leading to the accumulation of endogenous DNA lesions , and hinders DNA replication possibly due to the accumulation of RNAPs , whether or not stalled at DNA damages , thus rendering the G1/S checkpoint mandatory to prevent genomic instability ( see Figure 6 ) . Our findings emphasize the importance of coordinating transcription termination , DDR and replication in the maintenance of genomic stability and suggest that CFI plays a fundamental function in the coupling of these processes .
All strains used were isogenic to W303 , and are listed in Table S3 . Newly generated strains were obtained either by direct transformation or by genetic crosses . Plasmids used for recombination tests were pRS314-LYΔNS , pRS316-TINV , pRS314-LY and pRS314-SU [28] . For cell survival , yeast cells were grown in YEPD rich medium to an OD600 of 0 . 6 . 10-fold serial dilutions were dropped on YEPD plates , irradiated with the indicated dose of UV-C light , and incubated in the dark at 30°C for 3 days . For the 4-NQO , Phleomycin , MMS , CPT and HU sensitivity assays , the serial dilutions were dropped on YEPD plates containing the indicated amounts of genotoxic agents and incubated in the dark at 30°C for 3 days . UV survival curves were performed as described [9] . UV-C irradiation was performed using a BS03 UV irradiation chamber and UV-Mat dosimeter ( Dr . Gröbel UV-Elektronik GmbH ) . CPD repair at the RPB2 gene was analysed as described [64] . Briefly , cells were grown at 30°C in YEPD rich medium , irradiated in SD medium w/o amino acids with 200 J/m2 UV-C light ( BS03 UV irradiation chamber ) , the medium supplemented to YEPD rich and the cells incubated at 30°C in the dark for recovery . DNA from the different time-points was extracted , cut with NsiI and PvuII restriction enzymes ( Roche ) and aliquots were either treated with T4-endonuclease V ( Epicentre ) or left untreated . DNA was electrophoresed in 1 . 3% alkaline agarose gels , blotted to Nylon membranes and hybridized with radioactively labelled strand-specific DNA probes , which were obtained by primer extension . Sequences of the primers are listed in Table S4 . Membranes were analysed and quantified with a PhosphorImager ( Fujifilm FLA5100 ) . The average of the initial damage generated was 0 . 025 CPD/kb . To allow direct comparison between different strains , repair curves were calculated as the fraction of CPDs removed versus time . The initial damage was set to 0% repair . Cells were grown at 30°C in YEPD medium to an OD660 of 0 . 6 . Total RNAs were purified ( RNeasy Midi kit , Qiagen ) and expression profiling performed using the Affymetrix platform ( see Table S1 ) . The relative RNA levels for all yeast genes were determined using an Affymetrix microarray scanner and processed using the robust multiarray average method . Statistical data analyses were performed using the limma package ( affylmGUI interface ) of the R Bioconductor project ( http://www . bioconductor . org ) . For each strain , microarray analysis was conducted in triplicate . All values presented represent the average of these three determinations . Genes were considered significantly up- or down-regulated when their expression values were > or < 2-fold , respectively ( parameters: false discovery rate-adjusted p-value<0 . 01 , B-statistic value>2 , and average log2intensity A>7 ) . The expression data for each mutant has been deposited in NCBI's Gene Expression Omnibus ( accession number GSE50947 ) . Plasmid loss was monitored as the percentage of cells that lost centromeric plasmid pRS315 upon growth in non-selective media . Individual transformants were inoculated in 5 ml YEPD and grown at 30°C to OD660 0 . 6 . Cells were plated on YEPD or SC-leu to determine the percentage of plasmid loss . Six individual transformants were analysed for each genotype . Recombination frequencies were determined as the average value of the median frequencies obtained from at least three independent fluctuation tests performed at 26°C each from six independent colonies according to standard procedures [28] . Isogenic wild-type and rna14-1 strains deleted for the BAR1 gene and carrying several copies of the Herpes simplex thymidine kinase ( TK ) under the control of the strong constitutive GPD promoter were obtained by genetic crosses with strain SY2201 ( E . Schwob ) . Cells were grown in YEPD , incubated for 2 . 5 h with 0 . 125 µg/ml α-factor , washed twice with pre-warmed YEPD and released into S phase by addition of 1 µg/ml pronase . BrdU ( 200 µg/ml ) was added to the cultures prior to release . Cell cycle progression was monitored by flow cytometry on a FACSCalibur ( BD Bioscience ) using CellQuest software . Chromatin immunoprecipitation was carried out as described [65] with minor modifications . Briefly , Sodium Azide ( 0 . 1% ) was added to each sample and cells were broken in a multi-beads Shocker ( MB400U , Yasui Kikai , Japan ) at 4° in lysis buffer ( 50 mM HEPES-KOH pH 7 . 5 , 140 mM NaCl , 1 mM EDTA , 1% triton X-100 , 0 . 1% sodium deoxicholate ) and sonicated . Immunoprecipitation was performed using anti-BrdU antibody ( MBL ) attached to magnetic beads coated with Protein A ( Invitrogen ) . Input and precipitated DNA were analysed by RT qPCR ( 7500FAST Applied Biosystems ) . Relative BrdU incorporation at a given region was calculated relative to the signal at a late replicating region ( Chr . V , position 242210–242280 , [66] ) in the same sample . Primer sequences are listed in Table S4 . Rad52-YFP foci from log-phase cells transformed with plasmid pWJ1344 were visualized with a DM600B microscope ( Leica ) as previously described [67] with minor modifications . Individual transformants were grown to early-log-phase at 26°C , incubated at 30°C for 4 hours , fixed for 10 minutes in 0 . 1 M KiPO4 pH 6 . 4 containing 2 . 5% formaldehyde , washed twice in 0 . 1 M KiPO4 pH 6 . 6 , and resuspended in 0 . 1 M KiPO4 pH 7 . 4 . A total of 617 wild type , 947 rna14-1 , 733 sic1Δ , and 820 rna14-1 sic1Δ cells derived from at least three different transformants were analysed . Detection of Rpb1 , Rad53 , H2A-P , and β-actin was accomplished by Western analysis of TCA-precipitated proteins separated in 4–20% Cristerion TGX gradient PAGE ( Biorad ) . Antibodies 8WG16 ( Rpb1 , Covance ) , sc-20169 ( Rad53 , Santa Cruz Biotechnology ) , ab15083 ( H2A-P , Abcam ) and ab8224 ( β-actin , Abcam ) were used . For quantification , secondary antibodies conjugated to IRDye 680CW or 800CW ( LI-COR ) were used , the blot scanned in an Odyssey IR scanner and analysed with Image Studio 2 . 0 software ( LI-COR ) . For Western analysis after UV irradiation , cells were grown in YEPD rich medium to mid-log-phase , resuspended in SD media lacking amino acids to an OD660 of 0 . 6 and irradiated with UV-C light in a BS03 UV irradiation chamber ( Dr . Gröbel UV-Elektronik GmbH ) at 100 J/m2 . Medium was supplemented to YEPD rich and cells incubated in the dark at 30°C for recovery . | DNA damage occurs constantly in living cells and needs to be recognized and repaired to avoid mutations . DNA repair is particularly relevant for lesions occurring in actively transcribed DNA strands because the RNA polymerase cannot proceed through a damaged site . Stalled RNA polymerases and persisting DNA lesions can lead to genome instability or cell death . Specific mechanisms to repair obstructing DNA lesions are found from bacteria to higher eukaryotes , their malfunction leading to severe genetic syndromes in humans . Termination of transcription comprises cleavage and polyadenylation of the nascent transcript and displacement of the RNA polymerase from its DNA template . These processes , which are crucial for cell viability and growth in eukaryotes , require two major multi-subunit complexes in budding yeast . Here , we found that one of these complexes , Cleavage Factor I ( CFI ) , participates in the cellular response to DNA damage . In addition , we found that CFI dysfunction leads to replication defects , conceivably mediated by stalled RNA polymerases , rendering cell cycle checkpoints mandatory to prevent genomic instability . Our findings emphasize the importance of coordinating transcription termination , DNA damage response and replication in the maintenance of genomic stability suggesting that CFI plays a fundamental function in the coupling of these processes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cellular",
"stress",
"responses",
"model",
"organisms",
"rna",
"processing",
"dna",
"replication",
"nucleic",
"acids",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"molecular",
"genetics",
"yeast",
"and",
"fungal",
"models",
"biology",
"saccharomyces",
"cerevisiae",
"molecular",
"cell",
"biology",
"dna",
"dna",
"recombination",
"dna",
"repair",
"dna",
"transcription",
"gene",
"function"
] | 2014 | Cleavage Factor I Links Transcription Termination to DNA Damage Response and Genome Integrity Maintenance in Saccharomyces cerevisiae |
Hox genes in species across the metazoa encode transcription factors ( TFs ) containing highly-conserved homeodomains that bind target DNA sequences to regulate batteries of developmental target genes . DNA-bound Hox proteins , together with other TF partners , induce an appropriate transcriptional response by RNA Polymerase II ( PolII ) and its associated general transcription factors . How the evolutionarily conserved Hox TFs interface with this general machinery to generate finely regulated transcriptional responses remains obscure . One major component of the PolII machinery , the Mediator ( MED ) transcription complex , is composed of roughly 30 protein subunits organized in modules that bridge the PolII enzyme to DNA-bound TFs . Here , we investigate the physical and functional interplay between Drosophila melanogaster Hox developmental TFs and MED complex proteins . We find that the Med19 subunit directly binds Hox homeodomains , in vitro and in vivo . Loss-of-function Med19 mutations act as dose-sensitive genetic modifiers that synergistically modulate Hox-directed developmental outcomes . Using clonal analysis , we identify a role for Med19 in Hox-dependent target gene activation . We identify a conserved , animal-specific motif that is required for Med19 homeodomain binding , and for activation of a specific Ultrabithorax target . These results provide the first direct molecular link between Hox homeodomain proteins and the general PolII machinery . They support a role for Med19 as a PolII holoenzyme-embedded “co-factor” that acts together with Hox proteins through their homeodomains in regulated developmental transcription .
The finely regulated gene transcription permitting development of pluricellular organisms involves the action of transcription factors ( TFs ) that bind DNA targets and convey this information to RNA polymerase II ( PolII ) . Hox TFs , discovered through iconic mutations of the Drosophila melanogaster Bithorax and Antennapedia Complexes , play a central role in the development of a wide spectrum of animal species [1] , [2] . Hox proteins orchestrate the differentiation of morphologically distinct segments by regulating PolII-dependent transcription of complex batteries of downstream target genes whose composition and nature are now emerging [3]–[7] . The conserved 60 amino acid ( a . a . ) homeodomain ( HD ) , a motif used for direct binding to DNA target sequences , is central to this activity . Animal orthologs of the Drosophila proteins make use of their homeodomains to play widespread and crucial roles in differentiation programs yielding the very different forms of sea urchins , worms , flies or humans [8] . They do so by binding simple TAAT-based sequences within regulatory DNA of developmental target genes [9]–[14] . One crucial aspect of understanding how Hox proteins transform their versatile but low-specificity DNA binding into an exquisite functional specificity involves the identification of functional partners . Known examples include the TALE HD proteins encoded by extradenticle ( exd ) /Pbx and homothorax ( hth ) /Meis , which assist Hox proteins to form stable ternary DNA-protein complexes with much-enhanced specificity . This involves contacts with the conserved Hox Hexapeptide ( HX ) motif near the HD N-terminus , or alternatively , with the paralog-specific UBD-A motif detected in Ubx and Abdominal-A ( Abd-A ) proteins [15] , [16] . Other TFs that can serve as positional Hox partners include the segment-polarity gene products Engrailed ( En ) and Sloppy paired , that collaborate with Ubx and Abd-A to repress abdominal expression of Distal-less [17] . Finally , specific a . a . residues in the HX motif , the HD and the linker separating them play a distinctive role in DNA target specificity , allowing one Hox HD region to select paralog-specific targets [18] , [19] . Contrasting with our knowledge of collaborations involving Hox and partner TFs , virtually nothing is known of what transpires at the interface with the RNA Polymerase II ( PolII ) machinery itself to generate an appropriate transcriptional response . The lone evidence directly linking Hox TFs to the PolII machine comes from the observation that the Drosophila TFIID component BIP2 binds the Antp HX motif [20] . Another key component of the PolII machinery is the Mediator ( MED ) complex conserved from amoebae to man that serves as an interface between DNA-bound TFs and PolII . MED possesses a conserved , modular architecture characterized by the presence of head , middle , tail and optional CDK8 modules . Some of the 30 subunits composing MED appear to play a general structural role in the complex while others interact with DNA-bound TFs bridging them to PolII . Together , these subunits and the MED modules they form associate with PolII , TFs and chromatin to regulate PolII-dependent transcription [21]–[28] . Our analysis of a Drosophila skuld/Med13 mutation isolated by dose-sensitive genetic interactions with homeotic proboscipedia ( pb ) and Sex combs reduced ( Scr ) genes led us to view MED as a Hox co-factor [29] . However , how MED might act with Hox TFs in developmental processes has not been explored . The work presented here pursues the hypothesis that Hox TFs modulate PolII activity through direct binding to one or more MED subunits . Starting from molecular assays , we identify Med19 as a subunit that binds to the homeodomain of representative Hox proteins through an animal-specific motif . Loss-of-function ( lof ) Med19 mutations isolated in this work reveal that Med19 affects Hox developmental activity and target gene regulation . Taken together , our results provide the first molecular link between Hox TFs and the general transcription machinery , showing how Med19 can act as an embedded functional partner , or “co-factor” , that directly links DNA-bound Hox homeoproteins to the PolII machinery .
To search for MED subunits that contact Hox proteins directly , a Hox/MED binding matrix was established with the in vitro GST-pulldown assay . GST fusions of the eight Drosophila Hox proteins ( full-length Labial ( Lab ) , Deformed ( Dfd ) , Scr , Ubx , Abd-A and Abdominal-B ( Abd-B ) , or portions of Pb and Antp ) were probed with 35S-labelled MED proteins in standard conditions for 11 MED subunits that are not required for cell viability in yeast and/or that are known to interact physically with TFs in mammalian cells ( Materials and Methods ) . The Med19 subunit stood out , binding strongly to multiple GST-Hox proteins in this assay ( mean binding , ≥5% of input in multiple tests , except for Abd-B , 0 . 8%; Figure 1A ) . To provide an independent and in vivo test for direct Med19/Hox binding , we used the Bimolecular Fluorescence Complementation ( BiFC ) assay . Auto-fluorescence from the Venus variant of Green Fluorescent Protein ( GFP ) , abolished by truncating Venus protein into N- and C-terminal portions ( VN and VC ) , can be effectively reconstituted when VC and VN fragments are coupled to interacting protein partners that reunite them in cultured cells [30] or living organisms [31] , [32] . BiFC tests made use of the Gal4/UAS system [33] to direct co-expression of chimeric VN-Hox and Med-VC proteins . The functional validation of VN-AbdA relative to AbdA has been described [32] . VN-Ubx and VN-Dfd were likewise validated by their gain-of-function ( gof ) transformations in embryos ( Figure S1 ) . Fused Med19-VC was validated by its capacity to rescue lethal Med19 mutants ( described below ) . UAS-directed co-expression of Med19-VC with VN-Ubx , VN-Dfd or VN-AbdA in embryos under engrailed-Gal4 control ( en>Med19-VC+VN-Hox ) resulted in serial stripes of clear nuclear fluorescent signal for VN-Ubx and VN-Dfd ( Figure 1B–C ) . VN-AbdA was negative in this test ( Figure 1D , but see Figure 1H , I below ) . Nevertheless , the concordant results from GST pulldown and BiFC tests indicate direct Med19 binding to Ubx and Dfd . The observed direct Med19/Dfd or Ubx binding led us to ask whether these interactions utilize the homeodomain ( HD ) common to all Hox proteins . Consistent with this possibility , GST fusions containing the HX-HD regions of Pb ( middle , a . a . 119–327 ) and Antp ( C-ter , aa 279–378 ) bound Med19 at levels 15- to 50-fold superior to Pb N-ter ( a . a . 1–158 ) , Pb C-ter ( a . a . 267–782 ) and Antp N-ter ( a . a . 1–90 ) peptides ( Figure 1A ) . On dissecting the HD-containing C-terminal regions of Ubx , GST-Med19 bound similarly to wild-type and HX-mutated Ubx [15] , or to the linker-shortened C-terminal region of crustacean Artemia salina Ubx ( Figure S2 ) . Tests with the Antp C-ter peptide ( a . a . 279–378 ) containing its HX , linker and HD , led to the same conclusion ( not shown ) . Confirmation came from binding experiments using immobilized GST-Med19 and Antp , Ubx , Dfd and Abd-B HD peptides , where strong binding was observed for all four homeodomains despite marked divergence of their primary sequences ( 60% identity of Abd-B and Ubx HD ) ( Figure 1E ) . Contrary to the four Hox HD , neither Engrailed HD ( 43% identity with Ubx HD ) nor the TALE class Hth HD ( 21% identity ) bound detectably ( Figure 1E ) . Since Med19 binds Dfd , Antp , Ubx and Abd-B HD but not those of En or Hth , we infer that it is specific and can discriminate among homeodomains . In BiFC assays for Med19/HD association in vivo , comparable signals were observed on co-expressing Med19-VC with full-length VN-Ubx , or its HD alone ( VN-HDUbx ) , in the normal Ubx expression domain ( Ubx-Gal4; Figure 1F , G ) . Co-expressing VN-AbdA with Med19-VC under abdA-Gal4 did not give rise to fluorescence ( Figure 1D , H ) . By contrast , AbdA HD ( VN-HDAbdA ) yielded a strong fluorescent signal ( Figure 1I ) , confirming that the Hox HD is sufficient for direct Med19 binding . The differing responses obtained for full-length AbdA versus its HD raise the interesting possibility that AbdA sequences outside the HD limit its access to Med19 in vivo . Taken together , these biochemical and in vivo results indicate that the Hox HD is sufficient for direct binding to Med19 . Apart from yeast , no mutants of Med19 have been described . To address Med19 gene function in vivo , we employed imprecise excision of a viable P element insertion mutant , Med19P to generate two loss-of-function ( lof ) mutations , Med191 ( a pupal-lethal hypomorphic allele harboring a 14 base pair ( bp ) upstream deletion ) , and Med192 ( deleted for much of its protein coding sequence; see Figure 2A ) . Homozygotes for the presumptive null mutation Med192 die at the end of embryogenesis but do not show cuticular defects . To remove maternally contributed protein and/or mRNA [34] that might mask early requirements for Med19 , we used the Dominant Female Sterile technique to generate mitotic germ-line clones . When clones were induced in females heterozygous for Med192 and for ovoD1 , egg-laying was observed . Such embryos devoid of maternally contributed Med19 product undertake development ( as visualised by nuclear DAPI staining , Figure S3 ) . Though the first nuclear divisions proceed normally ( Figure S3 , ≈1 hr ) , abnormalities are already visible in pre-cellular blastoderm ( Figure S3 , ≈2 hr ) , leading to massively disorganised cellular embryos that die soon after ( Figure S3 ) . We conclude that a major maternal contribution to embryonic Med19 activity masks its zygotic roles . As an alternative approach to examining Med19 function in embryonic development , we made use of Med19-directed RNAi . Ubiquitous RNAi expression under daughterless-Gal4 control gave rise to cuticular defects in the spiracles and head of L2/L3 larvae ( Figure S4 ) reminiscent of the embryonic consequences of Abd-B lof in the posterior spiracles , or of lab/Dfd/Scr lof in the head . However , the late appearance of these defects , their incomplete penetrance and variable expressivity made it difficult to ascertain a functional link to embryonic Hox activities . We therefore decided to examine post-embryonic development , making use of partial loss-of-function combinations and of clonal analysis . Med191/Med192 animals die as pupae , but adult viability is restored by ubiquitous transgenic expression of Med19 ( Ub-Med19 or arm>Med19-VC; Figure S9 ) . These results show that lethality is due solely to loss of Med19 function . They also functionally validate the UAS-Med19-VC element used in BiFC assays . To better characterize the consequences of Med19 loss-of-function , we employed FLP/FRT-mediated mitotic recombination [35] to generate clones of cells homozygous for Med192 ( −/− ) . In “twin spot” experiments where −/− and +/+ cells arise from mitotic recombination in a single −/+ cell during mitosis , only +/+ cells were subsequently detected ( Figure 2B ) . This cell lethality is due to loss of Med19 , since expressing Med19-VC protein in mutant cells restores viability ( Figure 2C ) ; indeed , large Med192/Med192 clones are observed even though Med19-VC accumulation is less than for wild-type protein in adjacent cells ( Figure S5 , A–A″ ) . Strikingly , Med192/Med192 cells also survived in the presence of a Minute ( M ) mutation [36] that slows growth of surrounding heterozygous M−/+ cells ( Figure 2D ) . Immune staining with anti-Med19 sera confirmed that Med192 is a protein-null mutation ( Figure S5 , B–B″ ) . Thus the existence of these clones shows that Med19 is not strictly required for cell viability . The influence of cell environment on cell lethality suggests a role for Med19 in the control of cell competition . If Hox/Med19 binding is functionally relevant to homeotic activity , Med19 mutants might provoke Hox-like phenocopies or modify Hox-induced homeotic defects . In light of the strong maternal contribution of Med19 present in embryos , we turned our attention to later developmental stages . Hypomorphic Med191/Med192 animals , or Med192/Med192 animals with low-level ubiquitous expression from the Ub-Med19 transgene , survive to the pupal stage and show a fully penetrant loss of anterior spiracles ( Figure 2E , F ) . Rare adult Ub-Med19; Med192/Med192 survivors showed defects including loss of maxillary palps ( Figure 2G , H ) . Tissue-directed induction of Med192/Med192 clones in the dorsal compartment of the haltere imaginal disc ( apterous-Gal4>UAS-Flp ) is associated with disorganization of the distinctive pedicellar sensillae in halteres [37] , where these sensory organs are reduced in number and their well-ordered rows disrupted in adult halteres ( Figure 2I–I′ vs . 2J–J′ ) . All the preceding defects ( Figure 2E–J′ ) resemble Hox loss-of-function phenotypes: of Antp for the pupal anterior spiracles [38]; of Dfd for the adult maxillary palps [39] , [40]; and of Ubx for haltere sensory organs [37] . We therefore asked whether Med19 mutants can act as dose-sensitive modifiers of Hox activity in genetic interaction tests . For Antp , the fully penetrant spiracle loss observed in Med191/Med192 pupae ( Figure 2F ) is absent from Med19− or Antp− heterozygotes but can be detected in Med19−/Antp− double heterozygotes ( Figure S6 ) . This synergistic interaction links Med19 to normal Antp function . Conversely , ectopic expression from the gain-of-function ( gof ) AntpNs allele directs a fully penetrant transformation of antenna toward leg ( Figure 3A , B ) that is partially suppressed in Med192 heterozygotes , as shown by the replacement of distal claws by antennal aristae ( Figure 3C , Figure S6 ) . Adult maxillary ( Mx ) palps ( Figure 3D , arrowhead ) require Dfd function in a territory abutting the antennal primordium of the eye-antennal imaginal disc [40] , and the palp loss noted above ( Figure 2H ) suggested a link to Dfd . In interaction tests , palp loss provoked by Dfd-specific RNAi ( patched>dsRNADfd ) was enhanced in Med192 heterozygotes ( Figure S6 ) . Conversely , ectopic Mx organs observed in heterozygotes for the gof allele Dfd1 ( 20% of adult heads ) were fully suppressed in Dfd1/Med192 double heterozygotes ( Figure 3E , F and Figure S6 ) . Similarly , the transformation of the posterior wing ( Figure 3G ) to a hemi-haltere in UbxCbx1 homozygotes ( Figure 3H ) was synergistically modified on reducing Med19 activity ( Figure 3I ) . These dose-sensitive modifications of Antp , Dfd and Ubx-dependent phenotypes by Med19 lof mutants support a functional Hox/Med19 link in vivo . The intensive attention given to Ubx target gene regulation in the haltere imaginal disc [3]–[5] , [41] , [42] makes it an excellent paradigm for understanding Hox interplay with Med19 in a developmental program . We therefore examined the effect of removing Med19 function on Ubx activity towards selected target genes in the haltere imaginal disc ( Figure 4A ) . Clones of −/− cells induced in haltere discs in the presence of a Minute mutation showed Ubx levels comparable to their +/− neighbors ( Figure 4B–B′″ ) . The presence of normal nuclear Ubx signal in mutant cells after several mitotic divisions shows that the Med19− condition has not generally affected transcription . We next examined the effects of Med19−/− mitotic clones on several Ubx target genes in haltere development [41] . Ubx is known for its role in suppressing wing development , and acts to repress a number of prominent wing developmental genes in the haltere imaginal disc [41] . With the combined use of ectopic Hox expression and analysis of the transcriptome , additional Ubx targets have emerged . The first example of a Ubx-activated target was CG13222 , which is regulated through an autonomous cis-regulatory region called “edge” for its expression in a band of cells along the posterior border of the haltere disc [42] . Expression of the edge-GFP reporter construct , that recapitulates Ubx-dependent CG13222 expression [42] is cell-autonomously abolished in Med19−/− haltere clones ( Figure 4C , D; identified with anti-Med19 sera ) . This shows that Med19 is required for activation of the direct Ubx target CG13222 in the haltere disc . A second positively-regulated target identified in recent whole-genome analyses , bric-à-brac2 ( bab2 ) , is induced by ectopic Ubx [5] and correlates with direct Ubx binding to regulatory DNA in vivo [3] , [4] . Using mitotic recombination , we examined the expression of bab2 in Ubx−/− haltere cells and found that Bab2 accumulation is cell-autonomously abolished ( Figure 4E–E′″ ) . On examining bab2 expression with respect to Med19 activity , Bab2 accumulation was cell-autonomously down-regulated in Med19−/− cells ( Figure 4F–F′″ ) . This shows that Med19 activity is required in the haltere disc for normal Ubx-mediated activation of bab2 . However , contrary to edge , residual low-level bab2 expression is present in some Med19− cells ( Figure 4F–F′″ ) . The responses of these two regulatory sequences to Med19 lof indicate that Med19 is required for Ubx target gene activation . Several targets whose expression is repressed by Ubx were tested for a requirement for Med19 ( Figure 4A ) . For example , the broad central band of spalt ( sal ) expression in the wing pouch is absent from Ubx-expressing cells of the haltere pouch . sal is de-repressed in Ubx−/− haltere disc cells [41] , as shown by the appearance of Sal protein in Ubx mutant cells ( Figure 4G–G′″ ) . By contrast , no new expression of Sal is detected in equivalently placed Med19−/− clones ( Figure 4H–H′″ ) . De-repression was likewise not observed for other tested Ubx-repressed targets ( not shown ) , lending molecular support to the interpretation that Med19 is not involved in Ubx repressive activities . While further examples will be required to determine whether this illustrates a general property of Med19 action in transcriptional regulation , these results suggest that Med19 collaboration with Ubx is limited to gene activation . Binding of Hox proteins to Med19 specifically involves their conserved homeodomain . To identify Med19 sequences involved in HD binding , we used GST-pulldown to test full-length or deleted versions of Med19 with GST-HDAntp . Binding was retained on truncating the terminal regions of Med19 ( Figure 5A , constructs 1–2 ) , but was abolished on deleting an internal 70 aa region ( Figure 5A , constructs 3–4 ) . Smaller deletions confirmed that this region contains sequences required for full binding ( Figure 5A , constructs 5–6 ) . This 70 a . a . Med19 peptide is not only required but also proved sufficient to bind the HD ( Figure 5A , construct 7 ) , leading us to call it Homeodomain Interacting Motif ( HIM ) . The HIM interval of Med19 was then compared with Med19 orthologs from a spectrum of eukaryotes . Sequence alignments reveal a lysine/arginine-rich sequence that is strongly conserved in Med19 orthologs from six vertebrate or insect species ( Figure S7 ) . This striking conservation suggests that the contribution of HIM to Med19 function is subject to strong selective pressure . Med19 is required for Ubx-mediated activation of specific target genes in vivo , and directly binds Hox HDs in vitro through its conserved HIM motif . This suggested that Ubx function passes through Med19 , potentially via its HIM sequence . We therefore sought evidence for Med19/Hox binding in cellulo . Cultured Drosophila cells expressing UAS-Med19-VC or UAS-Med19ΔHIM-VC were used for co-immunoprecipitations with anti-GFP ( VC ) . As shown in the Western blot of Figure S8 , the three endogenous Med1 isoforms were associated with both Med19-VC and Med19ΔHIM-VC . This is consistent with the incorporation of full-length and HIM-deleted forms into the MED complex . We next co-expressed these proteins with Ubx-HA and tested for their association in cellulo . As seen in Figure 5B , Ubx-HA co-precipitates with Med19-VC , indicating the association of Ubx transcription factor with Med19 . By contrast , less Ubx-HA was detected on co-precipitating with Med19ΔHIM-VC ( relative to a non-specific band that serves as a de facto internal loading control , * in Figure 5B ) . These results indicate that the HIM domain contributes to Ubx-Med19 interaction . To further investigate the contribution of the HIM domain to HD binding in vivo , we generated transgenic lines containing the same UAS-Med19-VC , -Med19ΔHIM-VC and -HIM-VC constructs used above , and tested each protein's ability to bind to VN-HDUbx in the BiFC assay . In control experiments , Med19-VC , HIM-VC and Med19ΔHIM-VC accumulated at comparable levels in wing imaginal discs ( Figure S8 ) . Med19-VC and HIM-VC are fully nuclear . The Med19ΔHIM-VC protein ( lacking the highly basic HIM element ) is seen to accumulate in both the cytoplasm and the nucleus ( Figure S8 ) . This indicates that the HIM element contributes to nuclear localisation , together with other Med19 sequences . As noted in embryos ( Figure 1G ) , co-expressing Med19-VC with VN-HDUbx under dpp-Gal4 control in wing imaginal discs resulted in clear fluorescent signal ( Figure 5C ) . When HIM-VC was tested for its ability to interact with the Ubx HD , it gave rise to a fluorescent signal stronger than for intact Med19; by contrast , the Med19ΔHIM-VC fusion yielded only a background-level signal with VN-HDUbx ( Figure 5C ) . Taken together , these results of biochemical and BiFC experiments indicate that Med19 HIM is necessary and sufficient for full HD binding . While Med19 bereft of its conserved HIM element can be incorporated into MED , as shown above , its functional requirements in vivo remained an open question . Accordingly , we tested whether Drosophila HIM is relevant to Med19 developmental functions . In a genetic rescue test , ubiquitous Med19-VC expression restored adult viability to pupal-lethal Med191/Med192 hypomorphs , whereas Med19ΔHIM-VC did not . Med19ΔHIM-VC also showed a reduced aptitude to rescue pupal spiracles , adult maxillary palps and haltere sensillae compared with Med19-VC ( Figure S6 ) . These results indicate a requirement for Med19 HIM in several Hox-dependent developmental processes . We therefore sought to test the influence of the Med19 HIM peptide on Ubx-dependent transcriptional activation of the direct target CG13222/edge . To this end , ( i ) FLP/FRT-mediated mitotic recombination was used to generate cells devoid of wild-type protein , while ( ii ) UAS/Gal4-directed expression supplied normal or HIM-deleted Med19-VC , and ( iii ) the edge-GFP reporter was employed to assess Ubx-mediated activation of CG13222 . En-Gal4-directed UAS-Flp expression in the posterior haltere disc compartment served to induce mitotic recombination there , while en-Gal4 simultaneously directed expression of Med19-VC or Med19ΔHIM-VC in the posterior compartment ( Figure 6A–B , stained with anti-VC , blue ) . These twin-spot experiments provided two important observations . Firstly , large −/− clones ( RFP- ) were observed not only in Med19-VC but also in Med19ΔHIM-VC expressing discs ( Figure 6A′ , B′ ) . The existence of −/− clones is in marked contrast with their complete absence in Figure 2B . This shows that both Med19-VC and Med19ΔHIM-VC restore cell viability . We conclude that HIM is not necessary for cell viability . Further , it indicates that not only are both forms of Med19-VC incorporated into MED , but they are functional there . Secondly , Med19-VC and Med19ΔHIM-VC differed markedly in their capacities to ensure activation of the Ubx target gene CG13222 . Reporter expression was observed within all appropriately positioned clones of −/− cells expressing Med19-VC ( 11 of 11 clones; Figure 6A″ , C ) . By contrast , edge-GFP expression was entirely absent from most clones expressing Med19ΔHIM-VC ( 7 of 11 clones; Figure 6B″ , C ) . The existence of HIM-independent cell proliferation/survival shows that this cellular function of Med19 can be uncoupled from its Hox-related role . These results also provide clear functional evidence that Ubx-dependent activation of its edge-GFP target requires HIM-endowed Med19 .
Hox homeodomain proteins are well-known for their roles in the control of transcription during development . Further , much is known about the composition and action of the PolII transcription machine . However , virtually nothing is known of how the information of DNA-bound Hox factors is conveyed to PolII in gene transcription . The Drosophila Ultrabithorax-like mutant affecting the large subunit of RNA PolII provokes phenotypes reminiscent of Ubx mutants [43] , but the molecular basis of this remains unknown . The lone direct evidence linking Hox TFs to the PolII machine is binding of the Antp HX motif to the TFIID component BIP2 [20] . Here , we undertook to identify physical and functional links between Drosophila Hox developmental TFs and the MED transcription complex . Our results unveil a novel aspect of the evolutionary Hox gene success story , extending the large repertory of proteins able to interact with the HD [44] to include the Drosophila MED subunit Med19 . HD binding to Med19 via the conserved HIM suggests this subunit is an ancient Hox collaborator . Accordingly , our loss-of-function mutants reveal that Med19 contributes to normal Hox developmental function and does so at least in part via its HIM element . Thus this analysis reveals a previously unsuspected importance for Med19 in Hox-affiliated developmental functions . A fundamental property of the modular MED complex is its great flexibility that allows it to wrap around PolII and to change form substantially in response to contact with specific TFs [45] . Recent work in the yeast S . cerevisiae places Med19 at the interfaces of the head , middle and CDK8 kinase modules [46] , [47] . Med19 is thus well-positioned to play a pivotal regulatory role in governing MED conformation ( Figure 7 ) . Our results raise the intriguing possibility that MED structural regulation and physical contacts with DNA-bound TFs can pass through the same subunit . In agreement with this idea , recent work identified direct binding between mouse Med19 ( and Med26 ) and RE1 Silencing Transcription Factor ( REST ) [48] . This binding involves a 460 a . a . region of REST encompassing its DNA-binding Zn fingers [48] . The present work goes further , in identifying a direct link between the conserved Hox homeodomain and Med19 HIM ( Figure 7 ) that is , to our knowledge , the first instance for a direct , functionally relevant contact of MED with a DNA-binding motif rather than an activation domain . Med19 contributes to developmental processes with Antp ( spiracle eversion ) , Dfd ( Mx palp ) , and Ubx ( haltere differentiation ) . Other phenotypes identified with our mutants indicate further , non-Hox related roles for Med19 . As shown here , complete loss of Med19 function leads to cell lethality that can be conditionally alleviated when surrounded by weakened , Minute mutation-bearing cells . These observations , that uncouple HIM-dependent functions from the role of Med19 in cell survival/proliferation ( Figure 6B ) , are compatible with reports correlating over-expression of human Med19/Lung Cancer Metastasis-Related Protein 1 ( LCMR1 ) in lung cancer cells with clinical outcome [49] . Further , RNAi-mediated knock-down of Med19 in cultured human tumor cells can reduce proliferation , and tumorigenicity when injected into nude mice [50]–[58] . A recent whole-genome , RNAi-based screen identified Med19 as an important element of Androgen Receptor activity in prostate cancer cells where gene expression levels also correlated with clinical outcome [59] . It will be of clear interest to examine how , and with what partners , Med19 carries out its roles in cell proliferation/survival . The role played by mammalian Med19 and Med26 in binding the REST TF , involved in inhibiting neuronal gene expression in non-neuronal cells [48] , [60] , provides an instance of repressive Med19 regulatory function . We found that Med19 activity is required in the Drosophila haltere disc for transcriptional activation of CG13222/edge and bab2 , but is dispensable for Ubx-mediated repression of five negatively-regulated target genes ( Figure 4 ) . Ubx can choose to activate or it can repress , at least in part through an identified repression domain at the C-terminus just outside its homeodomain [61] . Conversely Med19 , which binds the Ubx homeodomain , appears to have much to do with activation . Concerning the mechanisms of Ubx-mediated repression , one illuminating example comes from analyses of regulated embryonic Distal-less expression [17] . Ubx can associate combinatorially with Exd and Hth , plus the spatially restricted co-factors Engrailed or Sloppy-paired in repressing Distal-less [17] . Engrailed in turn is able to recruit Groucho co-repressor [62] , suggesting that localized repression involves DNA-bound Ubx/Exd/Hth/Engrailed , plus Engrailed-bound Groucho . Groucho has been proposed to function as a co-repressor that actively associates with regulatory proteins and organizes chromatin to block transcription . Wong and Struhl [63] demonstrated that the yeast Groucho homolog Tup1 interacts with DNA-binding factors to mask their activation domains , thereby preventing recruitment of co-activators ( including MED ) necessary for activated transcription . The number of targets remains too small to be sure Med19 is consecrated to activation . Nonetheless , it will be of interest to determine whether Groucho can play a role in blocking MED/Ubx interactions that could provide an economical means for distinguishing gene activation from repression . The conserved Hox proteins and the gene complexes that encode them are well-known and widely used to study development and evolution . As to the evolutionary conservation of the Mediator transcription complex , the presence of MED constituents in far-flung eukaryotic species from unicellular parasites to humans [21] indicates that this complex existed well before the emergence of the modern animal Hox protein complexes . The DNA-binding domains are often the most conserved elements of TF primary sequence , and in the case of the Hox HD , recent forays into “synthetic biology” agree that this was the functional heart of the ancestral proto-Hox proteins [64]–[66] . Indeed , Scr , Antp and Ubx mini-Hox peptides containing HX , linker and HD motifs behave to a good approximation like the full-length forms , directing appropriate gene activation and repression resulting in genetic transformations [64]–[66] . Our results showing direct HD binding to Med19 HIM , and thus access to the PolII machinery , allow the activity of these mini-Hox proteins to be rationalized . We surmise that at the time when the Hox HD emerged to become a major developmental transcription player , its capacity to connect with MED through specific existing sequences was a prerequisite for functional success . One expected consequence of this presumed initial encounter with Med19 – a selective pressure on both partners and subsequent refinement of binding sequences – is in agreement with the well-known conservation of Hox homeodomains , and with the observed conservation of the newly-identified HIM element in Hox-containing eumetazoans . We imagine that subsequent evolution over the several hundred million years separating flies and mammals will have allowed this initial contact to be consolidated through subsequent binding to other MED subunits , ensuring versatile but reliable interactions at the MED-TF interface ( Figure 7 ) . Hox homeodomain proteins are traditionally referred to as selector or “master” genes that determine developmental transcription programs . The low sequence specificity of Hox HD transcription factors is enhanced by their joint action with other TFs , of which prominent examples , the TALE homeodomain proteins Extradenticle/Pbx and Homothorax/Meis are considered to be Hox co-factors . However , a Hox TF in the company of Exd and Hth could still not be expected to shoulder all the regulatory tasks necessary to make a segment with all the coordinated cell-types it is made up of , and collaboration with cell-type specific TFs appears to be requisite . A useful alternative conception visualizes Hox proteins not as “master-selectors” that act with co-factors , but as highly versatile co-factors in their own right that can act with diverse cell-specific identity factors to generate the cell types of a functional segment [67] . We envisage a model where a Hox protein would be central to assembling cell-specific transcription factors into TF complexes that interface with MED ( Figure 7 ) . Such Hox-anchored TF complexes could make use of selective HD binding to Med19 as a beach-head for more extensive access to MED , such that loss of the Hox protein would incapacitate the complex: in the case of Ubx− cells , inactivating bab2 or de-repressing sal . Accordingly , three observations suggest that binding of Hox-centered TF complexes involves additional MED subunits surrounding Med19 ( Figure 7 ) : ( i ) bab2 target gene expression is entirely lost in Ubx-deficient cells but can persist in some Med19− cells; ( ii ) edge-GFP in Med19− cells expressing Med19ΔHIM-VC was not altogether refractory to Ubx-activated edge-GFP expression ( Figure 6 ) ; and ( iii ) Med19ΔHIM-VC is not entirely impaired for Ubx binding , as seen in co-immunoprecipitations ( Figure 5 ) . Thus Hox protein input conveyed through Med19-HIM at the head-middle-Cdk8 module hinge might provide an economical contribution toward organizing TF complexes that influence overall MED conformation [45] and hence transcriptional output . Decoding how the information-rich MED interface including Med19 accomplishes this will be an important part of understanding transcriptional specificity in evolution , development and pathology .
Our work using Drosophila was performed in conditions in conformity with French and international standards . Culture and preparation of GST-fusion proteins , preparation of 35S protein probes , and pulldowns were carried out essentially as described in [68] . Chimeric GST-Hox constructs fused GST to Hox cDNAs . Full-length Hox fusions were used for Labial , Deformed , Sex combs reduced , Ultrabithorax , Abdominal-A and Abdominal-B . Fragments of Pb and Antp were present in the fusion proteins: Pb1 ( N-ter , a . a . 1–158 ) , Pb2 ( middle with HD , a . a . 119–327 ) and Pb3 ( C-ter , a . a . 267–782 ) . For Antp , two GST fusions were used: Antp1 ( N-ter , 1–90 ) and Antp4 ( C-ter with HD , 279–378 ) . Eleven MED putative surface subunits [21] could be expressed at useable levels in coupled in vitro transcription/translation reactions: Med1/Trap220 , Med2/Med29/Ix , Med6 , Med12/Kto , Med13/Skd , Med15/Arc105 , Med19 , Med25 , Med30 , Cdk8 and CycC . Cultured Drosophila S2 cells were transfected using FuGENE HD transfection reagent ( Roche ) with pActin-V5 ( negative control; pActin-GAL4 driver with either pUAS-Med19-VC or pUAS-Med19ΔHIM-VC ( MED co-IP ) ; or adding pUAS-Ubx-HA ( Med19-Ubx co-IP ) . 107 cells were transfected with driver plasmid plus the UAS responder plasmid ( s ) . After 72 hr , cells were harvested by scraping and pooled , collected by centrifuging then washed with 1x PBS . All subsequent steps until Western blotting were carried out at 4°C . Cell pellets were resuspended in IP buffer ( 50 mM TrisHCl , pH = 8 , 150 mM NaCl , 0 . 5% NP40 , 1 mM EGTA and Roche complete protease inhibitor cocktail ) , lysed by four-fold passage through a 27G needle , then centrifuged for 10 min at 14 , 500 rpm . Immunoprecipitation from 1 . 5 mg of total protein extract ( 5 µg/µl in IP buffer ) was performed with mouse anti-GFP ( ROCHE 4 µg/IP ) , with gentle agitation overnight . 15 µl of G-protein-coupled Sepharose beads ( SIGMA , P3296 ) were added , then gently agitated for 2 hr . The non-bound fraction was discarded . Beads were washed 4 times with fresh IP buffer , taken up in 2X Laemmli buffer containing DTT and SDS , heated to 95°C , and centrifuged . Supernatants were then submitted to polyacrylamide gel electrophoresis . Med1 and Med19 were revealed using polyclonal sera from guinea-pig ( diluted 1∶500 ) , while Ubx-HA was detected with rabbit anti-HA ( SIGMA ) diluted 1∶1000 . Constructions corresponding to UAS-VN-Ubx , UAS-VN-AbdA and UAS-VN-HDAbdA transgenic lines are described in [32] . UAS-VN-Dfd , UAS-VN-HDUbx: Dfd and Ubx HD sequences were cloned into XhoI-XbaI sites downstream of the Venus VN fragment into the pUAST or pUASTattB plasmids described in [32] . UAS-Med19-VC , UAS-HIM-VC: Full-length Med19 coding sequences , or the internal HIM sequence generated from Med19 cDNA by PCR , were introduced as EcoRI-XhoI fragments to replace Hth coding sequences of pUaHth-VC [32] . For UAS-Med19ΔHIM-VC , internally deleted Med19 was generated from the full-length construct by double PCR , using the overlap extension method . The PCR-derived internal deletion product was cleaved by RsrII and XhoI , then cloned in place of the equivalent fragment of UAS-Med19-VC . All constructs were sequence-verified before fly transformation . Transgenic lines were established by classical P-element mediated germ line transformation or by site-specific integration using the ΦC-31 integrase . Embryos were analysed as described in [32] . For BiFC in imaginal discs , late third-instar larvae of appropriate genotypes were cultured in parallel in the same environmental conditions of temperature and larval density , then were dissected at the same time and fixed in the same solution ( 20′; 4% para-formaldehyde , 0 . 5M EGTA , 1X PBS ) . Wing and haltere imaginal discs were dissected and mounted in Vectashield ( Vector Labs ) . Image acquisition was performed on a Leica SP5 using the same laser excitation , brightness/contrast and z settings . Confocal projections from at least 10 distinct wing discs per genotype were analyzed with ImageJ software . Stocks and crosses were maintained at 25°C on standard yeast-agar-cornmeal medium . Mutant stocks harboring AntpNs , Dfd1 , UbxCbx1 and Df ( 3L ) BSC8 were from the Bloomington Drosophila Stock Collection . AntpNs+Rc3 was provided by R . Mann . Edge-GFP and vgQ-lacZ originate from the S . Carroll lab . The Ub-Med19 transgenic line expressing full-length Med19 cDNA under ubi73 control is a homozygous-viable insertion on the X chromosome at attP site ZH 2A . These transgenic elements carry the visible marker mini-white . UAS-RNAi against Dfd is from a non-directed insertion on chromosome 2 ( Vienna Drosophila Research Collection stock 50110 ) . UAS-RNAi lines against Med19 are stocks 27559 and 33710 from the Bloomington collection . Gal4-expressing driver lines used were: dpp-Gal4 ( imaginal disc-specific , AP boundary , anterior compartment ) , arm-Gal4 ( ubiquitous ) , ptc-Gal4 ( AP boundary , anterior compartment ) , en-Gal4 ( posterior compartment ) , ap-Gal4 ( dorsal compartment of wing and haltere discs ) , Ubx-Gal4 and abdA-Gal4 ( abdominal expression under Ubx or abdA control , respectively ) . Loss-of-function Med19 alleles were generated by imprecise excision , mobilizing the viable P{EPgy2}EY16159 insertion marked with mini-w+ ( see Flybase ) . Among 154 white-eyed candidates , two Med191 and Med192 ( described in Figure 2 ) were hemizygous-lethal with Df ( 3L ) BSC8 . The following stocks were employed for rescue tests: Ub-Med19; Med192/TM6B , Hu Tb; Med191/TM6B , Hu Tb; Med192/TM6B , Hu Tb; Arm-Gal4; Med191/TM6B , Hu Tb; UAS-Med19-VC; Med192/TM6B , Hu Tb; UAS-Med19ΔHIM-VC; Med192/TM6B , Hu Tb . Mitotic clones were induced by Flp recombinase expressed from a hsp70-Flp transgene on heat induction ( 30′ at 38°C ) , or from a UAS-Flp element under Gal4 control as indicated above ( en>Flp , ap>Flp ) . Clones were generated and identified in marked progeny from crosses using the following stocks: Med192 FRT-2A/TM6B , Hu Tb hsp70-Flp; Ub-GFP FRT-2A/TM6B , Hu Tb Ub-Med19; Ub-GFP FRT-2A/TM6B , Hu Tb UAS-Med19-VC; Ub-GFP FRT-2A/TM6B , Hu Tb en>Flp; Med192 FRT-2A/TM6B , Hu Tb Ub-GFP M FRT-2A/TM6B , Hu Tb ap>Flp; Med192 FRT-2A/TM6B , Hu Tb hsp70-Flp; Ub-GFP M FRT-2A/TM6B , Hu Tb edge-GFP/CyO , Cy; M FRT-2A/TM6B , Hu Tb ap>Flp; FRT-82B Ub-GFP/TM6B , Hu Tb FRT-82B Ki pb5 pp Ubx1 e/TM6B , Hu Tb en>Flp; FRT-82B Ub-GFP/TM6B , Hu Tb edge-GFP , UAS-Med19-VC/CyO , Cy; His2Av-mRFP1 FRT-2A/TM6B , Hu Tb edge-GFP , UAS-Med19ΔHIM-VC/CyO , Cy; His2Av-mRFP1 FRT-2A/TM6B , Hu Tb After crossing y w hsp70-Flp; Med192 FRT-2A/TM6B , Hu Tb females with w; P[ovoD1] FRT-2A/TM3 , Sb males , progeny at L3/early pupal stages were subjected to heat shocks ( 1 hr , 37°C ) on two successive days . Resulting y w hsp70-Flp/w; Med192 FRT-2A/P[ovoD1] FRT-2A adult females were crossed with Med192 FRT-2A/TM3 , Ser twist>GFP males . Embryos resulting from germline clones were collected on egg lay plates , then analysed by confocal microscopy after mounting in DAPI-containing Vectashield medium . In positive controls where Med19+ replaced Med192 , all expected zygotic classes were obtained as viable , fertile adults . In the absence of heat shock , no eggs were laid . Performed as described in [40] . Antibodies used were: rat anti-Bab2 ( J-L Couderc , used at 1∶3000 ) ; mouse anti-GFP ( VC ) ( Roche , 1∶200 ) or chicken anti-GFP ( VC ) ( Invitrogen ) ; rabbit anti-Spalt ( R . Barrio , 1∶100 ) ; mouse anti-Col ( M . Crozatier/A . Vincent , 1∶200 ) ; mouse anti-dSRF ( M . Affolter , 1∶1000 ) ; rabbit anti β-Gal ( Cappel 1∶2500 ) , mouse anti-Wg ( 1∶200 ) and anti-Ubx ( 1∶50 ) from the Developmental Studies Hybridoma Bank , University of Iowa . Guinea pigs were immunized ( Eurogentec ) with GST-Med19 or GST-Med1 proteins extracted from E . coli and enriched by affinity chromatography . Anti-Med19 sera from terminal bleeds was used for immunocytology without purification at a 1∶500 dilution after prior pre-absorption on wild type larvae . Adult phenotypes were analyzed by light microscopy ( Zeiss Axiophot ) of dissected samples mounted in Hoyer's medium or by scanning electron microscopy ( Hitachi TM-1000 Tabletop model ) of frozen adults These were generated with the T-Coffee Program , employing the methodology described by [21] . | Mutations of Hox developmental genes in the fruit fly Drosophila melanogaster may provoke spectacular changes in form: transformations of one body part into another , or loss of organs . This attribute identifies them as important developmental genes . Insect and vertebrate Hox proteins contain highly related homeodomain motifs used to bind to regulatory DNA and influence expression of developmental target genes . This occurs at the level of transcription of target gene DNA to messenger RNA by RNA polymerase II and its associated protein machinery ( >50 proteins ) . How Hox homeodomain proteins induce fine-tuned transcription remains an open question . We provide an initial response , finding that Hox proteins also use their homeodomains to bind one machinery protein , Mediator complex subunit 19 ( Med19 ) through a Med19 sequence that is highly conserved in animal phyla . Med19 mutants isolated in this work ( the first animal mutants ) show that Med19 assists Hox protein functions . Further , they indicate that homeodomain binding to the Med19 motif is required for normal expression of a Hox target gene . Our work provides new clues for understanding how the specific transcriptional inputs of the highly conserved Hox class of transcription factors are integrated at the level of the whole transcription machinery . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"animal",
"genetics",
"protein",
"interactions",
"enzymology",
"dna-binding",
"proteins",
"animals",
"cell",
"differentiation",
"dna",
"transcription",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"organism",
"development",
"morphogenesis",
"pattern",
"formation",
"enzyme",
"chemistry",
"drosophila",
"research",
"and",
"analysis",
"methods",
"proteins",
"gene",
"expression",
"organogenesis",
"insects",
"arthropoda",
"biochemistry",
"genetics",
"cofactors",
"(biochemistry)",
"biology",
"and",
"life",
"sciences",
"organisms",
"evolutionary",
"developmental",
"biology"
] | 2014 | Drosophila melanogaster Hox Transcription Factors Access the RNA Polymerase II Machinery through Direct Homeodomain Binding to a Conserved Motif of Mediator Subunit Med19 |
Bacterial and human voltage-gated sodium channels ( Navs ) exhibit similar cation selectivity , despite their distinct EEEE and DEKA selectivity filter signature sequences . Recent high-resolution structures for bacterial Navs have allowed us to learn about ion conduction mechanisms in these simpler homo-tetrameric channels , but our understanding of the function of their mammalian counterparts remains limited . To probe these conduction mechanisms , a model of the human Nav1 . 2 channel has been constructed by grafting residues of its selectivity filter and external vestibular region onto the bacterial NavRh channel with atomic-resolution structure . Multi-μs fully atomistic simulations capture long time-scale ion and protein movements associated with the permeation of Na+ and K+ ions , and their differences . We observe a Na+ ion knock-on conduction mechanism facilitated by low energy multi-carboxylate/multi-Na+ complexes , akin to the bacterial channels . These complexes involve both the DEKA and vestibular EEDD rings , acting to draw multiple Na+ into the selectivity filter and promote permeation . When the DEKA ring lysine is protonated , we observe that its ammonium group is actively participating in Na+ permeation , presuming the role of another ion . It participates in the formation of a stable complex involving carboxylates that collectively bind both Na+ and the Lys ammonium group in a high-field strength site , permitting pass-by translocation of Na+ . In contrast , multiple K+ ion complexes with the DEKA and EEDD rings are disfavored by up to 8 . 3 kcal/mol , with the K+-lysine-carboxylate complex non-existent . As a result , lysine acts as an electrostatic plug that partially blocks the flow of K+ ions , which must instead wait for isomerization of lysine downward to clear the path for K+ passage . These distinct mechanisms give us insight into the nature of ion conduction and selectivity in human Nav channels , while uncovering high field strength carboxylate binding complexes that define the more general phenomenon of Na+-selective ion transport in nature .
Voltage-gated sodium ( Nav ) channels are widely distributed in the central and peripheral nervous systems where they participate in essential functions , including heartbeat , muscle contraction and brain activity [1 , 2] . Dysfunctional Navs are associated with several physiological disorders , including epileptic seizures and chronic pain [3 , 4] , and they are therefore a major target for new drugs [5] . However , our understanding of the fundamental mechanisms governing these channels remains incomplete , largely due to the lack of high-resolution structural data on mammalian Navs . There have , however , been recent breakthroughs in the structural determination for several Nav channels that enable investigations into molecular mechanisms . The first high-resolution structure of a Nav channel was the X-ray structure of the bacterial channel NavAb [6] , which was followed by numerous other high resolution structures , including for NavRh [7] , NavAep1 [8] and NavMs [9 , 10] . Although no mammalian Nav structure is yet available , two eukaryotic Nav structures have recently been solved using cryo EM; NavPaS from cockroach [11] and a newly resolved EeNav1 . 4 from electric eel [12] , released after the simulations in this study were completed . NavPaS has a 36–46% sequence identity to human Nav channels , but is 300–500 residues shorter . NavPaS has several residues in and around the SF unresolved , has reduced charge in the vestibule of the SF compared to human Navs and a resolution of 3 . 8 Å [11] . EeNav1 . 4 has a 65% sequence identity to the human Nav1 . 4 and a resolution of 4 Å [12] . The lower resolution of these structures is significant given the small size of the Nav SF ( diameter of ~3 . 6 Å at its narrowest point [12] ) , and thus while these new structures help guide and validate studies , there remains much to be learned from the existing high-resolution bacterial structures . Bacterial and human Nav channels share several features including Na+-selective conduction , voltage-dependent activation , pore-based inactivation and drug modulation [13–16] . While there is only a 25–30% sequence identity between bacterial and human Navs [10] , there exists evidence for overall shared structure [13] . Human Navs consist of four domains , DI-DIV , linked together to form one long polypeptide chain , whereas the simpler bacterial channels are made up of four identical subunits [13] ( Fig 1A ) . Each of these domains/subunits consists of 6 helical trans-membrane spanning segments , S1-S6 , where S1-S4 make up a voltage sensor domain ( VSD ) and S5-S6 the pore domain ( PD ) . Between these two latter segments is a P-loop that includes a narrow ion selectivity filter ( SF ) [13] ( Fig 1B and 1C ) , which establishes an ion preference “fingerprint” ( permeability Li+~Na+ > K+~Cs+~Rb+ ) that is the same for both bacterial and human Navs [1 , 17 , 18] . Both eukaryotic and bacterial Navs are Na+ over K+ selective with eukaryotic Navs selecting for Na+ with PNa+/PK+ ~10–30 [1 , 19] and bacterial Navs with PNa+/PK+ ~5–170 [17 , 18 , 20] . Due to the structural and functional similarities , the bacterial Nav channels offer an excellent template , or scaffold , to support investigation into the core functional activities of mammalian Navs . Despite overall conservation of structure and function , the channels present distinct SF sequences ( Fig 1A ) , with bacterial Nav channels making use of a ring of four Glu side chains ( EEEE ) , to create a high field strength binding site ( SHFS ) thought to favor Na+ [13] , whereas all mammalian Nav utilize a ring of Glu , Asp , Lys and Ala ( DEKA; Fig 1A ) [13 , 19 , 21] , containing not only acidic residues , but also a basic lysine side chain that is known to be crucial to selectivity in mammalian channels [19 , 21] . So how do these different channels achieve similar ion selectivity , and does the existence of disparate sequences mean that bacterial and mammalian channels impose their ion preference via different molecular strategies [18] ? We seek to reveal and compare the fundamental rules of selective ion permeation for the whole family of sodium channels , by exploring ion conduction mechanisms in a human Nav channel to compare to previous simulations in bacterial channels . There are 9 different Navs comprising Nav1 . 1 to Nav1 . 9 [22] , with highly conserved sequences throughout these different subtypes [23] . Nav1 . 2 is abundant in the human nervous system and has been investigated extensively using methods of mutagenesis and electrophysiology ( e . g . [21 , 24 , 25] ) . Here we take advantage of the conserved structural and functional features between mammalian and bacterial channels to generate a model of Nav1 . 2 , which incorporates the key SF and vestibular regions , into a high-resolution bacterial structure . Such grafting has previously been successful experimentally for imparting Ca2+ selectivity on a bacterial Nav channel [26] . The bacterial NavRh structure was chosen as the scaffold due to its higher sequence similarity in and around the SF ( Fig 1A; only shown for NavAb , NavRh and Nav1 . 2 , however , it can be seen that the same is true also for NavMs , NavAe and NavCt [16] ) . In particular , other bacterial Navs have one residue extra just above the SF ( Gly182; NavAb numbering ) compared to Nav1 . 2 ( Fig 1A ) . Furthermore , those channels have an arginine ( Arg185; NavAb numbering ) close to the SF that was found to interfere with the side chains of the Nav1 . 2 SF in our separate models using NavAb as a scaffold ( not shown ) . We also note that proposed open structures of a bacterial Nav have been presented [9 , 16 , 27] , but our simulations ( not reported here ) have suggested that the proposed open NavMs structure does not stay open without strong constraints , as discovered previously [28] . Furthermore , previous studies of ion selectivity in bacterial Navs using a closed lower gate have shown good sampling of ion translocation throughout the SF [18 , 29–31] , and are thus capable of shedding light on selectivity mechanisms by revealing the underlying equilibrium free energy surfaces governing multi-ion movements . There is minimal thermodynamic perturbation for ions in the SF and cavity due to the closed lower gate , with the free energy of an ion inside open and closed NavAb pores having been shown to be similar [27] . Previous simulations of bacterial channels have taught us much about the potential behavior of the human sodium channel . The bacterial SF contains a highly conserved ring of four Glu side chains forming a high field strength site ( SHFS ) [6] . Early simulation studies revealed ion binding sites and indications of a multi-ion conduction mechanism [28 , 31–36] . These studies suggested favoring of a partly-hydrated Na+ due to the SF geometry in the NavAb crystal structure and the strength of interaction of ions with glutamate side chains [28 , 31 , 36] . They also demonstrated the ability of the SF to bind 2 ions concurrently within the SHFS site , with higher affinity for Na+ potentially creating a reduced permeation barrier [36] . Simulations have shown that the SF is flexible and wide enough to house multiple Na+ ions [37] , and is occupied by an average of two to three ions [29 , 30 , 38]; although simulations under significant membrane potentials have indicated reduced occupancy [28] . However , microsecond-length MD simulations have shown that the symmetric arrangement of Glu side chains in the crystal structure is broken on long time scales , affecting ion occupancy and having significant implications for the permeation mechanism [29 , 30 , 36 , 39] . These studies demonstrated that there is coupling between ion translocation and SF conformation , with isomerization of glutamate side chains catalyzing Na+ conduction . These long unbiased simulations have described a stable 2-ion state , where the ions are trapped [30] . While some studies have focused only on this 2-ion state [31 , 36] , it has been shown that it is when a third ion enters the SF that efficient knock-on of the bottom ion into the central cavity occurs [18 , 29 , 30] . The top ion can either knock-on or pass-by the middle ion in SHFS , with both permeation pathways experiencing similar energy barriers ( S2 Fig; [30] ) . Regardless , binding of Na+ to the SHFS Glu side chains is central to understanding Na+ over K+ selectivity [18 , 30 , 31 , 36] . Conduction is reliant on the flexibility of these side chains [29 , 30 , 36] , where Na+ ions form favorable 2 ion-2 carboxylate clusters ( see S2 Fig , states C2 and C3 ) that are not stable for K+ [30] . Human Nav channels do not possess a symmetric E ring , but contain a set of carboxylates that might mediate similar complex formation . In particular , the DEKA signature sequence ( Fig 1 , right inset ) contains two carboxylates that may be sufficient . However , human channels also possess a well-conserved [24] charged ring in the outer vestibule , immediately adjacent to the SF , consisting of Glu , Glu , Asp and Asp residues ( EIEIIDIIIDIV; with the exception of Nav1 . 7 that contains EIEIIIIIIDIV ) [40] . The residues in these two rings have been shown to be asymmetrical in position [41] and highly flexible [42] , and thus may facilitate multi-ion conduction , akin to the bacterial channels . Substitution experiments have shown that the DEKA-ring is crucial for the selection of Na+ ions in mammalian Nav channels [19 , 21 , 25] , with at least one of the D and E necessary to preserve selectivity [19] . The most important carboxylate in the DEKA ring is the E , which together with the K ( residue 1422 in human Nav1 . 2 ) , maintains wildtype ( WT ) selectivity , despite mutation of the D [19] . However , if the positions of these two residues ( E and K ) are swapped , into DKEA , selectivity is reduced nearly 4-fold [25] , showing the importance of their precise locations , potentially due to interactions with surrounding amino acids , or implying a specific coordinating complex for Na+ during conduction . Mutagenesis has also shown that both the charge and length of the K1422 are necessary for maintaining Na+ selectivity; substitution of Lys with neutral side chains causing a complete loss of Na+ selectivity , and substitution with negatively charged side chains even reversing selectivity [19] . Previous computational studies examining the role of Lys in conduction have also been based on models using bacterial structures , and concluded on a passive blocking role for Lys . Xia et al . replaced the ring of four Ser in the bacterial channel NavRh with DEKA and performed MD simulations to find that only when the Lys was constrained to EIV was the SF permeable to Na+ and K+ ions [43 , 44] . Mahdavi et al . threaded the amino acids from rat Nav1 . 4 onto the backbone of a pore only model of NavAb , also suggesting that ion translocation requires displacement of Lys out of the permeation pathway , by hydrogen bonding to the neighboring DIV Ser [45] . We will show in this study , using extended length simulations , that Lys in fact plays an active role in conduction ( not just based on pore occlusion ) that is quite different for Na+ and K+ ions . The vestibular EEDD-ring has also been shown to be important for ion conduction , possibly because it increases the electrostatic attraction of extracellular cations [24 , 25 , 46 , 47] . However , all four residues are not equally important , with cysteine mutations showing the greatest decrease in conduction occurs when domain II Glu is mutated ( effect on conduction: EII>EI> DIV~DIII ) [25 , 41] . Furthermore , mutagenesis experiments have shown that replacement of EEDD residues can be detrimental to Na+ selectivity . In particular PNa/PK has been shown to decrease when DIV is replaced with Cys [40 , 41 , 48] , while other experiments have shown little or no effect [25 , 49] . Functionally , therefore , the exact roles of the outer ring carboxylates in selectivity are not well defined . Previous models based on bacterial structures have implied a greater thermodynamic preference for Na+ in the lower SF , and an additional outer binding site at the EEDD-ring [43 , 45] . However , we will show that on long timescales , SF rearrangements allow for stable complexes with carboxylates from both the SHFS and outer ring that are important to Na+ permeation . Experiments have suggested that Navs possess multiple binding sites inside the SF [50–53] . However , the evidence for their ability to bind multiple ions at the same time is inconclusive [50–55] . Investigations of flux coupling suggest a flux ratio exponent close to unity for mammalian Nav channels , which is generally thought to imply a 1-ion conduction mechanism [51 , 54 , 55] . Furthermore , anomalous mole fraction effects , considered evidence of ion cooperation during conduction , have only been observed in bacterial [18] and not mammalian Navs [53] . Such effects are normally expected in single-file channels that impose ion correlation [56] , but may also arise from preferential and localized ion binding [57–59] . Single-file conduction is unlikely in Nav channels given their size [60] and flexibility [42] . However , absence of an anomalous mole fraction effect does not rule out ion cooperativity . The fact that anomalous mole fraction effects are seen in bacterial Nav channels in the absence of bona fide single-file permeation , as suggested by several simulation studies , as well as the decoupling of ion and water fluxes [28] , demonstrates limitations in the ability of this measure to identify a multi-ion mechanism . Likewise , there exists contradictory evidence from single channel conductance measurements for mammalian Nav channels , which have shown a saturating dependence on concentration in some experiments , indicative of a single ion mechanism [50–52] , whereas others , where a wider range of concentrations have been examined , witness a complex relationship that may be explained by a multi-ion conduction mechanism [53] . Thus the experimental evidence is inconclusive when it comes to the ion occupancy and potential multi-ion conduction mechanism in mammalian Navs . Molecular dynamics ( MD ) simulations can therefore help us elucidate the mechanisms involved in ion selectivity in Navs . Bacterial and human Nav channels therefore both possess several carboxylate side chains in ( or adjacent to ) the SF , allowing for multiple high field strength sites that may bind one or more ions , possibly leading to increased Na+ selectivity [61] . In a multi-ion/multi-ligand complex there will be competition between favorable ion-ligand and unfavorable ligand-ligand and ion-ion energies [62] , needing long time scale simulations to capture the ion and protein configurations [29 , 30] . The mechanism is further complicated by the presence of the Lys , which may increase the sampling challenges due to side chain isomerizations . We thus turn to long MD simulations to capture those ion and slowly interconverting protein movements using the DE Shaw Anton supercomputer . We have performed multi-μs simulations where Na+ and K+ are permitted to freely diffuse into and out of the SF of the NavRh/Nav1 . 2 model , observing the involvement of DEKA and outer EEDD rings , as well as participation of the signature Lys in multi-ion binding complexes , revealing distinct permeation mechanisms for Na+ and K+ ions .
We have created a model of human Nav1 . 2 that includes the most important residues for ion permeation and selectivity , using a structurally well-defined bacterial channel as a scaffold . Although other parts of the channel may be important for attracting ions [11] , the main functional region involves the SF and outer vestibular regions [24] . The sequence alignment in Fig 1 was used to create a model of human Nav1 . 2 based on the bacterial NavRh structure ( PDB:4DXW [7] ) , due to their higher sequence homology in and around the SF region , as discussed above . The mutated region ( framed in cyan in Fig 1A ) , spanning residues 174–184 ( NavRh numbering used herein ) , was chosen to introduce the SF and vestibular sequence with minimal perturbation . This includes the human DEKA and EEDD rings ( residues 180 and 183/184 ) in the upper SF . Furthermore , residues 174–177 are located behind the SF and may interact with the side chains of the SF . In particular , within that range is NavRh Q175 , which is a positively charged Arg in Nav1 . 2 DI and DII , and may influence both structure and ion conduction . To minimize perturbation , side chains of the chosen amino acids were removed from NavRh and the corresponding Nav1 . 2 side chains rebuilt manually using the CHARMM program [63] . Herein the residues from the DEKA ring will be referred to by their 1-letter abbreviations . To avoid confusion with the symbol for Lys ( K ) , potassium will always be referred to in ionic form , K+ . The Asp and Glu residues from the vestibular EEDD ring will be referred to according to their domain numbers ( EI , EII , DIII and DIV ) . Between the DEKA ring and the EEDD ring the Nav1 . 2 sequence is one residue shorter in DII than in the bacterial channels . This residue ( number 181 ) was removed , and neighboring residues ( 180–182 ) joined with constrained MD simulation ( see below ) . We note that the original NavRh SF structure was closed at the lower gate as well as having a slightly collapsed SF . The collapsed SF has been proposed to be associated with the orientation of the Ser181 side chains , blocking the SF [64] , however , after removing the Ser side chains as well as patching Nav1 . 2 residues and equilibration ( described below ) , the filter forms an open SF conformation that allows us to study ion permeation ( Fig 1B and 1C ) . Other changes to the model , including the rebuilding of missing intracellular loops connecting S2 and S3 helices in the VSD using Rosetta [65] , and the maintenance of a key hydrogen bond between Thr178 and Trp182 that has been postulated to be important for keeping the shape of the SF in bacterial Nav [6 , 7] , are described in the Supplementary Information . It is difficult to predict or measure the pKa shift of Lys in a non-aqueous microenvironment with the coming and going of conducting ions , such as in the SF of a Nav channel [66] . Large pKa shifts have been recorded in channel environments , and protonation states depend on the exact environment and ion occupancy , which will fluctuate over time [67] . The dependence of Nav conduction on the protonation states of key residues in the SF has therefore been investigated in several studies of bacterial Navs [30 , 36 , 68] . Furthermore , quantum mechanical calculations using simplified models of the DEKA ring alone have demonstrated that a thermodynamic preference for Na+ may be achieved with either protonation state of the Lys [61] . However , the signature Lys has been shown to be crucial to Na+ selectivity [19] , intuitively suggesting that its charge may be important . To cover both possibilities , we have examined Na+ ion movements with both charged and neutral lysine , which will help us isolate the role of that charge in ion conduction . The proteins were embedded in lipid bilayers of dipalmitoyl-phosphatidylcholine ( DPPC ) , being the best characterized lipid for MD simulations [69] , with explicit TIP3P water molecules [70] and 150 mM of NaCl or KCl solution . Systems were built and pre-equilibrated with CHARMM [71] and further equilibrated using NAMD [72] prior to unbiased production simulations carried out on the purpose-built supercomputer Anton [73 , 74] for 4 μs for NavRh/Nav1 . 2 with charged Lys both for NaCl and KCl , and 2 μs for neutral Lys ( less time required due to better sampling of ion movements ) , totaling 10 μs . Simulations all used the CHARMM36 [69] lipid and CHARMM22 protein and ion parameters [72] with CMAP corrections [75] , chosen to provide direct comparison to our past simulations of the bacterial NavAb channel [30] . However , attention to ion-carboxylate parameters was given to ensure accurate interactions for Na+ and K+ inside the SF , with standard parameters for the ion-carboxylate interaction shown to lead to reasonable agreement with both ab initio MD free energies of binding and osmotic pressure data [76] . Descriptions of both ion-carboxylate and corrected ion-carbonyl interactions are discussed in the Supplementary Information . The free energy map , or potential of mean force ( PMF ) , for ion movement was calculated for each system using trajectory data with specific pore ion occupancies , from unbiased simulations as W ( {zi} ) = −kBT ln ρ ( {zi} ) + C , where ρ is the unbiased probability distribution as a function of reaction coordinate ( s ) {zi} , being the vertical positions of one or more ions ( z1 , z2 , or z3; defined with z1 being the lower-most ion ) , or their centroids ( e . g . z12 = ( z1 + z2 ) /2 ) , each relative to the center of mass of the protein , zref , and where C is a constant . See Supplementary Information for more details , including error calculations . The most important states involved in the permeation of Na+ and K+ ions across the SF were identified from the PMFs and cluster analysis . The frames were broken down according to ion occupancy as well as ion-ion and ion-carboxylate distances ( S1 Table ) . Different states were identified according to how many carboxylates were interacting with the ion/ions using a cutoff defined from the radial distribution functions in S3 Fig , with further details of the clustering method described in the Supplementary Information . These states include complexes involving one or more carboxylate groups , Na+ or K+ ions and/or the signature Lys ammonium group . To evaluate the relative binding affinities of these states , Free Energy Perturbation ( FEP ) [77] calculations were performed to obtain the relative binding free energies of Na+ and K+ to each particular complex . Details of these , and other calculations , including structural comparisons of available X-ray and Cryo-Em structures , and continuum electrostatic calculations to examine ion-binding propensities , are provided in the Supplementary Information .
After patching of Nav1 . 2 SF and vestibular residues into the bacterial channel , followed by equilibration , the overall structures of the PD and SF were stable and settled down to RMSD 2–3 Å ( S4 Fig ) after initial changes during the first μs of simulation , with the level dependent on the choice of Lys protonation state ( see below ) . Late changes are seen , especially for simulations in KCl , due to the onset of permeation following conformational changes within the SF , described below . The first 1 μs of the Nav1 . 2 model with charged Lys with NaCl and KCl and the first 0 . 25 μs of the shorter simulation with uncharged Lys have been discarded as equilibration ( S5 Fig ) . We observe that the Nav1 . 2 SF is relaxing into an asymmetrical shape where the DEKA and EEDD-ring span ~10–15 Å , allowing interactions with residues other than their immediate neighbors ( S5 Fig ) , explaining the increased RMSD . Previous experiments involving cysteine mutations have shown that the residues in the filter of an Nav channel are indeed asymmetrical in height and that the inner and outer rings span up to ~15 Å [41] . This asymmetry destabilizes the Thr/Cys178-Trp182 bond , important for keeping the shape of the SF in bacterial channels [6 , 7] , which are seen to break and form several times during the simulations . Flexibility has been shown to be a critical property of the bacterial Nav channel SF to facilitate ion conduction [29 , 30] . Furthermore , pairwise cysteine mutations have suggested significant flexibility ( up to 7 Å ) of the residues in the SF of a mammalian Nav [48] [42] . We observe high flexibility in the NavRh/Nav1 . 2 SF ( S6 Fig , with mean SF backbone fluctuations being approximately twice as large as those for NavAb ( RMSF~1 . 5 Å and RMSF~0 . 8 Å for NavRh/Nav1 . 2 and NavAb , respectively ) ( S6A and S6B Fig ) . These RMSF values ( S6A Fig ) are mostly due to asymmetric subunit movements , with the RMSF values halved when computed based on subunit-by-subunit orienting ( S6B Fig ) . The recent eukaryotic NavPaS [11] and EeNav1 . 4 [12] cryo EM structures offer related structural data to validate our model , although at limited resolution where several SF side chains were not resolved , preventing immediate conclusions about specific interactions . Structural alignment between the NavRh/Nav1 . 2 model and the NavPaS or EeNav1 . 4 structures ( S1 Fig ) shows structural similarities between the proteins , with key residues at the same positions . Due to the highly dynamic nature of the SF and its side chains , structural comparisons are limited and will depend on the choice of structure from the model simulation . Overall , however , structures are consistent , with the RMSD of the SF and vestibular region ( residue 178 to 184 ) backbone being 2 . 0 Å and 2 . 3 Å when aligning to all four subunits ( S1A Fig ) , and 1 . 8 Å and 2 . 1 Å when aligning subunit by subunit ( S1A Fig ) , for NavPaS and EeNav1 . 4 , respectively . The residues of the DEKA and EEQD/EEDD rings are asymmetrical in position in all three structures , with these residues spanning 10–15 Å in the model , ~10 Å in NavPaS and ~14 Å in EeNav1 . 4 ( S1 Fig ) . The spread of these side chains in both our model and the NavPaS and EeNav1 . 4 structures , combined with the high flexibility and lack of experimental resolution in this region , indicates that SF dynamics may play an important role in conduction . The DEKA Lys has been shown to be crucial to Na+ selectivity in mammalian Nav channels [19] . However , the protonation states of amino acids in a dynamic microenvironment , like that seen here within the SF ) , are hard to pinpoint experimentally [66] and may have large fluctuations [67] . We have therefore investigated ion permeation in Nav1 . 2 not only with a protonated ( Lys+ ) but first also with a deprotonated Lys ( Lys0 ) . Importantly , this allows us to isolate the role that the positive charge may play in the SF of Nav1 . 2 for Na+ permeation , and offers a comparison to existing bacterial Nav simulations . Permeation of Na+ in the SF of the Nav1 . 2 model when its DEKA Lys is neutral involves both 2-ion and 3-ion conduction mechanisms with an average number of 2 . 1±0 . 2 Na+ ions in the SF , similar to ( albeit slightly less than , though within the errors ) that seen in the bacterial channel NavAb ( 2 . 3±0 . 5 ions; [30] ) . The ions are partly hydrated as they permeate the SF ( S7 Fig ) . In the time series for ion movements shown in Fig 2 , several ions ( colored lines ) are observed to enter and exit the central cavity ( below −5 Å ) , often coexisting within the SF ( S1 Movie ) . We see 61 complete permeation events through the SF ( inward or outward moving involving 16 distinct ions ) and the ions appear to move independently of the neutral Lys side chain ( black line ) . Snapshots in Fig 2 show representative configurations of a 2-ion and 3-ion state , where we see multi Na+-ion/multi-carboxylate complexes similar to those in for the bacterial Nav ( [30]; see also Fig 3E and 3F , to be discussed below ) . However , in the Nav1 . 2 channel it most commonly not only involves carboxylates only from the inner DEKA-ring , but also from the outer EEDD-ring , whose carboxylates reach down to help coordinate the ions , as opposed to only using two carboxylates from the inner ring , as was the case of the bacterial NavAb channel [30] . This cooperation between the DEKA and EEDD-rings creates a longer SF binding region with a broad span of negative electrostatic potential ( Fig 3A and 3B ) , leading to less distinct binding sites . While NavAb has all of its charged side chains in one symmetrical ring , Nav1 . 2 has its charges spread out in two asymmetrical and highly flexible rings creating a broad SHFS ( ion density map shown in Fig 3C ) , compared to NavAb ( Fig 3D ) . Commonly a binding site is created with two Na+ and one or two carboxylates from the DEKA and EEDD rings ( S1 Table; column 2 and S8 Fig ) . The specific residues involved vary and we see several distinct carboxylate-ion complexes forming during the simulation ( S1 Table; column 2 and S8 Fig ) . Double ion occupancy dominates in the SF , but 3-ion occupancy is also common ( S8A Fig ) . When the SF is occupied by a single ion it is most commonly unbound , however , it is also likely to be bound to one or more carboxylate side chains ( S8B Fig ) . When there are two ions in the filter they most likely form single ion complexes with multiple carboxylates ( S8C Fig ) . They are also often singly bound or bound together in a tight multi-ion/multi-carboxylate complex ( defined by the radial distribution function in S3 Fig; with ions within 4 . 7 Å of each other and 2 or more carboxylate groups ( based on central C atom ) from the DEKA and EEDD-rings within 3 . 8 Å of the ions ) . When there are three ions in the SF , two of these are most likely to be bound together in a tight multi-carboxylate complex ( S8D Fig ) . We have previously shown that multi-ion/multi-carboxylate complexes play a role in Na+ selectivity in acid sensing ion channels [76] . When these complexes are present , the ions are mostly bound by E and/or D from the DEKA ring , together with the outer ring DIII ( S1 Table; column 2 ) . This cooperativity of inner and outer charge rings is common and while all carboxylate side chains from either of the two rings ring are not necessary , at least one is needed for Na+ binding in the SHFS . In particular , the DEKA ring is involved ~80% of the time , the EEDD ring ~80% of the time , and both rings cooperatively involved ~60% of the time . This need for only one of the carboxylates in the DEKA ring to maintain cation permeability has previously been implicated by mutagenesis experiments [19] . Despite the longer SF binding region in Nav1 . 2 with neutral Lys , we see a conduction mechanism resembling that of the bacterial Nav , relying on multi ion-multi carboxylate clusters for an efficient knock-on conduction . When there are 2 ions within the SF , at z1 ( lower ) and z2 ( upper ) , the free energy map ( as function of z1 and z2 ) in Fig 4A exhibits three states; A2 , B2 and C2 . In state A2 , the upper ion is in the vestibular region of the SF ( z2~14Å ) and the lower ion is collectively bound by the inner ( purple side chains ) and outer rings ( green side chains ) ( z1~7Å ) . In state B2 the top ion has joined the bottom ion and they are both bound by the inner and outer rings ( z1~z2~7Å ) forming cooperative tight multi-ion/multi-carboxylate clusters , commonly made up by D , E and DIII , and reminiscent of those seen in the bacterial Nav ( compare Fig 3E and 3F ) . In state C2 the top ion has either pushed the bottom ion into the cavity by Coulomb repulsion ( dashed line ) or it has passed by the bottom ion and entered the cavity itself ( dotted line crossing the diagonal , y = x , represented by black dashed line ) , in both cases translocating from z2~7 Å to z2~-10 Å , while the other ion remains collectively bound to the inner and outer rings . These distinct permeation pathways are equally likely , with the largest barrier encountered being 1 . 9±0 . 8 kcal/mol . Fig 4B–4D show free energy projections for the 3-ion occupancy state , with ion positions specified as z1 ( lower ion ) , z2 ( middle ion ) , z3 ( upper ion ) or z23 ( COM of the two upper ions ) . Here state A3 has one ion in the middle of the two rings ( z1~7 Å ) and one ion in the vestibule ( z2~14 Å ) , and an additional ion entering from the bulk above ( z3~15 Å ) . In Fig 4B and 4C we see how the upper ion pushes the middle ion downward to bind together with the lower ion at the inner ring ( z1~z2~6 Å ) , represented by state B3 , before eventually entering this site itself and pushing the bottom ion into the cavity , state C3 , completing the conduction event . The broad free energy surface in panel c is due to the range of binding sites offered by the inner and outer carboxylate rings , as well as the fact that the centroid of 2 ions may span a wide range . Fig 4D shows another projection involving only the top two ions ( z2 and z3 ) . The dashed line in Fig 4D shows how the top ion moves from state B3 to state C3 by entering the SHFS and knocking the bottom ion downward . The dotted line shows an alternative path where the top ion instead passes by the middle ion before knocking the bottom ion into the cavity . Regardless of the permeation pathway , the largest barrier experienced during permeation through the SF in the 3-ion state is 1 . 4±0 . 6 kcal/mol , similar to that experienced for Na+ in the bacterial channel ( S2 Fig ) . These results demonstrate that low barrier conduction may occur via knock-on or pass-by mechanisms in either 2- or 3-ion occupancy states for Nav1 . 2 , but where 3-ion conduction is energetically more favorable , with reduced activation barrier . However , considering a 2-ion occupancy is more common ( S8 Fig ) , the 2-ion mechanism is likely also contributing significantly to the overall ion flux . We therefore have observed that the mammalian SF conducts Na+ ions well with a neutral/deprotonated Lys in the DEKA ring , providing a useful comparison to the bacterial channel , with similar multi-ion/multi-carboxylate complexes forming during permeation ( although involving both SF DEKA and vestibular EEDD carboxylates ) . However , because Lys is so important to permeation and selectivity experimentally [19 , 21 , 25] , but when neutral apparently plays only a passive role that would not implicate it , this suggests that we must turn our attention to the protonated Lys case . When Lys is charged , there is an average of 1 . 5±0 . 4 Na+ ions in the SF and permeation involves either 1 or 2-ion conduction ( S2 Movie ) . This reduction in ion occupancy ( compared to the neutral Lys case above and the bacterial channel ) is presumed to be due to the reduced negative charge inside the SF . If the ammonium group of the Lys was considered as another ion , we would have , in total , a similar ion occupancy of 2–3 ions in the SF . ( n . b . Lys+ may be considered another ion , as Nav channels conduct ammonium ions nearly as well as Na+ , with relative permeability PNH4+/PNa+ = 0 . 16 [60] ) . The time series in Fig 5 reveals that several ions ( colored lines ) enter and leave the SF , but it is not until the second half of the 4 μs simulation that ions begin to cross the charged Lys and enter the cavity beneath , after which we observe 28 complete permeation events ( defined as any ion crossing downward or upward past the Lys , and involving 6 distinct ions that experience repeat crossings ) . Snapshots in Fig 5 show representative configurations of 1- and 2-ion carboxylate complexes . As with the neutral Lys case , we see multi-ion/multi-carboxylate complex formation where the residues from the EEDD ring ( green ) bend down into the SF region , leading to the binding of ions together with the carboxylates from the DEKA ring ( purple; see also S1 Table; column 3 ) . The DEKA ring is involved in binding ~50% of the time , the EEDD-ring ~90% of the time , and both rings cooperatively involved ~40% of the time . These complexes are most commonly ( 69±11% ) involving D from DEKA and EII from EEDD . The ammonium group of the Lys from DEKA is also involved , binding 61±10% of the time to the E or D from the DEKA-ring . The salt bridge between Lys and one or both of these residues has previously been suggested by mutagenesis experiments as being important for Na+ selectivity [19] , and may be responsible for the importance of the precise sequence position of the Lys in the DEKA ring [25] . We also observe that when ions enter the SF , the ammonium group of Lys ( black line in Fig 5 ) is displaced downward; being knocked-on , like any other ion . Thus , the charged Lys is intimately involved in the multi-ion mechanism . Importantly , it also participates in a Lys+Na+/carboxylate complex , similar to the 2-ion/multi-carboxylate clusters , commonly coordinated by D and EII and sometimes additionally by E ( S1 Table; column 5 ) . The ability to form these complexes is likely dependent on the location of the Glu and Lys in the DEKA ring , and swapping their positions might affect this complex formation thus decreasing selectivity , as indicated by mutagenesis experiments [25] . We will see below how this tight cluster plays an important role in the conduction mechanism . Free energy projections reveal the relationship between the ions and the position of the ammonium group of Lys , zLys , demonstrating conduction with 1-ion and 2-ion occupancies in the SF . When there is a lone ion in the SF , at z1 , the free energy map ( as function of z1 and zLys ) in Fig 6 exhibits four states; A1 , B1 , C1 and D1 . In state A1 , the ion is in the outer vestibular region of the SF ( z1~14Å ) and the Lys is bound to one of the inner ring carboxylates ( zLys~1 Å ) . In state B1 , the ion has entered deeper into the SF and is bound to the inner and outer rings ( z1~6 Å ) . The dotted line shows an alternative state , B1’ , in which the ion entering the filter replaces the ammonium group of the Lys , which is displaced downward ( from z1~6 Å & zLys~1 Å to z1~4 Å & zLys~-2 Å ) . It is interesting to note that there is no conduction occurring when this happens . The ion is bound tightly to the DEKA ring and the ammonium group of the Lys is too repulsive for it to pass by in the lower part of the filter , where the electrostatic potential is not as negative . However , the dashed line shows how the ion can cross the ammonium group of the Lys in a region of higher field strength created by multiple carboxylates ( z~2 Å ) , leading to state C1 . In state C1 the ammonium group of the Lys remains bound to the inner ring ( zLys~2 Å ) and the ion is bound to the backbone carbonyl of the two residues underneath the DEKA ring ( z1~-2 Å ) . In state D1 the ion has left the SF and entered the cavity ( z1~-8 Å ) , completing the permeation event . The dashed arrow shows the minimum free energy path for conduction , where the largest barrier encountered is 1 . 7±0 . 3 kcal/mol . The 2-ion conduction mechanism is summarized in Fig 7 . The free energy projection in Fig 7A shows the position of the bottom ion ( z1 ) as a function of the Lys ( zLys ) , Fig 7B shows the COM of the two ions ( z12 ) as a function of the Lys , while Fig 7C shows bottom versus top ions . In this 2-ion case , we see five distinct states; state A2 is like the corresponding 1-ion state A1 , but with an additional ion entering from the bulk above ( z12~14 Å ) . In Fig 7B we can see how the two ions enter deeper into the SF and bind collectively to the inner and outer rings , just above the Lys in state B2 ( z1~z2~7 Å & zLys~2 Å ) . There is an additional state B2’ , related to B2 , where the ammonium group of the Lys is displaced downward ( from z1~6 Å & zLys~2 Å to z1~4 Å & zLys~-2 Å; dotted line ) . However , just as for the 1-ion state , this downward Lys displacement is not sufficient for conduction , instead requiring that the ammonium group of Lys rise up and join one of the ions in a new intermediate stable state C2 as shown in Fig 7A ( z1~zLys~2 Å ) . This state only appears when there are 2 Na+ ions in the filter and seems to be vital for efficient permeation , because it allows the ion to pass the Lys with reduced energetic barrier ( dashed line ) . The ion then binds below the Lys to the backbone carbonyls of the lower SF represented by state D2 ( z1~-2 Å & zLys~2 Å ) . In state E2 the top ion pushes the bottom ion into the cavity and binds to the inner ring ( z1~-6 Å , z2~6 Å & zLys~2 Å ) . Fig 7C shows only the two ions , at z1 ( lower ) and z2 ( upper ) , and the corresponding states . The top ion can either enter and knock on or pass by the bottom ion . The dashed arrows show the lowest free energy path for conduction . The greatest barrier to overcome in the 2-ion state is 1 . 6±0 . 7 kcal/mol . Importantly , while conduction was seen to be possible with a singly-occupied SF ( seen in Fig 6; dashed line ) , the entrance of a second Na+ ion enables a stable state ( Fig 7A; state C2 ) where the ammonium group of the Lys and the bottom ion are collectively bound by the carboxylates , this allows the ion to pass by the Lys and in to the cavity with a lower barrier . As we shall see below , this efficient conduction mechanism does not exist for K+ ions . The Nav1 . 2 SF with charged Lys , in the presence of KCl solution , has an average occupancy of 1 . 3±0 . 1 K+ ions , being slightly lower than for Na+ ions , but comparable within errors . Fig 8 shows several ions ( colored lines ) entering and exiting the SF but not interacting extensively with the Lys ( black line ) . We see 38 complete K+ ion permeation events ( involving 13 distinct ions with repeat crossings , predominantly in the latter half of the 4 μs simulation after Lys rotamer change; see below and S3 Movie ) , being similar but somewhat less than the Na+ case above . Most of the time there exists a single K+ ion in the SF ( S9 Fig ) . When there are two K+ ions in the SF they are generally further from each other , with a mean ion-ion distance of 15 . 1±0 . 8 Å , compared to 11 . 4±1 . 2 Å for Na+ . Snapshots in Fig 8 show representative configurations . The K+ ions most commonly bind to single carboxylates or stay unbound ( S9 Fig ) . The lower occupancy and the larger ion-ion distances mean that we do not see the same tight multi-ion/multi-carboxylate complexes as frequently as we did for Na+ ( S9 Fig ) , and when they do occur they are almost always singly coordinated by outer ring carboxylates ( S1 Table; column 4 ) . Instead , a single K+ ion typically enters the SF and binds solely to the D from the DEKA ring ( S1 Table; column 6 ) . The DEKA ring is involved in binding ~40% of the time , the EEDD ring ~70% of the time , and both rings cooperatively involved only ~10% of the time . This reduced binding and absence of complex formation deep in the SF slows K+ permeation . We instead observe K+ ions held bound to the D of DEKA , electrostatically repelled by the Lys ammonium group , until the Lys side chain changes rotamer downward to allow ion movement , in stark contrast to the Na+ case above . We again see conduction with either 1 or 2 K+ ion occupancy . However , unlike for Na+ ions , we see similar permeation mechanisms when 1 or 2 K+ ions are in the SF . This , together with the large K+- K+ distance of 15 . 1±0 . 8 Å , compared to 11 . 4±1 . 2 Å for Na+ , suggest that the second ion does not participate in conduction but rather is loosely associated around the vestibule region . Fig 9A shows the 1-ion conduction mechanism , with the ion at z1 , and the ammonium group of Lys , at zLys . We observe four states in the 2D PMF in Fig 9; A1 , B1 , C1 and D1 . In state A1 the ion is in the vestibular region of the SF ( z1~14 Å ) and the Lys is bound to one of the inner ring carboxylates ( z1~0 Å ) . The ion then moves down and binds to the outer ring ( z1~8 Å ) represented by state B1 . In state C1 , the ion has entered deeper into the SF and is now bound only to the carboxylate of the D1 from the DEKA ring ( z1~4 Å ) . The Lys is still bound to the inner ring in both these two states ( zK~0 Å ) . The ions do not have considerable effect on the position of the ammonium group of the Lys . However , occasionally ( ~10% of the time ) the Lys changes rotamer downward ( zLys<-2 Å ) , allowing leakage of K+ ions and thus permeation , represented by dashed line , leading to state D1 . After the conduction event , the ammonium group of the Lys again bends upward to bind to the inner ring carboxylates ( zLys~2 Å ) . For this to happen the K+ ion has to pass by the ammonium group of the Lys in the lower part of the SF ( z1~zLys~-2 Å ) where the electrostatic potential is less negative , leading to a larger energy barrier of 2 . 8±0 . 3 kcal/mol , as seen in Fig 8A along the pathway represented with a dashed arrow . This increased barrier ( by 1 . 1±0 . 4 kcal/mol ) would suggest a relative permeability for the channel of the order of 10 , which is consistent with the experimental value for mammalian Nav channels ( e . g . PNa+/PK+ ~10 in Nav1 . 2 from rat [25] ) . This estimate , however , assumes the rate-limiting step for conduction does not involve Lys rotameric change , which was observed here on the multi-μs timeframe; apparently exceeding the sub-μs scale of permeation , but which may not have been reliably quantified based on 4μs simulation . The 2-ion conduction mechanism in Fig 9B looks very similar , with the large K+- K+ distance of 15 . 1±0 . 8 Å showing that it is the same as the 1-ion conduction mechanism , only with an additional ion in the vicinity . Fig 9B shows the 2D PMF with the bottom ion , at z1 , and the ammonium group of Lys , at zLys , being almost identical to the 1-ion case in Fig 9A; supporting the lack of a concerted 2-ion conduction mechanism . The lower ion enters the channel , binds collectively to the DEKA and EEDD rings in state B2 , before binding solely to the DEKA ring in state C2 . To enter the cavity , the ion has to wait for the Lys to change rotamer , it can then pass the Lys ammonium group in the lower SF , and then move further down into the cavity ( state D2 ) . In the meanwhile , the upper ion generally sits in the vicinity of the SF . In Fig 9C we see how the two ions can cross each other in the top part of the channel . The top ion can be anywhere between z2~7 Å and z2~15 Å during conduction , however , with a slight preference around z2~12 Å . The 2-ion energy barrier is similar to the 1-ion case at 2 . 7±0 . 3 kcal/mol . The Na+ and K+ barriers differ by 1 . 1 kcal/mol , however , the slowest coordinate in the K+ translocation pathway appears to involve the structural isomerization of the Lys side chain downward to permit conduction , leading to a permeation event that is more costly for K+ . Key to the low barrier for Na+ permeation is its ability to form tight multi-ion/multi-carboxylate clusters as well as a complex with Lys and 2 carboxylates , allowing pass-by conduction in the SHFS . We do not see the same tight K+-K+ clusters nor the simultaneous binding of K+ and Lys near the DEKA ring . Instead we see a lone K+ ion that is singly bound and forced to pass by the Lys lower in the SF ( S9 Fig ) , where the electrostatic field is less favorable ( Fig 3A ) , making the whole process far less likely . We now explore the thermodynamic stabilities of the complexes observed during permeation events to understand the causes of the different Na+ and K+ mechanisms that may underlie selectivity . Multi-ion complex formation with carboxylates is required for efficient permeation as it helps draw in and stabilize the ions in the SF . This acts as a precursor for the ion+Lys/multi-carboxylate complex that aids in the crossing of the SF . The analysis in S9 Fig suggests that such multi-ion/multi-carboxylate complexes appear more favorable for Na+ over K+ . When the SF is occupied by a single ion , Na+ is most commonly bound to one or more carboxylate side chains ( 45% and 26% ) , whereas K+ is more likely to be unbound ( 61% ) ( S9B Fig ) , and Na+ is several times more likely than K+ to form multi-carboxylate complexes ( 26% vs . 6% ) . Of the multi-ion occupancies observed , Na+ ions are most likely to be either singly bound or in loose multi-carboxylate complexes ( 42% and 40% ) , whereas K+ is most likely to be either singly bound or unbound ( 52% and 36% ) . Na+ is several times more likely to form tight multi-ion/multi-carboxylate clusters than K+ ( 12% vs . 3% ) ( S9C Fig ) . For Na+ ions , these clusters are predominantly made up by outer ring carboxylates together with either the E or D from the DEKA ring , most commonly EII together with D ( S1 Table; column 3 ) . For K+ such complexes are almost always made up only by the outer ring carboxylates ( S1 Table; column 4 ) . Thus , tight 2-ion/multi-carboxylate complexes , similar to the complex that is formed before the low free energy pass-by conduction events for Na+ ( Fig 7; state B2 ) , are more common for Na+ than K+ , and may be very important for selectivity ( S9 Fig ) . FEP calculations were used to investigate the relative stabilities of Na+ and K+ in such a complex . The most representative multi-ion/multi-carboxylate complex was identified to be bound collectively by D and EII ( Fig 10B and 10C; inset ) ( 69% ) ( S1 Table; column 3 ) . Results from FEP calculations show preferential binding by Na+ with 4 . 8±0 . 1 kcal/mol for single ion occupancy ( Fig 10A ) , showing a strong inherent preference by carboxylates to bind Na+ , in agreement with Eisenman high field strength theory on ion selectivity [78] . Quantum mechanical calculations using a model DEKA ring have demonstrated a similar preference for Na+ over K+ ( 4 . 8 kcal/mol [79] ) , revealing consistency between models . The same complex containing two ions shows a large preference of 8 . 3±0 . 1 kcal/mol for Na+ over K+ ions ( Fig 10B and 10C ) ; 4 . 2 kcal/mol from the transformation of the 1st ion ( Fig 10B ) and 4 . 1 kcal/mol from the transformation the 2nd ion ( Fig 10C ) . This shows that an additional ion creates extra stability for Na+ relative to K+ . When another ion is added to the complex , the cumulative ion-carboxylate attraction increases faster than the carboxylate-carboxylate and ion-ion repulsion , more so for Na+ than K+ , giving additional stability to Na+ complexes . These complexes create deeper binding in the SF that aids crossing the ammonium group of the Lys . We note that this large cumulative free energy difference between Na+ and K+ ions to form a 2-ion complex with multiple carboxylates is not the key energy controlling the different permeation mechanisms in Figs 7 and 9 , but represents the relative stability of a complex that is particular to Na+ , and largely unseen for the K+ ion because of its low stability , helping explain the distinct mechanisms . During the ion crossing of the charged Lys , the ion creates a stable transition state together with the Lys and multiple carboxylates ( Fig 10D; inset ) . This ion-Lys state is bound collectively by the D and E from the DEKA ring , as well as EII from the outer ring , and shows preferential binding of a single Na+ by 4 . 3±0 . 4 kcal/mol ( Fig 10D ) . This ability to create a joint ion-Lys complex to facilitate crossing of the Lys in the SHFS is the salient difference in Na+ and K+ permeation mechanisms . The K+ ion instead usually resides alone in the SF and crosses the ammonium group of the Lys further down , ( Fig 10E; inset ) , where the electrostatic fields arising from carboxylates in the upper SF is less attractive . If we examine the relative free energies of Na+ and K+ ions in this region of the SF near the Lys , where such crossing occurs for K+ , we calculate a similar but reduced value of 3 . 1±0 . 2 kcal/mol , still favoring Na+ ( Fig 10E ) . This value remains negative because the ion remains in contact with D and E side chains from the DEKA ring as it passes the Lys side chain . Theoretically , both K+ and Na+ should be able to translocate in this fashion when the Lys moves downward , but this was not observed for Na+ . We may understand this from the relative destabilization of Na+ in this region relative to the selective multi-carboxylate complexes that form above it , making downward movement of Na+ less likely . We therefore hypothesize that selective permeation in Nav1 . 2 arises from the distinct mechanisms ( Lys-complex pass-by Na+ permeation versus electrostatically Lys plugged K+ permeation ) , which originates from the difference in stabilities of multi-ion and ion-Lys complexes in the high field strength region of the SF . Multi-ion/multi-carboxylate complexes have also been seen to be crucial for Na+ permeation in the bacterial NavAb channel , where a preference for single Na+ ions of -2 . 2±0 . 6 kcal/mol has been estimated [30] . In the bacterial channel , we also suggest that optimal conduction requires the greater occupancy of the 3-ion state which is achieved when the crystallographic E177-S178 H-bond is lost on 2 opposite monomers , allowing the SHFS carboxylate groups to reach toward the center of the SF and coordinate 2 Na+ ions concurrently . This two-Na+/two-carboxylate complex offers additional stability , allowing those 2 ions to coexist in SHFS , and we calculate the relative stability of Na+ vs . K+ in SHFS to become -3 . 8±0 . 2 kcal/mol in NavAb . This observation suggests that Na+ ions adopt a specific 3-ion arrangement with 2 ions at the level of SHFS , that does not occur in simulations of K+ , which closely parallels our observations in the Nav1 . 2 channel . The ability to select for native ions while discriminating against other similar ions , is one of the key features of voltage gated ion channels . Ion conduction is a dynamic process that involves conformational isomerizations of the residues of the SF , as well as cooperation between multiple ions , requiring long timescale simulations to extract the underlying energetic landscape . In the absence of a high-resolution structure of a mammalian channel to study selectivity and conduction at the molecular level , we designed a model of the human Nav1 . 2 by patching the essential sequence of residues in and around the SF , including the key DEKA and EEDD rings , into the NavRh bacterial channel . This NavRh/Nav1 . 2 channel relaxes into an asymmetrical configuration , where the DEKA and EEDD rings are highly flexible and cooperate to ensure efficient conduction and selectivity for Na+ over K+ , in line with previous studies of bacterial Nav channels where similar flexibility has been shown to be critical for ion conduction [29 , 30] . The presence of a positively charged Lys in Nav channels has long been a source of intrigue; being essential for selectivity , but its specific role unexplained . Previous computational studies of eukaryotic Nav models point toward a passive , hindering role , where the Lys side chain needs to bind away from the middle of the pore lumen in order to allow conduction of Na+ ions . We reveal a more intimate role for this residue , depending on its protonation state . When Lys is deprotonated ( neutral ) , we observe a 2 to 3-ion occupancy in the SF where both 2- and 3-ion knock on conductions are common , facilitated by tight multi-ion/multi-carboxylate complexes that are made possible by the flexible side chains of the SF . When Lys is instead protonated ( charged ) , we see reduction to 1 to 2-ion occupancy , due to the extra charge introduced by the ammonium group of the Lys . In fact , we see the Lys actively participating in the conduction mechanism with a role similar to that of a permeating ion . In the presence of K+ ions , the conduction mechanism instead relies on single ion permeation , where multi-ion/multi-carboxylate complexes are less likely . These multi-Na+/multi-carboxylate complexes help attract Na+ in to the SF of Nav1 . 2 , where binding deeper in the SF stabilizes the ions . Interestingly , the carboxylates involved in these key multi-Na+/multi-carboxylate complexes generally include only one carboxylate from DEKA and one or two from the EEDD ring in the vestibular region , rather than both D and E from the DEKA ring . This observation is consistent with mutagenesis experiments showing the need for only one of the D and E from the DEKA ring , supporting the idea of involvement of the outer ring [19] . These Na+ favorable complexes are more commonly bound collectively by D and EII . Cysteine mutations of the residues in the EEDD ring show the greatest decrease in conduction when this domain II Glu ( EII in our model ) is mutated [25 , 41] , consistent with our observation that this residue from the outer ring is particularly important . In the case of charged Lys we see conduction that is possible with a singly occupied SF , however , the binding of a second Na+ ion reduces the conduction barrier by creating an intermediate state that enables efficient multi-ion conduction . The second Na+ ion helps to push the ion already in the SF and stabilize a state where the ammonium group of the Lys and the bottom ion are collectively bound by the carboxylates , allowing the ion to pass by the Lys into the cavity . The facilitating of inward conduction by the presence of an additional external Na+ ion may explain the robust inward rectification of current in mammalian Nav channels seen experimentally [1] . Even though more space for conduction can be created in the SF by the steric unplugging of the ammonium group of the Lys , no conduction occurs in that state . This behavior is consistent and comparable with previous results in NavAb , where pass by of ion by another would only occur at the level of SHFS . When the charged ammonium group of Lys is in the SHFS , where the electrostatic potential is most negative , it creates a smooth electrostatic environment leading into the cavity , whereas when it is in the down state it creates a zone of high electrostatic potential that cuts the cavity off from the SF . The Na+ ion can benefit from the smooth energetic surface by binding at SHFS in a multi-ion complex with Lys , whereas K+ cannot; instead becoming partially electrostatically blocked by the Lys in the lower SF . This illustrates the fundamental role of the Lys as an active participant in the selective conduction mechanism . K+ is far less likely to form these multi-ion/multi-carboxylate complexes; in particular , we see preferential binding of 8 . 3±0 . 1 kcal/mol for Na+-Na+ over K+-K+ in these complexes . Furthermore , K+ does not form the stable multi-carboxylate/Lys state . In fact , the ion+Lys/multi-carboxylate cluster favors Na+ over K+ by 4 . 3±0 . 4 kcal/mol . The consequence of this difference in affinity is a completely different conduction mechanism for K+ , having to pass the ammonium group of Lys in the lower part of the SF , where the electrostatic environment is not as favorable . Interestingly , the apparently independent conduction pathways for Na+ and K+ ions could potentially explain absence of anomalous mole fraction effects in mammalian Nav channels [53] . In the future , simulations of a model human Nav channel in an open conformation , based on constrained-open NavAb or NavMs structures [9 , 16 , 27] , could be used to explore the competition of Na+ and K+ ions under the driving force of a membrane potential . It has been proposed that mammalian Navs select for Na+ via a different mechanism to bacterial Navs , owing to their divergent SHFS site forming residues ( EEEE Vs DEKA ) [18 , 80] . While this appears to be the case based on our observations , we do see evidence for common features that may be central to Na+ selectivity . Both bacterial and human channels make use of an efficient multi-ion process enabled by flexible carboxylates binding two Na+ ions during ion knock on or pass by conduction events . The efficiency of this multi-ion mechanism is reliant on the formation of 2-Na+/2-carboxylate complexes , whose thermodynamic stabilities are increased for Na+ compared to K+ . Even though bacterial channels possess 4 carboxylates at the SHFS site , only two are needed for high affinity binding [30] . Thus , the bacterial channel uses two of its four EEEE Glus , whereas the human channel uses carboxylates both from the DEKA and EEDD rings , leading to a Na+ permeation process that closely resembles our observations in the model human channel . However , the Lys residue in the human DEKA signature participates in the multi-ion knock-on mechanism in ways that impose a unique ion discrimination process , perhaps explaining the higher level of ion selectivity in mammalian Navs [1] . We propose that this additional selectivity is achieved by establishing distinct permeation mechanisms for Na+ and K+ ions , where the Lys expedites Na+ translocation through high field strength complex formation , while partially blocking the K+ ion . | Ion channels can rapidly and selectively conduct an ionic species , essential for the firing of neurons , where sodium and potassium channels respond to changes in membrane potential to release stores of sodium and potassium ions in succession . The ability of a protein pore to discriminate between these two nearly identical ions has remained an intriguing problem for decades . In particular , the origins of sodium selectivity have been obscured by diverse protein chemistries that exhibit sodium-selective conduction in prokaryotes and eukaryotes . Here we use multi-microsecond atomistic simulations to observe and contrast the permeation mechanisms of sodium and potassium ions in model human and bacterial sodium channels . These channels exhibit shared features of conduction , centered on the involvement of charged protein groups that form complexes with the smaller sodium ion . In the human channel model , we observe a special role for the signature lysine that allows sodium to pass , but partially blocks potassium . As sodium channels are vital to heartbeat , sensation , muscle contraction and brain activity , these insights could assist developments in improved therapeutics for disorders such as epilepsy and chronic pain , with mechanisms relevant to a range of ion transport processes in biology and materials . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"electricity",
"electrophysiology",
"neuroscience",
"organic",
"compounds",
"simulation",
"and",
"modeling",
"ion",
"channels",
"electrostatics",
"basic",
"amino",
"acids",
"amino",
"acids",
"thermodynamics",
"research",
"and",
"analysis",
"methods",
"sodium",
"channels",
"proteins",
"chemistry",
"biophysics",
"free",
"energy",
"physics",
"mutagenesis",
"biochemistry",
"biochemical",
"simulations",
"eukaryota",
"organic",
"chemistry",
"physiology",
"genetics",
"lysine",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"neurophysiology",
"organisms"
] | 2018 | Selective ion permeation involves complexation with carboxylates and lysine in a model human sodium channel |
The gene encoding the GroEL chaperonin is duplicated in nearly 30% of bacterial genomes; and although duplicated groEL genes have been comprehensively determined to have distinct physiological functions in different species , the mechanisms involved have not been characterized to date . Myxococcus xanthus DK1622 has two copies of the groEL gene , each of which can be deleted without affecting cell viability; however , the deletion of either gene does result in distinct defects in the cellular heat-shock response , predation , and development . In this study , we show that , from the expression levels of different groELs , the distinct functions of groEL1 and groEL2 in predation and development are probably the result of the substrate selectivity of the paralogous GroEL chaperonins , whereas the lethal effect of heat shock due to the deletion of groEL1 is caused by a decrease in the total groEL expression level . Following a bioinformatics analysis of the composition characteristics of GroELs from different bacteria , we performed region-swapping assays in M . xanthus , demonstrating that the differences in the apical and the C-terminal equatorial regions determine the substrate specificity of the two GroELs . Site-directed mutagenesis experiments indicated that the GGM repeat sequence at the C-terminus of GroEL1 plays an important role in functional divergence . Divergent functions of duplicated GroELs , which have similar patterns of variation in different bacterial species , have thus evolved mainly via alteration of the apical and the C-terminal equatorial regions . We identified the specific substrates of strain DK1622's GroEL1 and GroEL2 using immunoprecipitation and mass spectrometry techniques . Although 68 proteins bound to both GroEL1 and GroEL2 , 83 and 46 proteins bound exclusively to GroEL1 or GroEL2 , respectively . The GroEL-specific substrates exhibited distinct molecular sizes and secondary structures , providing an encouraging indication for GroEL evolution for functional divergence .
Chaperonins are essential cellular components that are responsible for protein folding , assembly and transport [1]–[6] . Chaperonins are also a major group of heat shock proteins that are over-expressed at high temperatures and have fundamental roles in growth and survival at non-permissive temperatures [6]–[8] . GroEL is a type I chaperonin , and in Escherichia coli , the GroEL chaperonin is required in vivo for the proper folding , at all temperatures , of approximately 300 newly translated polypeptides ( accounting for approximately 10% of the total ) that participate in various physiological processes [9] . Because of its importance in many cellular processes , the groEL gene is ubiquitously distributed in bacteria . Most bacterial species , such as E . coli , possess a single groEL gene , whereas other species ( nearly 30% of bacteria with sequenced genomes ) have evolved multiple groEL copies [1] . The paralogous GroEL proteins are highly similar in sequence and , most likely , in structure . However , some differences exist between duplicated groEL genes , and these duplicated GroEL proteins have evolved to play divergent roles in many different cellular processes in different bacterial species [10]–[15] . Although the mechanisms of functional divergence are important for our understanding of the complexity of evolution , these mechanisms have not been characterized to date . Myxobacteria are δ-proteobacteria with unique and complex multicellular behaviors , such as movement in swarms on solid surfaces , cooperative feeding on macromolecules or other microbial cells and the development of multicellular fruiting bodies containing numerous myxospores against adversity conditions [16] , [17] . Myxococcus xanthus DK1622 is a model myxobacterium with a large genome ( 9 . 14 Mbp ) that includes many duplicated crucial genes [18] . It has been suggested that such duplication is responsible for the complex social behavior of these cells , although this hypothesis has not been experimentally validated . There are two copies of the groEL gene in the genome of M . xanthus DK1622 . Previous studies indicate that either of the two paralogous groEL genes can be deleted in strain DK1622 without affecting cellular viability , although the deletion does result in distinct defects in the cellular heat-shock response , predation and development [15] . In this study , we investigated the effects of the substrate selectivity of the GroEL proteins and the expression levels of the duplicated groEL genes on the functional divergence of heat-shock responses , development and predation . We performed a comparative proteomics analysis of the substrate specificity of the two GroELs , and the relationships between the structural differences and substrate specificity were investigated using bioinformatics , molecular swapping and site-directed mutagenesis .
Otani et al . found that , although GroEL1 and GroEL2 are among the major proteins induced by heat shock , the density of GroEL2 spots in two-dimensional electrophoretic gels is much lower than that of GroEL1 [19] . It was also noted that the expression levels of groEL1 and groEL2 were not equal in the wild-type strain DK1622 in CTT growth medium and TPM development medium and that the two groEL genes played distinct roles in heat-shock responses , development and predation [15] . To assess the changes in groEL1 or groEL2 expression in groEL-deletion mutants and whether these changes contribute to functional divergence , we inserted groEL1 or groEL2 , each with its own promoter , into the genome of groEL1- or groEL2-deletion mutants at the attB integration site using pSWU30 , producing four groEL-complemented strains ( Table S1 ) . The groEL expression levels and cell viability were compared between these mutants and the wild-type strain DK1622 following heat shock at 42°C for 30 min . Quantitative PCR assays indicated that the expression of the groEL genes was regulated in a complex manner in different mutants upon heat shock ( Figure 1 ) . In the wild-type strain DK1622 , the groEL2 expression level was only one-quarter of that of groEL1 after heat shock . The deletion of groEL1 ( strain YL0301 ) led to an increase in the groEL2 expression level ( approximately twofold ) . The expression of groEL1 inserted in YL0301 ( strain YL0901 ) was approximately half that in DK1622 under the heat shock conditions , but the presence of exogenous groEL1 had no obvious effect on the expression level of groEL2 ( P>0 . 05 ) . Thus , the total expression of all the groEL genes in YL0901 was similar to that in DK1622 . In YL0902 , which contained an additional groEL2 gene , the total expression of groEL2 also doubled , reaching a level equal to four times that of groEL2 in DK1622 . In YL0302 , the deletion of the groEL2 gene led to reduced expression of the groEL1 gene under the heat shock conditions ( approximately 60% of that in DK1622 ) . Transformation of the YL0302 mutant ( strain YL0906 ) with another groEL1 gene increased the total groEL expression level to that of DK1622 ( P>0 . 05 ) . The total expression level of groEL1 and groEL2 in the groEL2-complemented YL0302 mutant ( strain YL0907 ) also reached the level in DK1622 ( P>0 . 05 ) . Interestingly , although YL0902 had two groEL2 genes and no groEL1 , the survival rates of both YL0901 and YL0902 were similarly increased after heat shock at 42°C for 30 min , paralleling the increase in the total groEL expression level in these two mutants ( Figure 1 ) . The survival rates of the YL0906 and YL0907 mutants after the heat shock treatment also corresponded to an increase in the total groEL expression level . These results suggest that the lethal nature of the heat shock in the groEL1 deletion mutant ( YL0301 ) and the increased sensitivity of the groEL2 deletion mutant ( YL0302 ) to high temperatures [15] result from a significant decrease in the total expression of GroEL , leading to an insufficient level of GroEL proteins to facilitate the refolding of denatured proteins . This is consistent with a model where there is a threshold level of GroEL beneath which cells cannot survive . When the total expression of groEL is higher than the threshold , there is a positive correlation between the survival rate and groEL expression in M . xanthus DK1622 cells after heat shock ( r = 0 . 98 , P<0 . 01 ) ( Figure S1 ) , a result which supports the hypothesis that , after duplication , both groEL1 and groEL2 retain fundamental functions by balancing their expression dosage [20] , [21] . Because the deletion of groEL1 and the deletion of groEL2 result in deficiencies in development and predation , respectively [15] , we performed a development assay on DK1622 , YL0301 , YL0901 and YL0902 and predation assays on DK1622 , YL0302 , YL0906 and YL0907 . Figure 2A shows the expression levels of groEL1 and groEL2 in different strains after 12 , 36 and 60 h of incubation on TPM development medium . When groEL1 was inserted into the genome of the groEL1-deletion mutant ( strain YL0901 ) , the developmental defect was mostly reversed , with sporulation reaching 70%–80% of that of DK1622 . However , although YL0902 ( containing two copies of groEL2 ) had a total groEL expression level similar to that of YL0901 at different developmental stages , YL0902 displayed a development defect similar to that of YL0301 , and the sporulation ability of YL0902 was only approximately 20% of that of the wild-type strain DK1622 ( Figure 2B and 2C ) . The insertion of groEL1 into YL0302 ( strain YL0906 ) did not improve the predation feeding ability of cells on an E . coli mat , and the swarming time of YL0906 to the edge of the E . coli colony was 60–65 h , which is similar to that of YL0302 ( P>0 . 05 ) . When groEL2 was inserted into YL0302 ( strain YL0907 ) , the swarming time to the E . coli colony edge decreased to 40 h ( Figure 3 ) . Because the presence of living E . coli cells in the E . coli predation experiments might affect the qPCR assay , we instead performed the analysis using a liquid feeding assay with casein as the only nutrient [15] . The total expression level of groEL1 and groEL2 was also similar in the YL0906 and YL0907 mutants , suggesting that the changes in expression are not the major contributors to functional divergence ( Figure S2 ) . The above results indicate that although the distinct heat-shock responses of the groEL1 and groEL2 mutants were determined by the total groEL expression level , the divergent functions of groEL1 and groEL2 in development and predation are the result of the substrate specificity of the corresponding GroEL chaperonins . To explore the evolutionary relationships of groEL , we compared the M . xanthus GroEL sequences with those of ten genome-sequenced bacteria , including three Myxobacteria , three Actinobacteria , three Cyanobacteria , and E . coli ( Figure 4 and Table S2 ) . With the exception of E . coli , these species possess duplicated groEL genes . The maximum likelihood tree showed that the GroELs from Myxobacteria , Actinobacteria and Cyanobacteria clustered separately ( Figure 4A ) , suggesting that the groEL gene duplication originated from three independent evolutionary events in these three taxa . We further calculated the Ka/Ks values of these orthologous groEL genes ( Figure 4B and Table S3 ) . The average Ka/Ks values for Actinobacteria groEL2 and Cyanobacteria groEL1 were less than 0 . 1 , suggesting that they are highly evolutionarily conserved . It has been reported that groEL2 in the three Actinobacteria species [12] , [22]–[24] and groEL1 in the three Cyanobacteria species [25]–[27] are housekeeping genes , which is consistent with their Ka/Ks values . In M . xanthus , the Ka/Ks value for groEL1 was significantly lower than that for groEL2 ( P<0 . 05 ) but significantly higher than that for the housekeeping groEL genes in Actinobacteria or Cyanobacteria ( P<0 . 01 ) . These results suggest that both of groEL1 and groEL2 in M . xanthus are suffered weak selection pressure , consistent to the finding that the deletion of either gene alone does not affect cell viability [15] . Based on their structural characteristics and sequence conservation , the GroEL protein sequences have been divided into five regions , i . e . , an N-terminal equatorial region , an N-intermediate region , an apical region , a C-intermediate region and a C-terminal equatorial region ( Figure S3 ) [28] . The two intermediate regions have the highest level of conservation between M . xanthus DK1622 GroEL1 and GroEL2 , i . e . , 97 . 7% and 97 . 2% identities for the N- and C-intermediate regions , respectively . The identities for the other three regions are 81% for the N-terminal equatorial region , 75 . 4% for the apical region , and 62 . 6% for the C-terminal equatorial region . Further Ka/Ks analysis showed similar sequence characteristics for the five GroEL1 and GroEL2 regions in the four Myxobacterial species referred to above ( Figure 4C ) . For example , the Ka/Ks values of the N- and C-intermediate regions were very low ( <0 . 05 ) , suggesting that these two regions are highly conserved; in contrast , the other three regions had higher Ka/Ks values ( >0 . 3 ) , suggesting these regions are most likely involved in the functional divergence of GroEL1 and GroEL2 . A sequence alignment showed that the high Ka/Ks values of the C-terminal equatorial regions were largely due to the variability of the C-terminal tail sequences . For example , the C-terminal tail of GroEL1 in M . xanthus was composed of six repeated GGM motifs , similar to that of E . coli GroEL , whereas the C-terminal tail of GroEL2 is greatly different . It was also noted that there are substantial differences in the C-terminal sequences between the duplicated GroELs ( Figure 4A ) [1] . To clarify the relationships between the structural and functional divergence , we designed a series of single region-swaps between the groEL1 and groEL2 genes to determine the roles of the regions and their contributions to functional divergence . The swapped regions included the N-terminal equatorial , apical , and C-terminal equatorial regions between GroEL1 and GroEL2; the two highly similar intermediate regions were not included . The groEL2 hybrids containing the N-terminal equatorial , apical or C-terminal equatorial region of groEL1 were inserted into the genome of the groEL1-deletion mutant YL0301 using pSWU30 , producing the region-swapped strains YL0903 , YL0904 and YL0905 , respectively . Similarly , the groEL1 hybrids with a swapped N-terminal equatorial , apical or C-terminal equatorial region of groEL2 were inserted into the genome of the groEL2-deletion mutant YL0302 to produce the mutant strains YL0908 , YL0909 and YL0910 , respectively ( Figure 5A ) . Because the groEL2 mutant displays defective cellular predation and the groEL1 mutant displays deficient development and sporulation [15] , region swapping was performed in YL0301 using the groEL1 chimeras and YL0302 using the groEL2 chimeras . Detailed descriptions of these mutants are listed in Tables S1 and S4 and Figure S4 . The development and predation phenotypes of the region-swapped mutants were assayed using the intact groEL1- and groEL2-complemented mutants as controls . The results showed that the developmental defect of the groEL1-deletion mutant was not reversed by GroEL2-equatorial-NGroEL1 ( YL0903 ) . The sporulation ability of YL0903 was approximately 20% of that of DK1622 , which was the same as that of YL0301 . However , the fruiting bodies of YL0904 ( YL0301 complemented with GroEL2-apicalGroEL1 ) were more similar to the fruiting bodies of the wild-type strain DK1622 than to the fruiting bodies of YL0301 , and the sporulation ability also increased to 55%–65% of that of DK1622 . The sporulation of the strain complemented with GroEL2-equatorial-CGroEL1 ( YL0905 ) was 30%–40% of that of DK1622 ( Figure 5B ) . In the predation experiments , the single swapped region in YL0908 did not noticeably improve the predation defects , which were similar to those of YL0906 ( P>0 . 05 ) . However , the YL0909 strain significantly recovered its predation ability , which was similar to that of YL0907 . These two mutants spread to the edge of the E . coli colonies within 40–45 h . The predation defect was also improved to some extent in YL0910 , which required 55–60 h to reach the edge of the E . coli mat ( Figure 5C ) . Accordingly , the apical region and the C-terminal equatorial region determine the substrate preference , thus causing the functional divergence of the duplicated chaperonins with respect to development and predation; conversely , the N-terminal equatorial region has almost no effect . In addition , we deleted the repeated GGM region ( GGMGGMGGMGGMGGMGM ) from GroEL1 in M . xanthus DK1622 , producing the YL1001 mutant ( Figure 5A ) . Compared with the wild-type DK1622 , the mutant was markedly defective in development , and the sporulation ability of YL1001 was only 32 . 6% of that of DK1622 ( Figure 5D ) . Furthermore , we swapped three of the six GGM repeats with YGGDDMDY in DK1622 , the corresponding sequence in GroEL2 , producing the YL1002 mutant . Similar to YL1001 , YL1002 was also defective in development and exhibited a decreased sporulation ability ( 38 . 1% of that of DK1622 ) ( Figure 5D ) . These results demonstrate that the GGM repeated region is necessary for GroEL1 to perform its normal functions in development . To identify the proteins that interact with GroEL1 and GroEL2 in M . xanthus DK1622 , immunoprecipitation assays were performed using the groEL1- and groEL2-deletion mutants ( strains YL0301 and YL0302 ) , and the bound proteins were subjected to mass spectral identification . Most of the non-specific substrates identified using two negative controls ( see Methods ) were ribosomal proteins ( Table S5 ) . This result was consistent with the results for E . coli [9] . After removing the non-specifically bound proteins , 151 and 114 proteins were found to be bound by GroEL1 and GroEL2 , respectively . Of the bound proteins , 68 were bound to both GroEL1 and GroEL2 ( GroEL1/2 ) , whereas 83 and 46 proteins bound exclusively to GroEL1 and GroEL2 , respectively ( Table S5 ) . Of the functionally annotated GroEL1/2 substrates ( 58/68 , 85 . 3% ) , many had functions or predicted functions related to fundamental physiological cellular processes; examples of such substrates are succinyl coenzyme A synthetase and isocitrate dehydrogenase , two key enzymes of the citric acid cycle [29] , [30] . This result is consistent with the fact that either the groEL1 or groEL2 gene could be deleted without affecting cellular growth but that the double deletion of groEL1 and groEL2 resulted in inviable cells [15] . However , except for PilA , no proteins involved in M . xanthus social behavior were found to bind to both GroEL1 and GroEL2 . In contrast , of those annotated proteins that were exclusively bound by GroEL1 or GroEL2 ( accounting for 75 . 9% and 76 . 1% of bound proteins , respectively ) , a considerable number are involved in the social behaviors of M . xanthus DK1622 ( Table S5 ) . For example , the frz signal transduction system is well known to play important roles in development process of M . xanthus DK1622 [31] . The frizzy aggregation protein FrzCD [31] is in the substrate list of GroEL1 . Besides , Flp pilus assembly protein CpaB [18] , sensor histidine kinase/response regulator CheA4 [32] and Type IV pilus secretin PilQ [33] were found to be specific substrates of GroEL1 , whereas type IV pilus assembly ATPase PilB [34] , gliding motility protein MglA [35] , type IV pilus biogenesis protein PilM [36] and several proteins related to the biosynthesis of secondary metabolites were found to be exclusively bound by GroEL2 . These results are consistent with the hypothesis that GroEL is an essential component and that the duplicated groEL genes evolved to participate in various complex physiological processes in Myxococcus cells . The structural characteristics of the substrate proteins were further analyzed by comparing their secondary structures with the known protein domain classification database CATH [37] . After excluding the proteins that had low E-values ( >0 . 001 ) , we obtained 36 reliable secondary structures for the 68 identified GroEL1/2 substrates , 38 for the 83 GroEL1-specific substrates and 34 for the 46 GroEL2-specific substrates . It is known that proteins with β-sheets exposed to the hydrophobic surface and packed with the hydrophobic surfaces of α-helices ( called the αβ domain ) have high-affinity interactions with the apical region of GroEL and are normally present as substrates of GroELs [38] . As expected , most GroEL1/2 substrates ( 34 of 36 ) contain at least one αβ domain . It is interesting that , although 31 of the 34 ( 91 . 18% ) GroEL2-specific substrates possess at least one αβ domain , only 27 of the 38 ( 71 . 05% ) GroEL1-specific substrates contain an αβ domain ( Figure 6A , Table S5 ) . Another interesting difference is the difference in the molecular sizes of the GroEL1 and GroEL2 substrates . According to the current model of GroEL [28] , [39]–[45] , substrate selection is heavily dependent upon the adaptation of a substrate molecule to the cavity volume of the GroEL chaperonin , which may change in response to GroEL sequence changes . Previous studies have shown that GroEL strongly prefers to act on proteins with a molecular weight ranging from 20 kDa to 60 kDa [46] . The average molecular weight of GroEL1-specific substrates was significantly smaller than that of GroEL2-specific substrates ( Figure 6B; P<0 . 05 ) . For example , while 51 . 8% ( 43 of 83 ) of the GroEL1-specific substrate proteins were less than 40 kDa , the molecular weights of only 21 . 7% ( 10 of 46 ) of the GroEL2-specific proteins were less than 40 kDa . A third important characteristic is the pI value of the substrate; there was no significant difference between the GroEL1 and GroEL2 substrates with respect to pI ( Figure 6B; P>0 . 05 ) .
Duplication is a major source of new genes and is equally important in Bacteria , Archaea and Eukarya [47] , [48] . The duplication of the GroEL chaperonin gene has occurred in many different bacterial cells as part of the evolution of complexity [49] . M . xanthus DK1622 is well known for its complex multicellular behaviors [16] , [17] , and this strain possesses a large genome ( 9 . 14 Mb ) in which there are many duplicated genes , including two copies of groEL [18] . In addition to participating in fundamental processes involved in cellular growth , the two duplicated groEL genes have been demonstrated to play distinct roles in heat-shock responses , development and predation in DK1622 [15] . The results described in this report show that the divergent functions of GroEL1 and GroEL2 in various physiological processes result from different mechanisms . The groEL expression level is the key reason for the difference in the heat-shock response after the deletion of groEL1 or groEL2 , suggesting that the duplicated groEL genes in Myxococcus have similar functions in cell survival . These functions are most likely similar to their fundamental function in cell growth at normal temperatures . Either of the two groEL genes can be deleted without significantly affecting cell growth , but at least one groEL gene is required for cell survival [15] . In contrast , the functional divergence of the duplicated GroELs with respect to their roles in development and predation processes reflects in their substrate specificity , which has been suggested to evolutionarily relate to the unusual social behavior of Myxococcus . The co-substrates of GroEL1 and GroEL2 have consistently been shown to be essential cellular components , but the duplicated GroEL chaperonins have also evolved their own substrate preferences related to late-appearing cellular processes , such as social behaviors and PKS/NRPS biosynthesis . The evolutionary models for the functional divergence of duplicated genes are likely to include neofunctionalization , subfunctionalization , or a combination of thereof [20] , [47] , [50] . Thus , the functional divergence of the duplicated groELs in M . xanthus is likely a combination of neofunctionalization and subfunctionalization , i . e . , the subneofunctionalization model [50] . Although extensive studies have demonstrated that the duplicated groEL genes play distinct roles in different cellular physiological processes [10]–[15] , an understanding of the mechanisms involved in their functional divergence will provide insight into bacterial evolution . The GroEL proteins have been divided into five regions based on structural characteristics and sequence conservation [28] . Previous studies have shown that GroEL chimeras bearing equatorial or apical regions exchanged between M . tuberculosis and E . coli retained the normal chaperonin functions of GroEL [51] . Bioinformatics analyses indicate that the duplicated GroELs from different bacteria share similar characteristics: the N- and C-intermediate regions are highly conserved , suggesting that these regions have essential functions in maintaining the functional structure of GroELs , and the apical , N-terminal and C-terminal regions are much more flexible , suggesting their possible roles in functional divergence . The region-swapping experiments indicated that the functional divergence of the duplicated GroELs in M . xanthus was caused by the apical and C-terminal regions . The GGM repeat at the tail of the GroEL1 C-terminal region , which is similar to that of E . coli GroEL , is important for the distinct functions of GroEL1 in development and sporulation . These results are consistent with the positions of these two regions in the GroEL oligomeric complex , i . e . , the apical region is at the opening through which substrates enter the central cavity , and the C-terminal equatorial region is at the bottom of the cavity [43] . However , it remains unclear whether region swapping has effects on in vivo chaperonin functions . To address the question , we assayed the survival rates of the mutants YL0903 , YL0904 and YL0905 in response to heat shock and found that all the mutants rescued the lethality of heat-shock observed for YL0301 . However , the survival rates of YL0903 , YL0904 , and YL0905 were low compared with the strains complemented with an intact groEL gene ( Figure S5 ) . This result suggests that the region-swapped GroELs function in M . xanthus cells but that these functions were affected , at least at non-permissive temperatures . It is still unclear whether the chimeras interact with intact GroELs and whether the in vivo functions of the chimeras result from mixed GroEL complexes . Furthermore , although the protein substrates of the M . xanthus GroELs were consistent with those of the single E . coli GroEL with regard to their secondary structural features [9] , the substrate spectra varied significantly . This variation is most likely due to the low level of sequence similarity between the E . coli GroEL and the M . xanthus GroELs ( E . coli GroEL is 67 . 3% and 65 . 2% similar to M . xanthus GroEL1 and GroEL2 , respectively ) and to the difference in the protein substrates between these two bacteria . Therefore , there are many questions related to the GroEL chaperonins and their functional divergence that need to be addressed .
The strains and plasmids used in this study are listed in Table S1 . For the growth assays , the M . xanthus strains were cultivated in the Casitone-based nutrient-rich CTT medium [52] . The E . coli strains were routinely grown on Luria-Bertani ( LB ) agar or in LB broth . E . coli was grown at 37°C , whereas the Myxococcus strains were incubated at 30°C . When required , 40 µg/ml of kanamycin ( Km ) and 10 µg/ml of tetracycline ( Tet ) ( Sigma ) were added to the medium . The groEL expression levels during heat shock and liquid predation were analyzed using quantitative real-time PCR . M . xanthus DK1622 and other mutants were harvested after 18 h and exposed to 42°C for 1 h . The RNA was extracted immediately using a total RNA extraction kit following the manufacturer's instructions ( Promega ) . Contaminating DNA was removed with a DNAfree kit ( Ambion ) . The purified RNA was transcribed to yield cDNA , which was stored at −70°C . The quantitative real-time PCR was performed using a Bio-Rad sequence detection system with 250 nM primers , 10 µl of SYBR Green PCR Master Mix ( Bio-Rad ) , 7 µl of RNase-free water , and 2 µl of cDNA template . The PCR was performed for 3 min at 95°C , followed by 40 cycles of 30 s at 95°C , 30 s at 59°C , and 15 s at 72°C . The 16S rRNA was used as a normalization signal . Calibration curves ( groEL1 , groEL2 , and 16S RNA ) were generated using 10-fold dilutions of M . xanthus DK1622 genomic DNA . The following pairs of forward and reverse primer pairs were used: groEL1 , 5′-CACCGAGACGGAGATGAAGG-3′ and 5′-TGAGGCAGCGGATGTAGGC-3′; groEL2 , 5′-ATCCGCACGCAGATTGAC-3′ and 5′-GCfCTTCTTCTCCTTCATCTCC-3′; and 16S rRNA , 5′-CGCCGTAAACGATGAGAA-3′ and 5′-TTGCGTCGAATTAAACCAC-3′ . The groEL expression levels during predation were analyzed using quantitative real-time PCR . The strains were cultured for 50 h in medium containing casein as a substrate instead of hydrolyzed proteins , and the RNA was extracted immediately . The method and the primers used were the same as those described above . The groEL expression level during different developmental stages was analyzed by measuring the β-galactosidase activity , as described by Li et al . [15] , [53] , with minor modifications . The cells were broken using a Mini-Beadbeater ( BioSpec ) at a speed of 2500 rpm . The β-galactosidase activity was determined using o-nitrophenyl-β-galactopyranoside ( Sigma ) , and the samples were analyzed at 420 nm . The total protein concentration was determined using the bicinchoninic acid protein assay ( Pierce ) . The specific activity was calculated as follows: specific activity = 213×A420/ ( sample volume×protein concentration×reaction time ) [15] , [54] . M . xanthus cells were harvested at mid-logarithmic phase and suspended to a final density of 5×109 cells/ml in TPM buffer . Aliquots ( 10 µl ) were spotted onto TPM agar , and the plates were cultivated at 30°C and observed every 24 h to monitor the formation of fruiting bodies . The sporulation rate was measured after 5 days as previously described . The assays were performed at least three times [15] . The predation assays were performed according to the method used in a previous study [15] . E . coli and M . xanthus cultures were harvested at mid-logarithmic phase and washed three times with 10 mM MOPS buffer ( pH 7 . 6 ) . The final cell densities of the cultures were 5×109 cells/ml for M . xanthus and 1×1011 cells/ml for E . coli . Then , 50 µl of E . coli was pipetted onto a plate to form a 1-cm-diameter colony , and 2 µl of M . xanthus was added to the center of the E . coli colony , with an inoculation diameter of 0 . 15 cm . The assay was repeated at least three times . The plates were incubated at 30°C for 6 days , during which time the size of the M . xanthus growth area was recorded every 12 h . The predation ability of M . xanthus was reported as the time required for M . xanthus to spread to the edge of the E . coli colonies . M . xanthus cultures were harvested as described above . The cells were heat shocked for 30 min at 42°C , serially diluted and plated on CTT agar . After 6 days incubation , the CFUs were calculated [15] . The groEL gene sequences from ten genome-sequenced bacterial strains were retrieved from the NCBI database ( Table S2 ) , and the amino acid sequences were aligned using the protein sequence alignment program in CLUSTALW [55] . A maximum likelihood tree was constructed using MEGA5 [56] . The Ka/Ks values among orthologous groEL genes or among paralogous groEL genes were calculated using KaKs_Calculator 1 . 2 [57] with the NG , MLWL and MLPB models [58] , [59] . The region-swapping assay was conducted according to a previously published method [51] . The regions responsible for the developmental defects of YL0301 and the predation defects of YL0302 were investigated by incorporating single groEL regions into YL0301 or YL0302 . The complementation mutants were constructed with the site-specific integration plasmid pSWU30 . The apical region of groEL1 was inserted into YL0301 to obtain YL0904 ( YL0301::pSWU- groEL2-apicalgroEL1 ) . Briefly ( Figure S4 ) , 0 . 5 kb of the upstream sequence and the N-terminal region ( bp 1–597 ) of groEL2 and the C-terminal ( 597 to the end ) of groEL1 were amplified by PCR . The two fragments were spliced by fusion PCR , digested with XbaI and BamHI , and ligated into pSWU30 digested with XbaI and BamHI . The plasmid was transferred to E . coli λ-pir cells , and the plasmid DNA was extracted from the Tet-resistant transformants using the eZNA Plasmid Mini Kit I ( Omega Bio-Tek ) according to the manufacturer's instructions . The correct plasmid was used as the template in the second round of fusion PCR . The plasmid containing the correct sequence was transferred by electroporation into YL0301 , and individual Tet-resistant colonies were screened . The mutant phenotypes were observed to determine the effects of the apical region on development . The same method was used to replace other regions . The primers used are listed in Table S4 . The GGM region deletion mutants were constructed using the positive-negative KG cassettes described by Ueki et al . Briefly , the upstream sequence ( before the GGM sequence ) and the downstream sequence ( after the GGM sequence ) were amplified by PCR . The two fragments were fused to the XbaI restriction site to construct homologous fragments with in-frame deletions . These homologous fragments were ligated into SmaI-digested pBJ113 . The resulting plasmid containing the correct sequence was transferred by electroporation into DK1622 . The second round of screening was then performed on CTT plates containing 1% galactose ( Sigma ) . The deletion mutants that grew on galactose but were sensitive to kanamycin were identified and verified by PCR and sequencing . The GGM sequence-swapping mutants were constructed in a similar manner . The upstream sequence ( 739–1623 of groEL1+the DNA sequence corresponding to the last eight amino acids in groEL2 ) and the downstream sequence ( the DNA sequence corresponding to the last eight amino acids in groEL2+815 bp downstream of groEL1 ) were PCR amplified . The two fragments were fused to the XbaI site to construct in-frame deletion fragments that were ligated into SmaI-digested pBJ113 . The resulting plasmid was subjected to two rounds of screening to obtain the GGM-swapping mutant . The phenotypes of the mutants were observed to determine the effects of the GGM repeat region on development . The primers used are listed in Table S6 . groEL1-ko and groEL2-ko mutants were resuspended in Tris buffer ( 50 mM Tris , 150 mM NaCl , 5 mM EDTA , and 20 µl PMSF ) after cultivation and washed three times . The cells were lysed using a high-pressure homogenizer . Aliquots of 50 µl of protein A/G beads were added to 10 ml of supernatant solution to remove the proteins that non-specifically bound to the beads . An anti-GroEL antibody ( 3 mg/ml ) was added , and the solution was shaken at low speed at 4°C . Aliquots of 100 µl of protein A/G beads were mixed with 50 ml of solution and incubated for another 2 h . The beads bound by the GroEL substrates were washed with Tris buffer three times , and the beads and proteins were separated with lysis buffer . The substrates of GroELs were identified by high-pressure liquid chromatography-tandem mass spectrometry ( LC-MS/MS ) by Shanghai Zhongke Biotech Company . A negative control assay was performed with DK1622 using the same protocol in the absence of the antibody against GroEL to exclude non-specific binding between the beads and proteins [38] . To exclude non-specific binding between the antibody and proteins and to exclude non-specific binding between GroEL and proteins after cell lysis , another negative control was performed by adding 0 . 1% SDS to the lysis buffer to separate GroEL from its substrates [60] . The solution was diluted 50 fold after cell lysis and the addition of extra GroEL1 and GroEL2 protein . The antibody against GroEL was then added to identify non-specifically bound proteins [38] , [60] . | GroEL is a type I chaperonin , involved in protein folding , assembly , and transport . It is a major group of heat-shock proteins that are over-expressed at high temperatures and has fundamental roles in growth and survival at non-permissive temperatures . Because of its importance in many cellular processes , the groEL gene is ubiquitously distributed in bacteria . Most bacterial species possess a single groEL gene , while others ( close to 30% of sequenced bacterial genomes ) have two or more groEL copies . Many studies have described the functional divergence of duplicated groEL genes in different bacterial species , but the involved mechanisms have not yet been characterized . Myxobacteria are characterized by their unique multicellular behaviors . Myxococcus xanthus DK1622 , the model strain of myxobacteria , possesses a large genome ( 9 . 14 Mb ) , containing many gene duplications , including two copies of the groEL gene . Gene duplications and their functional divergence are suggested for complex cellular behaviors , which , however , have not yet been testified . In this paper , using combined proteomic and genetic approaches , we explored how the duplicated groEL genes of M . xanthus DK1622 evolved to fit the functional divergence for social behaviors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genetics",
"biology",
"evolutionary",
"biology",
"genomic",
"evolution",
"genetics",
"and",
"genomics"
] | 2013 | Mechanisms Involved in the Functional Divergence of Duplicated GroEL Chaperonins in Myxococcus xanthus DK1622 |
Zinc finger MYND-type-containing 10 ( ZMYND10 ) , a cytoplasmic protein expressed in ciliated cells , causes primary ciliary dyskinesia ( PCD ) when mutated; however , its function is poorly understood . Therefore , in this study , we examined the roles of ZMYND10 using Zmynd10–/–mice exhibiting typical PCD phenotypes , including hydrocephalus and laterality defects . In these mutants , morphology , the number of motile cilia , and the 9+2 axoneme structure were normal; however , inner and outer dynein arms ( IDA and ODA , respectively ) were absent . ZMYND10 interacted with ODA components and proteins , including LRRC6 , DYX1C1 , and C21ORF59 , implicated in the cytoplasmic pre-assembly of DAs , whose levels were significantly reduced in Zmynd10–/–mice . LRRC6 and DNAI1 were more stable when co-expressed with ZYMND10 than when expressed alone . DNAI2 , which did not interact with ZMYND10 , was not stabilized by co-expression with ZMYND10 alone , but was stabilized by co-expression with DNAI1 and ZMYND10 , suggesting that ZMYND10 stabilized DNAI1 , which subsequently stabilized DNAI2 . Together , these results demonstrated that ZMYND10 regulated the early stage of DA cytoplasmic pre-assembly by stabilizing DNAI1 .
Primary ciliary dyskinesia ( PCD ) is an autosomal-recessive disorder caused by defective motile cilia or flagella that is characterized by respiratory distress , impaired mucociliary clearance , chronic cough , sinusitis , bronchiectasis , male infertility , laterality defects , and cardiac anomalies in term neonates [1 , 2] . To date , mutations in about 30 genes have been linked to PCD in approximately 50–70% of cases [3] . In motile cilia or flagella , the outer dynein arm ( ODA ) and inner dynein arm ( IDA ) are attached to the peripheral microtubules of the 9+2 axoneme with a fixed periodicity and generate ATP-dependent motion . Dynein arms are large , multisubunit protein complexes comprised of light , intermediate , and heavy chains [4] . The latter have ATPase activity , which provides power for the sliding between microtubules in the beating cilia . The capacity of the dynein arm to function as molecular motors depends on the integrity of its components . ODA consists of two to three heavy chains , two or more intermediate chains , and a cluster of four to eight light chains , whereas IDA has a more diverse composition [4 , 5] . As such , several genes linked to PCD encode dynein arm components , including dynein axonemal light chain 1 ( DNAL1 ) , dynein axonemal intermediate chain 1 ( DNAI1 ) , DNAI2 , dynein axonemal heavy chain 5 ( DNAH5 ) , and DNAH11 [6–10] . Mutations in DNAH6 , a heavy chain of IDA , were detected in individuals with heterotaxy and ciliary dysfunction [11] . Dynein arms are pre-assembled in the cytoplasm and transported into motile cilia , where they are docked onto peripheral microtubules; however , the underlying mechanisms are poorly understood [12] . Dynein axonemal assembly factors ( DNAAFs ) are involved in the pre-assembly of dynein arms , and their mutation is linked to PCD [13–16] . There are five known DNAAFs—leucine-rich repeat-containing 50 ( LRRC50 , DNAAF1 ) [13] , kintoun ( KTU , DNAAF2 ) [14] , DNAAF3 [15] , dyslexia susceptibility 1 candidate 1 ( DYX1C1 , DNAAF4 ) [16] , and HEAT repeat-containing protein 2 ( HEATR2 , DNAAF5 ) [17] . KTU and DYX1C1 interact with chaperone proteins , including heat shock protein 70 ( HSP70 ) , HSP90 , and T-complex chaperonin complex [14 , 16] . DNAAFs work with a chaperone complex to facilitate the proper folding of heavy chains and their assembly with intermediate chains [15] . Defects in DNAAFs result in the loss of ODAs and IDAs from axonemes [13–16] . In addition to DNAAFs , mutations in several cytoplasmic proteins , including LRRC6 [18 , 19] , zinc finger MYND-type-containing 10 ( ZMYND10 ) [20 , 21] , chromosome 21 open reading frame 59 ( C21ORF59 ) [22] , PIH1 domain-containing 3 ( PIH1D3 ) [23 , 24] , and armadillo repeat-containing 4 ( ARMC4 ) [25] , have been identified as those causing PCD when defective . Mutations in LRRC6 , ZMYND10 , C21ORF59 , and PIH1D3 cause ODA and IDA defects [18–24] , whereas those in ARMC4 cause ODA defects [25] . Thus , these proteins likely function at different stages of pre-assembly or have ODA- or IDA-specific roles . Overall , these proteins are expected to be involved in the pre-assembly of dynein arms based on their cytoplasmic localization and consequences upon loss of expression . However , the specific functions of these proteins and their relationships with DNAAFs are not well understood . ZMYND10 ( also known as BLU ) has a myeloid , nervy , and DEAF-1 ( MYND ) -type zinc finger domain at its C-terminus that engages in protein-protein interactions [26] . ZMYND10 is highly enriched in ciliated cells compared with that in nonciliated cells [27] and is expressed in motile ciliated tissues in mice [21] . ZMYND10 has been shown to interact with LRRC6 [20 , 21] , although its function in motile ciliated cells is not known . Therefore , in this study , we generated and characterized Zmynd10−/− mice and found that they recapitulate phenotypic aspects of human PCD , including the absence of ODA and IDA without defects in ciliogenesis or 9+2 axonemal structure . We also found that the levels of DNAI1 and DNAI2 in ODA were reduced in Zmynd10−/− mice . ZYMND10 binds to and stabilizes DNAI1 , thereby facilitating the assembly of intermediate chains . Our data suggested that ZMYND10 may play a role in the pre-assembly of dynein arms by regulating the expression of dynein arm components at the protein , and not the mRNA level and promoting their assembly into a cytoplasmic protein complex .
To investigate the function of ZMYND10 , we generated mice with targeted deletion of the Zmynd10 gene locus ( S1A and S1B Fig ) using a lacZ-containing targeting cassette ( Zmynd10tm1[KOMP]Wtsi ) . β-Galactosidase activity staining of Zmynd10+/− lung tissues on postnatal day ( P ) 1 revealed Zmynd10 expression in the bronchus and bronchioles , but not in the alveoli ( S1C and S1D Fig ) . This expression pattern was consistent with that in the previous in situ hybridization results in mouse lung tissues [21] . Zmynd10 expression was also observed in the testes of Zmynd10+/− mice at P28 as well as in the spermatids and earlier-stage germ cells ( S1E and S1F Fig ) . Deletion of coding exons 2–11 yielded a Zmynd10-null allele ( Zmynd10−/− ) , and western blotting and immunofluorescence analyses of the testis lysates and tracheal tissue , respectively , confirmed the absence of ZMYND10 ( S2 Fig ) . Zmynd10−/− mouse litters conformed to Mendelian ratios , and neonates showed no gross abnormalities , indicating that loss of Zmynd10 did not cause embryonic lethality; however , mutant mice exhibited growth retardation and were visibly smaller at P10 , with all eventually dying within 1 month of birth ( S3A–S3C Fig ) , with a mean survival of 14 days . Zmynd10−/− mice developed hydrocephaly and subsequent abnormal head morphology , with complete penetrance ( Fig 1A and 1B ) , and the cerebral ventricles were dilated with cortical tissue thinning ( Fig 1C–1F ) . In addition , 42% of the Zmynd10−/− mice showed laterality defects , including reversal of the heart apex , stomach , liver , or spleen ( S3E and S3F Fig ) . Alcian Blue staining of the paranasal cavities revealed mucosal congestion in Zmynd10−/− mice but not in wild-type littermates , suggesting defective mucociliary clearance ( Fig 1G and 1H ) , which is a prominent manifestation of PCD leading to recurrent airway infection . Inflammation in the lung tissue was never observed in Zmynd10−/− mice that died before P20; however , some of the mice that survived past P25 showed severe pulmonary inflammation in mutants , as evidenced by loss of alveolar architecture , thickening of alveolar septae , collapse of the alveolar space , and infiltration of inflammatory cells and fibroblasts ( Fig 1I–1J and S3D Fig ) . Fibrosis was not observed in Masson trichrome staining ( Fig 1K and 1L ) . Taken together , these data indicated that loss of Zmynd10 induced defects consistent with PCD . Since the Zmynd10−/− phenotype suggested defects in motile cilia , we examined their ultrastructure by transmission electron microscopy ( TEM ) . The tracheal cilia and basal bodies were abundant in the tracheal and lung epithelial cells in both Zmynd10+/+ and Zmynd10−/− mice ( Fig 2A–2D and S4 Fig ) ; however , in the latter , some areas of the tracheal epithelium were surrounded by cellular debris and mucus ( S4 Fig ) . Analysis of the tracheal cilia cross-sections revealed a typical 9+2 microtubular structure in both wild-type and Zmynd10−/− mice ( Fig 2E and 2F ) . In contrast , cilia in Zmynd10−/− mice lacked both ODA and IDA structures ( Fig 2H ) , which were present in the peripheral microtubules of Zmynd10+/+ mice ( Fig 2G ) . This is in agreement with the previous results of studies on humans subjects with ZMYND10 mutations who lacked ODA and IDA in the respiratory epithelium [20 , 21] . The observed defects in the ciliary structure resulted in a loss of ciliary motility and beating in the ventricles of the Zmynd10−/− brain ( S1 Movie and S2 Movie ) . However , ciliogenesis itself was normal , as demonstrated by the observation that the numbers of cilia and basal bodies were not different between Zmynd10+/+ and Zmynd10−/− mice ( Fig 2I and 2J ) . The subcellular distribution of the ODA intermediate chain protein DNAI2 in mouse tracheal epithelial cells ( mTECs ) cultured at the air-liquid interface ( ALI ) for 14 days was examined by immunofluorescence microscopy . In mTECs from Zmynd10+/+ mice , DNAI2 was detected in both cilia and cytoplasm and was found to colocalize with the ciliary marker acetylated α-tubulin ( Fig 2K and S5A Fig ) . In contrast , DNAI2 was only weakly expressed in the cytoplasm of mTECs from Zmynd10−/− mice and did not colocalize with acetylated α-tubulin , suggesting that DNAI2 was significantly decreased and failed to translocate to motile cilia in the absence of Zmynd10 ( Fig 2L and S5B Fig ) . Thus , the phenotypes observed in Zmynd10−/− mice resulted from a ciliary motility defect associated with the loss of axonemal ODA and IDA . The mRNA levels of DNAH5 and DNALI1—ODA heavy and IDA light-intermediate chain proteins , respectively—were downregulated by ZMYND10 knockdown in cultured human tracheal epithelial cells [20] . This was examined in Zmynd10+/+ and Zmynd10−/− mice by transcriptional profiling/RNA sequencing using total RNA extracted from the testis , lung , and brain tissues ( S1 Table ) . The transcript levels of ODA and IDA components in mutants were similar to those in wild-type mice ( Fig 3 ) , suggesting that ZMYND10 did not regulate the transcription of dynein arm components . Gene ontology analysis showed that genes involved in muscle function were significantly upregulated , whereas those involved in ion transport were reduced in Zmynd10-/- mice compared with those in Zmynd10+/+ mice ( S6 Fig ) . We and others previously reported that ZMYND10 interacts with LRRC6 [20 , 21] . Given that both proteins are cytoplasmic , we examined whether they interacted with other cytoplasmic proteins that are known to be defective in PCD by co-immunoprecipitation in human embryonic kidney ( HEK ) 293T cells . We also obtained potential interactors from a web-based protein-protein interaction database , PrePPI ( https://honiglab . c2b2 . columbia . edu/PrePPI/index_old . html ) . ZMYND10 was found to interact with C21ORF59 and DYX1C1 ( DNAAF4 ) , IQ motif and ubiquitin domain-containing protein ( IQUB ) , Tctex1 domain-containing D1 ( TCTEX1D1 ) , and DNAI1 , but not with DNAI2 ( S7 Fig ) . TCTEX1D1 is a dynein light chain of ODA [5] , and one of the intermediate chain proteins of ODA , DNAI1 , interacts with ZMYND10 . Interactions with LRRC6 and REPTIN , which are essential for cilia motility [28] , as well as interactions with C21ORF59 , DNAI1 , and IQUB were confirmed by glutathione S-transferase ( GST ) pulldown assays using lysates from mTECs ( Fig 4A ) . In addition , we found that ZMYND10 interacted with heat shock cognate protein 70 ( HSC70 ) , a constitutively expressed molecular chaperone ( S8 Fig ) , suggesting that ZMYND10 played a role in the folding and assembly of dynein arms through cooperation with HSC70 . These protein-protein interactions are illustrated in Fig 4B . REPTIN and C21ORF59 expression was also examined in mTECs at ALI 14 by immunofluorescence microscopy . REPTIN and C21ORF59 signals were reduced in Zmynd10−/− mTECs ( Fig 4C–4E and S9 Fig ) . Interestingly , C21ORF59 was recently shown to interact with LRRC6 and Dishevelled ( Dvl ) and is implicated in planar cell polarity as well as the correct localization of ODAs to motile cilia [29] . Our results demonstrated that ZMYND10 interacted with cytoplasmic proteins associated with PCD and that some interaction partners were downregulated in the absence of Zmynd10 . Our results in mTECs suggested that protein levels of dynein arm components ( Fig 2K and 2L ) and their interaction partners ( Fig 4C–4E ) were altered in Zmynd10−/− mice . To investigate this in detail , immunoblotting was carried out using the testis lysates . The levels of proteins that interacted with ZMYND10 , such as C21ORF59 , IQUB , and LRRC6 , as well as dynein arm subunits , including DNAI1 , DNAI2 , and DNAH7 , a heavy chain of inner dynein arm , were downregulated in mutant mice ( Fig 5A and 5B ) . This was also confirmed by immunofluorescence analysis of the mouse tracheal tissue . DNAH5 , DNAI2 , and IQUB signals were weaker in Zmynd10−/− mice than in Zmynd10+/+ mice ( Fig 5C–5E ) . A previous study in individuals with a homozygous truncating ZMYND10 mutation showed that DNAI2 was absent from bronchial epithelial cells , whereas DNAH5 and DNALI1 remained in the cytoplasm [30] . There were no differences in the mRNA levels of dynein arm components between Zmynd10+/+ and Zmynd10−/− mice ( Fig 3 ) . These data suggested that ODA and IDA were unstable , likely due to improper assembly in the absence of Zmynd10 . We speculated whether the decrease in protein levels of ZMYND10-interacting factors resulted from dysregulated transcription in Zmynd10−/− mice . To investigate this possibility , we examined the transcript levels of these factors and of DNAAFs , which are involved in dynein arm assembly [3] , by RNA sequencing . There were no differences in the mRNA levels of these proteins between Zmynd10+/+ and Zmynd10−/− mice ( Fig 5F–5H ) . These results suggested that ZMYND10 stabilizes dynein arm components and their interaction partners at the protein level , but not at the transcript level . To determine whether ZMYND10 regulated its interaction partners and dynein arms at the protein level , we investigated the effects of ZMYND10 on the stability of LRRC6 , DNAI1 , and DNAI2 in a heterologous system . HEK 293T cells overexpressing LRRC6 and/or ZMYND10 were treated with cycloheximide ( 100 μg/mL ) for up to 48 h to block protein synthesis ( Fig 6A ) . Only 7 . 8% of LRRC6 remained after 48 h of treatment . However , this value was increased to 44 . 4% when LRRC6 was co-expressed with ZMYND10 ( Fig 6B ) , suggesting that ZMYND10 prevented LRRC6 degradation . Similarly , the amount of DNAI1 protein was increased from 30 . 9% to 64 . 1% by co-expressing ZMYND10 ( Fig 6C and 6D ) , but not by co-expressing ZMYND10-p . Gln366* mutant protein , which lacked the MYND domain and was identified in a PCD patient [20] ( S10 Fig ) , indicating that the MYND domain was necessary for the stabilizing effect . In contrast , the stability of DNAI2 , which did not interact with ZMYND10 ( S7K and S7L Fig ) , was unaffected by ZMYND10 overexpression ( S11 Fig ) . Given that both DNAI1 and DNAI2 levels were downregulated in Zmynd10−/− mice , we speculated that ZMYND10 stabilized DNAI1 , which in turn stabilized DNAI2 . To evaluate this possibility , we compared the protein levels of DNAI2 upon co-expression with DNAI1 without or with ZMYND10 . DNAI2 was more stable in the presence of both DNAI1 and ZMYND10 than in the presence of DNAI1 alone ( Fig 6E and 6F , and S12 Fig ) . These results demonstrated that ZMYND10 stabilized some of its interaction partners at the protein level and modulated the pre-assembly of intermediate chains by stabilizing DNAI1 .
In this study , we generated and characterized Zmynd10−/− mice as a model for human PCD . The mice exhibited loss of ciliary motility and ODA and IDA components without disruption of ciliogenesis , thereby recapitulating the phenotypes associated with ZMYND10 mutations in humans and serving as an appropriate model to study PCD pathogenesis . The assembly of dynein arms into motile cilia is a complex process involving many regulatory factors , including DNAAFs and chaperones , that contribute to the stabilization , folding , and pre-assembly of dynein arm components [15 , 24 , 31] into multiprotein complexes that undergo intraflagellar transport into the axoneme for attachment to peripheral microtubules [31] . In Chlamydomonas , dynein arms exist as intermediate-chain/heavy-chain ( IC-HC ) , light chain , and docking complexes [12] . During ODA pre-assembly , HCs , such as DNAH5 or DNAH11 , are attached to ICs , DNAI1 ( IC1 ) , and DNAI2 ( IC2 ) . This process is facilitated by DNAAF1 , DNAAF2 , and DNAAF4 , while DNAAF3 is proposed to act during the final stages of chaperone dissociation [15] . IC-HC assembly fails in the absence of the IC subunit [12] . Biochemical analyses of Zmynd10−/− mice showed that DNAI1 and DNAI2 are downregulated , suggesting that ZMYND10 stabilizes these two proteins or mediates their assembly . Given that formation of the IC complex precedes IC-HC assembly , the reduced levels of DNAI1 and DNAI2 may account for the observed decrease in an ODA HC , DNAH5 . In this study , we demonstrated that ZMYND10 formed a cytoplasmic protein network comprised of LRRC6 , C21ORF59 , DYX1C1 , IQUB , REPTIN , and HSC70 . The interaction between LRRC6 and REPTIN is essential for cilia motility in zebrafish , although this function is independent of its known role as a transcriptional regulator [28] . Similarly , although ZMYND10 binds to REPTIN , the expression levels of various dynein arm components and interactors of ZMYND10 were not diminished in Zmynd10−/− mice , ruling out transcriptional regulation of these factors as a mechanism underlying ODA and IDA defects . We previously showed that DNAH5 and DNALI1 mRNAs were downregulated in human tracheal epithelial cells in which an shRNA targeting ZMYND10 was delivered by lentivirus . This discrepancy between in vivo mouse and cell line data may be due to artefacts resulting from lentiviral vector integration [32] . C21ORF59 interacts with LRRC6 , DNAAF1 , and Dvl to regulate polarization as well as ciliary motility [29] . DYX1C1 interacts with KTU , HSP70 , HSP90 , and T-complex chaperonin [14 , 16] , whereas REPTIN interacts with PIH1D1 [33] , which contains a PIH ( protein interacting with HSP90 ) domain implicated in the pre-assembly of dynein arms [34] . PIH1D3 interacts with KTU , DYX1C1 , and HSP90 [23 , 24] and is implicated in the formation of the IC complex , as evidenced by its interaction with DNAI2 and the downregulation of DNAI1 and DNAI2 in Pih1d3−/− mouse sperm [24 , 35] . ZMYND10 is functionally similar to PIH1D3 in that both proteins interact with DYX1C1 and heat shock proteins; moreover , ICs are reduced in mice lacking Zmynd10 or Pih1d3 [35] . ZMYND10 interacts with HSC70 , a member of the HSP70 family . HSC70 is involved in diverse cellular processes , including protein folding and protein degradation , and exerts its chaperone activity by cooperation with cochaperones and by binding to nascent or unfolded polypeptides through the substrate binding domain in an ATP-dependent manner [36] . Therefore , it is possible that ZYMND10 affects the stability of DNAI1 through cooperation with HSC70 . Currently , there is no curative therapy for PCD . For PCD resulting from a defective dynein arm component , the component should be replaced with a normal one . However , this will be challenging considering the huge size of some dynein components . For example , the coding region of DNAH5 , which is most frequently mutated in PCD [17] , is about 15 . 6 kb , encoding a protein of 529 kDa . In this regard , PCD resulting from defects in DNAAFs or other cytoplasmic proteins is different in that dynein arm components are not compromised , but their cytoplasmic assembly or trafficking is defective . In this study , we demonstrated that the protein levels of DNAI1 and DNAI2 were reduced in Zmynd10−/− mice due to the decreased stability of DNAI1 in the absence of ZYMND10 . Therefore , increasing protein stability of DNAI1 can be considered as a potential treatment . In conclusion , our results demonstrated that several cytoplasmic proteins , including ZMYND10 , formed a protein network in motile ciliated cells that , in conjunction with chaperone proteins , modulated various aspects of dynein arm pre-assembly . ZMYND10 specifically functioned in the early steps of this process by regulating DNAI1 stability or folding , thereby controlling IC assembly ( S13 Fig ) . These findings provide insights into the molecular mechanisms involved in dynein arm assembly and the pathogenic basis for PCD-associated defects . This will also help to develop pharmacological interventions for PCD caused by defects in the cytoplasmic nonaxonemal components of motile cilia .
The animal protocol was reviewed and approved by the Institutional Animal Care and Use Committee of University of Michigan ( #08619 ) , Boston Children's Hospital ( #13-01-2283 ) , and Yonsei University College of Medicine ( #2015–0178 ) . All animals were handled in accordance with the Guidelines for the Care and Use of Laboratory Animals . Targeted Zmynd10tm1 ( KOMP ) Wtsi embryonic stem cells were obtained from the Knockout Mouse Project Repository and injected into blastocysts . Chimeric mice were bred with C57BL/6J mice to establish germline transmission . Wild-type littermates were used as controls for Zmynd10−/− mice . Genotyping was performed by standard PCR using the primers Zmynd10-ex2F ( 5′-TGGAGGAGCTTGGAACTGAC-3′ ) , Zmynd10-ex2R ( 5′-GGAGGCAGACACAGTTAGGC-3′ ) , and CSD-RAF5-F ( 5′-ACACCTCCCCCTGAACCTGAAA-3′ , SR1 ( 5′-TGCTTTATTGTGCGAAAGGAAGAGGG-3′ ) . P1 or P28 mice were sacrificed , and the testes and lungs were dissected . After three washes with phosphate-buffered saline ( PBS ) , the tissues were fixed in 4% paraformaldehyde ( PFA ) /0 . 02% Nonidet ( N ) P-40 for 2 h at room temperature and permeabilized with 0 . 02% NP-40 in PBS for 1 h . Samples were incubated overnight at 37°C in X-gal staining solution composed of 5 mM K3Fe ( CN ) 6 , 5 mM K4Fe ( CN ) 6 , 2 mM MgCl2 , 0 . 01% sodium deoxycholate , 0 . 02% NP-40 , and 1 mg/mL X-gal in PBS . They were then washed three times with PBS for 5 min each and post-fixed with 4% PFA for 24 h before embedding within paraffin . Sections ( 10 μm thick ) were deparaffinized and rehydrated through a graded series of ethanol concentrations followed by counterstaining with Nuclear Fast Red ( Vector Laboratories , Burlingame , CA , USA ) . The lung and snout tissue specimens were fixed using 10% formalin for 24 h . The tissues were sectioned ( 5 μm thickness ) and stained with hematoxylin and eosin , or periodic acid-Schiff for histological examination . The tracheas of Zmynd10+/+ and Zmynd10−/− mice at P14 were dissected and fixed using 2 . 5% glutaraldehyde , 1 . 25% PFA , and 0 . 03% picric acid in 0 . 1 M sodium cacodylate buffer ( pH 7 . 4 ) overnight at 4°C . Samples were then processed for TEM analysis using standard techniques . P16 mice were deeply anesthetized and then decapitated . The brain was rapidly removed and immersed in ice-cold Dulbecco’s modified Eagle’s medium ( DMEM; Invitrogen , Carlsbad , CA , USA ) supplemented with 10% fetal bovine serum ( FBS; Sigma-Aldrich , St . Louis , MO , USA ) . Sagittal sections of 150-μm thickness were cut using a vibratome ( VT1200S; Leica , Wetzlar , Germany ) . Sections from the third ventricle were visualized on an Axio Observer A1 microscope using a 63× phase contrast objective lens ( LD Plan-Neofluor 0 . 75 Corr Ph2 M27; Carl Zeiss , Jena , Germany ) equipped with a high-speed charge-coupled device camera ( optiMOS sCMOS; QImaging , Surrey , BC , Canada ) . Movies were acquired at 100 frames/s . A polyclonal antibody recognizing the C-terminal sequence ( amino acids 339–362 , DRLERENKGKWQAIAKHQLQHVFS ) of mouse ZMYND10 was recovered from rabbits injected with the corresponding antigen ( AbFrontier , Seoul , Korea ) . LRRC6 and DNAH5 antibodies were previously described [37 , 38] . Antibodies against DNAI2 ( H00064446-M01; Abnova , Taipei , Taiwan ) ; REPTIN ( ab89942; Abcam , Cambridge , UK ) ; DNAH7 ( NBP1-93613 ) and DNAI1 ( SAB4501181; both from Novus Biologicals , Littleton , CO , USA ) ; IQUB ( HPA020621 ) and TCTEX1D1 ( HPA028420; both from Sigma-Aldrich ) ; acetylated α-tubulin ( T7451 from Sigma-Aldrich and 5335S from Cell Signaling Technology , Danvers , MA , USA ) ; FLAG ( #8146 ) and Myc ( #2276; both from Cell Signaling Technology ) ; and C21ORF59 ( sc-365792 ) and β-actin ( sc-1615; both from Santa Cruz Biotechnology ) were purchased from commercial sources . Secondary antibodies were purchased from Invitrogen and Santa Cruz Biotechnology for immunofluorescence and immunoblotting analyses , respectively . mTECs grown on inserts were fixed using 4% PFA for 10 min and permeabilized with 0 . 1% Triton X-100 for 20 min at room temperature . The tracheal tissue was fixed in 4% paraformaldehyde overnight at 4°C , embedded in a paraffin blocks , and cut into 5-μm-thick sections . The sections were then mounted on slides , deparaffinized , and rehydrated through a graded series of ethanol concentrations . After rehydration , antigen retrieval was performed by boiling sections for 30 min using a Retrieve-All Antigen unmasking system 1 ( pH 8; BioLegend , San Diego , CA , USA ) . Sections were permeabilized with 1% sodium dodecyl sulfate for 10 min at room temperature . mTECs and trachea samples incubated in blocking buffer containing 10% donkey serum and 1% bovine serum albumin for 1 h at room temperature . Samples were incubated overnight at 4°C with primary antibodies diluted in blocking buffer . After washes with PBS , samples were incubated with secondary antibodies and 4′ , 6-diamidino-2-phenylindole for 30 min at room temperature , washed , and covered with mounting medium and cover slips . Images were acquired using an SP5X laser scanning microscope ( Leica ) or LSM 700 microscope ( Carl Zeiss ) . mTECs were isolated from Zmynd10+/+ and Zmynd10−/− mice at P14 as previously described [39] . Briefly , cells were isolated by overnight digestion with pronase ( Roche Diagnostics , Indianapolis , IN , USA ) at 4°C and then separated from contaminating fibroblasts by incubation in mTEC basal medium on a Primaria cell culture plate ( Corning Inc . , Corning , NY , USA ) for 3–4 h . mTECs were seeded in collagen-coated apical chambers of transwell permeable supports ( 0 . 4-μm polyester membrane; Corning Inc . ) . Proliferation medium was applied to the apical and basal chambers of the wells , and cells were cultured at 37°C in 5% CO2 [40] . For ALI culture conditions , the medium was removed from the apical chamber when mTECs became confluent , and differentiation medium was added to the basal chamber . Total RNA was isolated from the brain ( P14 ) , lung ( P14 ) , and testis ( P21 ) tissues obtained from Zmynd10+/+ and Zmynd10−/− mice using a Qiagen RNA extraction kit ( Qiagen , Valencia , CA , USA ) . RNA sequencing was performed by Theragen Etex ( Suwon , Korea ) . Libraries were constructed with a TruSeq RNA Library Sample Prep kit ( Illumina , San Diego , CA , USA ) , and the enriched library was sequenced on an Illumina HiSeq 2500 system . Sequence reads were mapped against the mouse reference genome ( NCBI GRCm38/mm10 ) and analyzed using CLC Genomics Workbench v . 9 . 0 . 1 software ( CLC Bio , Cambridge , MA , USA ) . HEK 293T cells were maintained in DMEM supplemented with 10% FBS and penicillin ( 50 IU/mL ) /streptomycin ( 50 μg/mL ) . The cells were transfected with plasmids using Lipofectamine PLUS reagent ( Invitrogen ) . Experiments were performed as previously described [41] . Immunoblotting was quantified by densitometry using ImageJ software ( National Institutes of Health , Bethesda , MD , USA ) . Immunoprecipitation was performed using EZview Red anti-FLAG M2 or anti-c-Myc affinity gels ( Sigma-Aldrich ) . Pulldown assays with GST-ZMYND10 and GST-MYND were performed as previously described [20] . Cycloheximide chase was used to assess the stability of LRRC6 , DNAI1 , and DNAI2 . HEK293T cells were transfected with Myc-tagged LRRC6 , DNAI1 , or DNAI2 with or without FLAG-tagged ZMYND10; at 24 h post-transfection , cells were treated with 100 μg/mL cycloheximide ( C4859; Sigma-Aldrich ) to inhibit new protein synthesis . Cells were harvested at predetermined time points , and LRRC6 , DNAI1 , and DNAI2 levels were detected by western blotting . Results are presented as means ± standard errors or standard deviations for the indicated number of experiments . Statistical analysis of continuous data was performed with two-tailed Student’s t-test or one-way analysis of variance , with Dunnet’s , Bonferroni , or Dunn post hoc test , as appropriate . Results with P values of less than 0 . 05 were considered statistically significant . | Dynein arm defects are linked to primary ciliary dyskinesia ( PCD ) . ZMYND10 increased the stability of its interacting proteins and specifically regulated intermediate chain protein assembly , revealing tightly regulated mechanisms underlying dynein arm assembly and PCD-related pathogenesis . Increasing protein stability could be useful for developing PCD therapeutics . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"molecular",
"probe",
"techniques",
"immunoblotting",
"dyneins",
"molecular",
"motors",
"respiratory",
"system",
"immunoprecipitation",
"co-immunoprecipitation",
"molecular",
"biology",
"techniques",
"trachea",
"immunologic",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"motor",
"proteins",
"research",
"and",
"analysis",
"methods",
"proteins",
"immunoassays",
"molecular",
"biology",
"immunofluorescence",
"precipitation",
"techniques",
"cytoplasm",
"biochemistry",
"cytoskeletal",
"proteins",
"cell",
"biology",
"anatomy",
"cilia",
"biology",
"and",
"life",
"sciences",
"microtubule",
"motors"
] | 2018 | ZMYND10 stabilizes intermediate chain proteins in the cytoplasmic pre-assembly of dynein arms |
The ubiquitin-proteolytic system controls the stability of proteins in space and time . In this study , using a temperature-sensitive mutant allele of the cul-2 gene , we show that CRL2LRR-1 ( CUL-2 RING E3 ubiquitin-ligase and the Leucine Rich Repeat 1 substrate recognition subunit ) acts at multiple levels to control germline development . CRL2LRR-1 promotes germ cell proliferation by counteracting the DNA replication ATL-1 checkpoint pathway . CRL2LRR-1 also participates in the mitotic proliferation/meiotic entry decision , presumably controlling the stability of meiotic promoting factors in the mitotic zone of the germline . Finally , CRL2LRR-1 inhibits the first steps of meiotic prophase by targeting in mitotic germ cells degradation of the HORMA domain-containing protein HTP-3 , required for loading synaptonemal complex components onto meiotic chromosomes . Given its widespread evolutionary conservation , CUL-2 may similarly regulate germline development in other organisms as well .
The ubiquitin-proteolytic system has emerged as a central mechanism to regulate protein turnover spatially and temporally [1] , [2] . In this system , ubiquitin , a small polypeptide of 76 amino acids , is covalently linked to a target protein through an enzymatic cascade , and the assembly of a poly-ubiquitin chain typically specifies that target protein for rapid degradation via the 26S proteasome [3] . The system can rapidly turn “off” regulatory proteins with high selectivity and is essential for numerous cellular processes such as transcription , signaling , DNA replication , DNA repair , cell cycle progression and differentiation . Not surprisingly , defects in the ubiquitin-proteolytic system have been implicated in a number of human diseases including cancers and neurodegenerative disorders [4]–[6] . An enzymatic cascade of three enzymes mediates the attachment of ubiquitin to substrate protein: the ubiquitin-activating enzyme ( E1 ) , ubiquitin-conjugating enzyme ( E2 ) , and ubiquitin-ligase ( E3 ) [7] . Repeated cycles of ligation to the initial ubiquitin lead to poly-ubiquitination . The assembly of poly-ubiquitin chains can occur at different lysine residues within ubiquitin , with conjugation at lysines 11 and 48 typically leading to proteasomal degradation [8]–[11]–[12] . The paramount regulatory step in the cascade is the selective recognition of substrates , which is achieved by the E3-Ligase . Many E3 enzymes are nucleated around cullin scaffold subunits , with at least five cullin family members conserved in all metazoans [13] , [14] . Each cullin scaffold nucleates multiple E3-ligase complexes that all contain a similar catalytic core but use different substrate recognition modules to engage their cognate substrate ( s ) [15] , [16] . Over the past several years , Cullin RING E3-ligases nucleated around the cullins 1 , 3 and 4 have been investigated extensively . By comparison , the functions of CRL2 complexes are relatively less well understood , with the noticeable exception of the CRL2Von Hippel Lindau ( VHL ) complex that regulates the hypoxic response ( for review see [17] ) . A Cul2 knockout mouse has not been reported and the roles of Cul2 during vertebrate development remain largely unknown . In Caenorhabditis elegans , CUL-2 is highly expressed in the germline and in early embryos , where it regulates numerous developmental processes including germ cell proliferation [18]–[20] , sex determination [20] , progression through oocyte meiosis [21] , [22] , cell polarity [21] , cell fate determination [23] and cell cycle progression [19] . CUL-2 accomplishes these various functions by recruiting , via the adaptor protein elongin C ( ELC-1 ) , distinct substrate-recognition subunits ( SRS ) that specifically engage their substrates ( Figure 1A ) [17] . Several SRS have been identified in C . elegans , though in most cases their relevant targets remain to be identified . For instance , the Leucine Rich Repeat protein LRR-1 is an SRS expressed throughout the germline where it is essential for germ cell proliferation [19] , but the precise function of the CRL2LRR-1 E3-Ligase in the germline is still poorly understood . We report here the isolation of a temperature-sensitive mutation in the C . elegans cul-2 gene . The cul-2 ( or209ts ) mutation recapitulates all the cul-2 ( RNAi ) and cul-2 ( 0 ) phenotypes and also reveals several novel functions in the C . elegans germline . More specifically , we show that besides promoting germ cell proliferation , CUL-2 has at least two other functions in the germline: it participates in the proliferation versus meiotic entry decision and inhibits the first steps of meiotic prophase . Furthermore , we show that the CRL2LRR-1 E3-ligase inhibits meiotic prophase progression at least in part by promoting degradation of HTP-3 , a HORMA domain containing protein . HTP-3 is required for progression through meiotic prophase and for the assembly of the synaptonemal complex , the zipper-like structure that tethers homologous chromosomes together .
The evolutionarily conserved CRL2LRR-1 E3-ligase is essential for germline development in C . elegans: lrr-1 ( 0 ) and cul-2 ( 0 ) null mutants are sterile with a small germline [18] , [19] . However , the precise function of this enzyme is still poorly understood as the sterility in null mutants limits analysis of gene requirements . To circumvent this problem , we screened for temperature-sensitive ( ts ) mutants that resemble lrr-1 ( 0 ) mutants . While lrr-1 ( 0 ) animals are sterile , this phenotype is suppressed by inactivation of the DNA replication ATR/ATL-1 checkpoint pathway [19] . Therefore , we screened for ts mutants that , like lrr-1 ( 0 ) mutants , are sterile at the restrictive temperature of 25°C but recover fertility upon atl-1 depletion by RNAi ( Figure 1B ) . Such mutants might be specifically defective in the function of the CRL2LRR-1 complex and allow for a more careful analysis of gene requirements in the germline at different temperatures that only partially compromise E3-ligase function . After screening a collection of ts mutants ( Figure S1 , Table S1 ) we found one that fulfilled the above criteria: or209ts . Like lrr-1 ( 0 ) mutants , most or209ts mutant animals are sterile at 25°C but recover fertility upon atl-1 inactivation by RNAi ( Figure 1C ) . or209ts has previously been linked to chromosome III [24] , and we used single nucleotide polymorphisms to further map the or209ts mutation to the end of this chromosome , which contains the cullin gene cul-2 ( 21 . 36 cM ) ( Figure S1 ) . We next found that or209ts fails to complement a known cul-2 null allele ( material and methods ) , and we then sequenced the cul-2 gene from or209ts genomic DNA and found a single nucleotide change relative to wild-type in the splicing donor site at the fifth exon-intron boundary ( Figure 1D ) . This mutation substantially reduces CUL-2 protein levels even at the permissive temperature ( 15°C ) , but this effect is aggravated at the restrictive temperature ( 25°C ) ( Figure 1E ) . Consistent with the partially conditional effect on protein levels , cul-2 ( or209ts ) animals present high levels of embryonic lethality ( 30% ) and a severe reduction in brood size even at the permissive temperature of 15°C , compared to wild-type ( Figure S1 ) . Further analysis revealed that cul-2 ( or209ts ) animals present typical cul-2 loss-of-function phenotypes , including the accumulation of abnormally high levels of the TRA-1 protein , presumably causing feminization of the germline ( Figure 1F ) , and of the PIE-1 protein ( Figure S2 ) , both well-defined targets of CRL2FEM-1 and CRL2ZIF-1 E3-ligases , respectively [23] , [25] . We conclude that or209ts is a conditional allele of cul-2 . To our knowledge , or209ts is the first temperature-sensitive allele yet identified that affects a metazoan cullin gene . Using the or209ts allele , we re-examined CUL-2 function in the germline . The C . elegans germline is spatially organized and contains - from the distal to proximal end -mitotically proliferating stem cells , meiotic germ cells , and gametes ( sperm or oocyte ) . The mitotic zone extends 18–20 germ cell diameters ( gcd ) along the gonadal axis from the distal end , with the total number of germ cell nuclei exceeding 200 [26] ( Figure 2A ) . In the transition zone ( TZ ) , just proximal to the mitotic zone , cells enter into meiotic prophase [27] , [28] . At this stage , structural components accumulate on meiotic chromosomes ( e . g HIM-3 [29] ) such that germ cell nuclei present a characteristic crescent shape that is readily evident by DAPI staining [30] . Corroborating our previous observations indicating that loss of the CRL2LRR-1 E3-ligase results in a cell cycle arrest [19] , the number of germ cells in the mitotic zone was severely reduced and nuclei were enlarged in cul-2 ( or209ts ) mutants raised from early larval stages ( L1 ) at the restrictive temperature of 25°C . As expected , this phenotype was largely suppressed by inactivation of the ATL-1/DNA replication pathway ( Figure 2 ) . In addition , we noticed that the size of the mitotic zone was reduced in cul-2 ( or209ts ) animals , as determined by scoring the distance , in gcd , between the distal end of the germline and the appearance of both nuclear crescents and HIM-3 positive cells ( Figure 2B , arrows and 2C–D ) . The mitotic zone was reduced to 15 gcd in cul-2 ( or209ts ) mutants raised to adulthood at permissive temperature ( 15°C ) , and even more reduced in mutants raised at the restrictive temperature ( 25°C ) . However , it was difficult to rigorously quantify the phenotype at 25°C because germ cell nuclei were both enlarged and reduced in number ( Figure 2 ) . We thus analyzed the size of this region in atl-1 ( RNAi ) ; cul-2 ( or209ts ) mutant hermaphrodites and also observed a smaller mitotic zone in these animals . The size of the mitotic zone was similarly reduced in atl-1 ( RNAi ) ; lrr-1 ( 0 ) animals but was unaffected in the control atl-1 ( RNAi ) ( Figure 2B and 2D ) . Finally , ZYG-11 is another substrate-recognition subunit of a CRL2 complex that appears dispensable in germline stem cells [31] , and accordingly , the size of the mitotic zone was unaffected in zyg-11 ( b2ts ) mutant germlines at 25°C ( Figure 2B and 2D ) . Collectively , these results indicate that CUL-2 , in combination with LRR-1 , promotes germ cell proliferation by counteracting the ATL-1/DNA replication checkpoint but may also influence the size of the mitotic zone independently of ATL-1 . Indeed , atl-1 depletion in lrr-1 and cul-2 mutants does not fully rescue the size of the mitotic zone and atl-1 depletion in WT animals does not affect the size of the mitotic zone . We thus investigated whether cul-2 , in addition to regulating mitotic proliferation , might also have a role in preventing meiotic entry , by looking for genetic interactions between cul-2 and genes regulating the proliferation versus meiotic entry decision . GLP-1/Notch signaling controls the decision between self-renewal and entry into the meiotic cell cycle [28] . Downstream of Notch , the nearly identical Puf ( Pumilio and FBF ) -domain RNA-binding proteins FBF-1/2 prevent meiotic entry by repressing the translation of meiotic promoting factors and structural components of meiotic chromosomes ( Figure 2A ) [32] , [33] . We therefore constructed double mutants between cul-2 ( or209ts ) and glp1 ( bn18 ) , a temperature-sensitive glp-1 mutant in which the Notch signaling pathway is partially defective [34] , and between cul-2 ( or209ts ) and fbf-1 ( 0 ) or fbf-2 ( 0 ) , and scored meiotic entry in the double versus single mutants at the semi-permissive temperature of 20°C . At this temperature , germ cell proliferation is only modestly affected in cul-2 ( or209ts ) mutants ( Figure 3A ) . To score meiotic entry in these different mutant backgrounds , we used a combination of cellular morphology ( nuclear crescents ) and molecular markers . More specifically , we monitored the appearance of HIM-3 and SUN-1 Ser8-Pi ( P-SUN-1 ) on meiotic chromosomes [35] . P-SUN-1 is first detected in the mitotic zone , at nuclear periphery , in germ cells that are in mitosis from prometaphase onward ( Figure 3A , yellow arrows ) , and then in meiosis , in the transition zone ( TZ ) at foci and patches , as well as over the nuclear envelope , as reported previously [35] . As shown in Figure 3A , the cul-2ts mutant enhanced the premature meiotic entry phenotype of the glp-1 ( bn18 ) mutant , as evidenced by the premature meiotic entry in the glp-1 ( bn18 ) cul-2 ( or209ts ) double mutants , compared to the single mutants . Likewise , when combined with cul-2 ( or209ts ) , both fbf-1 and fbf-2 mutants showed a smaller mitotic zone than single mutants ( Figure 3B ) indicating that cul-2 influences the size of the mitotic zone . Collectively , these results indicate that CUL-2 acts with the Notch signaling pathway and FBF-1/2 to prevent meiotic entry .
Temperature-sensitive ( ts ) alleles have been instrumental for discovering the function of essential genes in C . elegans , in particular for genes like CUL-2 with roles in multiple processes . Indeed , ts alleles present numerous advantages: they often only partially reduce gene function even at the fully restrictive temperature , and they allow for modulation of activity through analysis at multiple temperatures . CUL-2 is an essential gene in C . elegans that is highly expressed in the germline and in the early embryo where it has been implicated in numerous processes . Our phenotypic analysis revealed that the cul-2 ( or209ts ) mutant recapitulates all the cul-2 loss of function phenotypes including feminization of the germline ( Figure 1F ) , defects in cell fate determination with PIE-1 accumulation in somatic blastomeres ( Figure S2 ) , and defects in cell cycle progression in the early embryo [24] ( data not shown ) . The penetrance of each phenotype varies with the duration of the shift at restrictive temperature , and thus this cul-2 ( or209ts ) mutation provides a useful sensitized genetic background for identifying new in vivo functions of CRL2 complexes and , most importantly , for identifying its targets . Finally , other temperature-sensitive mutants affecting the ubiquitin-proteolytic system have been identified in C . elegans [48]–[50] , but cul-2 ( or209ts ) is to our knowledge the first conditional mutation reported for a metazoan cullin . Using the cul-2 ( or209ts ) allele , we confirmed our previous observations indicating that loss of the CRL2LRR-1 E3-ligase causes hyperactivation of the ATL-1/DNA replication checkpoint in germ cells , resulting in a cell cycle arrest and adult sterility [19] . Germ cell nuclei were enlarged and reduced in number in the cul-2 ( or209ts ) mutant at 25°C , a phenotype that was partially suppressed by RNAi-mediated inactivation of atl-1 ( Figure 2 ) . Furthermore , reducing mus-101 and atl-1 function by RNAi increased the hatching rate of the cul-2 ( or209ts ) mutant ( Figure 5 ) , further demonstrating that loss of CRL2LRR-1 function triggers activation of the ATL-1/DNA replication checkpoint pathway . Why is the ATL-1 checkpoint pathway hyperactivated in cul-2 ( or209ts ) and lrr-1 ( 0 ) mutants ? We believe that one function of the CRL2LRR-1 complex is to regulate DNA replication integrity , both in germ cells and in early embryos [19] . The ATL-1 pathway is thus activated primarily in response to DNA replication defects in the lrr-1 ( 0 ) [19] and cul-2 ( or209ts ) mutants ( this study ) . In addition , our observations suggest that the inappropriate accumulation of structural components of meiotic chromosomes in lrr-1 ( 0 ) mutants may also contribute to the activation of the ATL-1 checkpoint pathway ( see below ) . Besides promoting germ cell proliferation , our results suggest that CUL-2 influences the balance between stem cell self-renewal and meiotic differentiation possibly by regulating the stability of meiotic promoting factors . In the C . elegans germline , GLP-1/Notch signaling controls the decision between self-renewal and entry into the meiotic cell cycle [28] . Downstream of Notch , the RNA-binding proteins FBF-1/2 prevent meiotic entry by repressing the translation of meiotic promoting factors , and , of structural components of meiotic chromosomes [32] , [33] , [51] . In addition to this regulatory network , which is largely translational in nature , there is increasing evidence that post-translational regulations play also a critical role in regulating the proliferation versus meiotic entry decision . For instance , the Cyclin E/Cdk2 kinase acts with the Notch pathway to promote the proliferative fate and to prevent meiotic entry [47] , [52] , and our results indicate that CUL-2 plays also a role in this process . Indeed , the cul-2ts mutant enhanced the premature meiotic entry phenotype of the glp-1 ( bn18 ) mutant and when combined with the cul-2ts mutant , both fbf-1 and fbf-2 mutants showed a smaller mitotic zone than single mutants ( Figure 3 ) . This function of CUL-2 in influencing the balance between germ cell self-renewal and meiotic entry is likely independent of the DNA replication checkpoint pathway given that atl-1 depletion does not affect the size of the mitotic zone ( Figure 2 ) . These observations suggest the existence of a complex network of post-transcriptional and post-translational regulatory mechanisms to regulate the balance between germ cell self-renewal and meiotic entry . What is the role of CUL-2 in this network ? CUL-2 might act independently at multiple levels , for instance by regulating the activity of the CYE-1/CDK-2 kinase , and by regulating the stability both of meiotic promoting factors and of structural components of meiotic chromosomes in the mitotic zone of the germline . A gradient of high/low CYE-1/CDK-2 kinase is established along the distal to proximal end of the germline , and this gradient appears important for the self-renewal versus meiotic entry decision [47] , [52] . Three complementary mechanisms establish this gradient: in the meiotic zone i ) GLD-1 inhibits CYE-1 translation [53] , ii ) CYE-1 subunit is targeted for degradation by an E3-ligase nucleated around CUL-1 [47] and iii ) the cyclin-dependent kinase inhibitor CKI-2 accumulates and inhibits CYE-1/CDK-2 activity [53] . In the mitotic zone , CKI-2 translation is inhibited by Notch and FBF-1/2 . CKI-2 is not essential for germ cell proliferation but plays a role in the maintenance of the germline by influencing the proliferation versus meiotic entry decision [54] . CUL-2 has been implicated in the regulation of CKIs stability [18] and thus CUL-2 might target CKI-2 for degradation in the mitotic zone of the germline to maintain high CYE-1/CDK-2 activity in this region . Alternatively , CUL-2 might act with CYE-1/CDK-2 to control the stability of meiotic promoting factors in the mitotic zone of the germline . Consistent with this hypothesis , substrate phosphorylation often is a pre-requisite for recognition by an E3 ligase [15] , and it has been shown recently that CYE-1/CDK-2 phosphorylates GLD-1 and thereby regulates its stability in the mitotic zone of the germline [52] , suggesting that CUL-2 might regulate GLD-1 degradation in germline stem cells . However , we failed to detect significant GLD-1 accumulation in germ cells located in the most distal part of the germline upon inactivation of cul-2 ( data not shown ) , suggesting that CUL-2 might not be involved in GLD-1 degradation . Nevertheless , we cannot exclude the possibility that a small fraction of GLD-1 accumulates upon inactivation of cul-2 and thereby contributes to the cul-2 phenotype . Although the mechanisms by which CUL-2 influences the proliferation versus meiotic entry decision remain to be identified , our results indicate that CRL2LRR-1 negatively regulates progression through meiotic prophase by controlling the stability of HTP-3 ( Figure 7 ) . HTP-3 accumulates in germ cells upon inactivation of cul-2 , lrr-1 and the proteasome , and HTP-3 physically interacts with LRR-1 in vitro . Furthermore , reducing htp-3 function , using the htp-3 ( vc75 ) allele or with RNAi , suppressed lrr-1 ( tm3543 ) sterility and increased the hatching rate of the cul-2 ( or209ts ) mutant , respectively . In addition , reducing syp-1 function also increased the hatching rate of the cul-2 ( or209ts ) mutant . These results suggest that premature accumulation of HTP-3 on chromosomes allows for inappropriate HIM-3 and subsequently SYP-1 binding . This inappropriate assembly of SC components on chromosomes likely contributes to DNA replication checkpoint activation . However , although SC components appear to be expressed and loaded inappropriately on chromosomes upon inactivation of the CRL2LRR-1 E3-Ligase , germ cells do not appear to enter into meiosis because we failed to detect SUN-1 Ser8-Pi on chromosomes of ectopic proliferating germ cells upon inactivation of cul-2 or lrr-1 . This observation indicates that HTP-3 accumulation is not sufficient to drive progression through meiotic prophase and that CUL-2 acts independently of HTP-3 to influence to proliferation versus meiotic entry decision , presumably by controlling the stability of meiotic regulator ( s ) in the mitotic zone of the germline ( Figure 7 ) . In conclusion , our results indicate that CUL-2 acts at multiple levels in the germline to coordinate germ cell proliferation and meiotic entry and thus emerged as a critical regulator of germline development in C . elegans .
C . elegans strains were cultured and maintained using standard procedures [55] . Strains of the following genotypes were used: N2 Bristol ( wild type ) ; lrr-1 ( tm3543 ) II/mIn1[mIs14 dpy-10 ( e128 ) ]II [19]; cul-2 ( or209ts ) III [24]; htp-3 ( vc75 ) I [42]; htp-3 ( vc75 ) I; lrr-1 ( tm3543 ) II/mIn1[mIs14 dpy-10 ( e128 ) ]II ( this study ) ; glp-1 ( bn18 ) [34]; atl-1 ( tm853 ) IV [56]; gld-3 ( q730 ) nos-3 ( q650 ) II/mIn1[mIs14 dpy-10 ( e128 ) ]II [57]; gld-3 ( q730 ) nos-3 ( q650 ) II/mIn1[mIs14 dpy-10 ( e128 ) ]II; cul-2 ( or209ts ) III ( this study ) , fbf-1 ( ok91 ) II [32] , glp-1 ( bn18 ) cul-2 ( or209ts ) III ( this study ) ; fbf-1 ( ok91 ) II; cul-2 ( or209ts ) III ( this study ) ; fbf-2 ( q738 ) II [58] , fbf-2 ( q738 ) II; cul-2 ( or209ts ) III ( this study ) . For this study , or209 worms were mated to the Hawaiian CB4856 C . elegans wild isolate , and recombinant F2 lines homozygous for the or209 mutation were isolated based on embryonic lethality at the restrictive temperature . Several SNPs located between −27 . 2 cM and 21 . 25 cM along the third chromosome were amplified by PCR from these recombinants , essentially as described [59] , and genotyped using pyrosequencing technology . Briefly , PCR amplifications were performed from single worms using Taq DNA polymerase ( New England Biolabs ) and specific primers for each SNP . The purification of single-stranded PCR amplicons and the pyrosequencing reactions were subsequently performed according to manufacturer's instructions using a Pyromark Q96 ID instrument ( Biotage ) . This analysis positioned the or209 mutation between +18 . 52 cM and the right end of chromosome III . This region contains cul-2 , and we showed that all embryos from or209/cul-2 ( ek1 ) trans-heterozygotes ( ek1 is a null allele of cul-2 [18] ) failed to hatch . Escherichia coli clones expressing dsRNA to deplete C . elegans genes were obtained from the MRC Geneservice ( Cambridge , U ) [60] , [61] . RNAi feeding was performed as described using 2 mM IPTG ( RNAi plates ) [62] . The results are presented as means ± S . E . M . In all graphs , data were compared by a Mann-Whitney test ( two-tailed p ) or Student test ( Figure 3 and Figure 5 ) . All calculations were performed with InStat3 software ( Graphpad ) . * p<0 . 05; ** p<0 . 01; *** p<0 . 001 . L4 animals were fed on NGM plates containing 10 mM final HU prior to analysis . Germlines were then dissected and stained with DAPI . Quantification of HU-induced cell cycle arrest was performed by counting the number of nuclei in a defined volume ( 20 000 µm3 ) . Germlines were dissected in PBS followed by freeze-crack , immersion in cold MeOH ( −20°C ) for 1 min , and fixation in 1× PBS , 0 . 08 M HEPES ( pH 6 . 9 ) , 1 . 6 mM MgSO4 , 0 . 8 mM EGTA and 3 . 7% paraformaldehyde for 30 min in a humidity chamber at room temperature . Slides were washed 3×5 min , blocked for 1 h in PBT ( 1× PBS , 0 . 1% Triton X-100 , and 5% BSA ) , and incubated overnight at 4°C with primary antibodies diluted in PBT . Working dilutions for the primary antibodies were 1∶1000 for rabbit anti-HIM-3 ( M . Zetka ) , 1∶100 for rabbit anti-HTP-3 ( M . Zetka ) and 1∶1000 anti-SUN-1 Ser8-Pi ( V . Jantsch ) . Slides were later incubated for 30 min at room temperature with secondary antibodies coupled to the Alexa 488 and 568 fluorophore ( 1∶600 , Molecular Probes ) . Next , germlines were mounted in Vectashield Mounting Medium with DAPI ( Vector ) . Fixed germlines were imaged using either a TCS SP5 confocal microscope ( Leica ) or a LSM 710 confocal microscope ( Zeiss ) with 40× objectives . Confocal images correspond to the projection of confocal Z-stacks spanning maximum 5 µm . Captured images were processed using ImageJ and Adobe Photoshop . For whole-worm images , worms were immobilised with 20 mM levamisole and mounted on 2% agarose pads . Images were then acquired using an Axiovert 200 inverted microscope equipped with DIC optics . Standard procedures were used for SDS–PAGE and western blotting . The following antibodies were used in this study: primary antibodies were directed against CUL-2 [19] , ELC-1 [19] , HTP-3 ( M . Zetka ) , HIM-3 ( M . Zetka ) , SUN-1 Ser8-Pi [35] , Tubulin ( Sigma ) , Actin ( Sigma ) , TRA-1 [63] and MUS-101 [36]; secondary antibodies conjugated to peroxidase against rabbit or mouse were purchased from Sigma . LRR-1/ELC-1/ELB-1 trimeric complexes were expressed in E . coli and purified as described [19] . Total worm extracts were prepared by cryolysis , as previously described [64] , and loaded onto trimeric complexes immobilised on T7-agarose beads ( Novagen ) for 2 hours at 4°C . After five washes , proteins were eluted with sample buffer and separated by SDS-PAGE . | Maintenance of the germline depends on the presence of a germline stem cell pool with self-renewal potential that produces gametes upon meiotic differentiation . Factors regulating the balance between germline stem cell self-renewal and meiotic differentiation ensure germline homeostasis , whereas disruption of these regulatory mechanisms can lead to sterility or cancer . In this study , we show that the ubiquitin-proteolytic system ( UPS ) , which selectively targets regulatory proteins for proteasomal degradation , controls germline development by acting at three different levels . The UPS promotes germ cell proliferation , regulates the balance between self-renewal and meiotic differentiation , and limits progression through meiotic prophase . In particular , we show that an E3 ubiquitin-Ligase nucleated around the cullin 2 ( CUL-2 ) protein and using the Leucine rich repeat protein LRR-1 as substrate recognition subunit regulates in germ stem cells the stability of HTP-3 , which is required for progression through meiotic prophase . These findings identify a previously unknown role for proteolytic regulation in germline development and also explain how the critical balance between germ cell proliferation and meiotic differentiation can be tightly and robustly controlled with multiple , parallel regulatory inputs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"germ",
"cells",
"meiosis",
"molecular",
"cell",
"biology",
"cell",
"biology",
"chromosome",
"biology",
"cellular",
"types",
"biology",
"genomics",
"genetics",
"and",
"genomics"
] | 2013 | CRL2LRR-1 E3-Ligase Regulates Proliferation and Progression through Meiosis in the Caenorhabditis elegans Germline |
Post-transcriptional regulation of gene expression plays a crucial role in many bacterial pathways . In particular , the translation of mRNA can be regulated by trans-acting , small , non-coding RNAs ( sRNAs ) or mRNA-binding proteins , each of which has been successfully treated theoretically using two-component models . An important system that includes a combination of these modes of post-transcriptional regulation is the Colicin E2 system . DNA damage , by triggering the SOS response , leads to the heterogeneous expression of the Colicin E2 operon including the cea gene encoding the toxin colicin E2 , and the cel gene that codes for the induction of cell lysis and release of colicin . Although previous studies have uncovered the system’s basic regulatory interactions , its dynamical behavior is still unknown . Here , we develop a simple , yet comprehensive , mathematical model of the colicin E2 regulatory network , and study its dynamics . Its post-transcriptional regulation can be reduced to three hierarchically ordered components: the mRNA including the cel gene , the mRNA-binding protein CsrA , and an effective sRNA that regulates CsrA . We demonstrate that the stationary state of this system exhibits a pronounced threshold in the abundance of free mRNA . As post-transcriptional regulation is known to be noisy , we performed a detailed stochastic analysis , and found fluctuations to be largest at production rates close to the threshold . The magnitude of fluctuations can be tuned by the rate of production of the sRNA . To study the dynamics in response to an SOS signal , we incorporated the LexA-RecA SOS response network into our model . We found that CsrA regulation filtered out short-lived activation peaks and caused a delay in lysis gene expression for prolonged SOS signals , which is also seen in experiments . Moreover , we showed that a stochastic SOS signal creates a broad lysis time distribution . Our model thus theoretically describes Colicin E2 expression dynamics in detail and reveals the importance of the specific regulatory components for the timing of toxin release .
Regulation of gene expression occurs at transcriptional and post-transcriptional levels , and has been studied intensively both experimentally and theoretically [1–10] . Bacterial stress responses , such as the well-studied production and release of the toxin colicin E2 in Escherichia coli , represent one setting in which post-transcriptional control is crucial [11–15] . Colicins are toxic proteins produced by certain E . coli strains in response to stress as a means to kill bacteria that compete with them for the same resources . More specificly , colicin E2 is a bacteriocin , which damages the DNA of bacterial cells that absorb it ( a DNAse ) . Once synthesized , colicin E2 forms a complex with an immunity protein , thus protecting its producer from its otherwise lethal action [14 , 16 , 17] . The toxin is released only upon cell lysis , which is triggered by the synthesis of a dedicated lysis protein [15 , 18–20] . As this inevitably entails the death of the producer cell [19] , it is vital for the persistence of the population that only a fraction of its members actually releases the toxin [14] . The genes for the colicin , immunity protein and lysis protein are organized into the colicin E2 operon , which is depicted in Fig 1 , together with the interaction network that controls colicin E2 expression and release . Each of the three components is encoded by a single gene—the colicin by cea , the colicin-specific immunity protein by cei , and the lysis factor by cel—and three regulatory regions control their transcription: an SOS promoter upstream of the cea gene [21] , and two transcriptional terminators T1 and T2 , located upstream and downstream of the cel gene , respectively [22] . The key transcriptional regulator of the SOS operon is the LexA protein ( reviewed in [23] ) , marked in orange in Fig 1 . LexA dimers repress the SOS promoter region of the ColE2 operon , but also block the transcription of over 30 other SOS genes [24 , 25] , many of which play an important role in DNA repair [26] . In the event of DNA damage , the LexA dimer undergoes auto-cleavage upon interaction with RecA [27] , and the transcription of SOS genes begins . The presence of the two transcriptional terminators in the ColE2 operon results in the production of two different mRNAs: A shorter transcript ( short mRNA , marked purple in Fig 1 ) that encompasses only the genes for the toxin colicin E2 and the immunity protein , and a longer transcript ( long mRNA , marked green in Fig 1 ) , which additionally includes the lysis gene [14 , 28 , 29–32] . Hence , lysis can only be initiated after the translation of long mRNA [18] , and this crucial operation is regulated at the post-transcriptional level , as described below . Post-transcriptional regulation makes use of many different mechanisms . Recent studies emphasize the particular importance of non-coding sRNAs [33] for various processes in E . coli , especially because of their ability to introduce delays and set up thresholds for translation [34–37] . This is done either directly , by sRNAs pairing with their target mRNA ( sRNA-mRNA interaction ) , or indirectly , by sequestering of specific mRNA-binding proteins ( mRNA-protein interaction ) [2 , 38 , 39] . For the latter form of regulation , recent studies highlighted the importance of the production rates of regulatory components [40] . In the case of the ColE2 system , the translation of the long mRNA is regulated by the carbon storage regulator protein CsrA [28] , marked red in Fig 1 . CsrA dimers destabilize target mRNAs by binding to a region that includes the ribosome-binding site ( Shine-Dalgarno sequence ) [41] . Masking of the ribosome-binding site by CsrA thus not only represses translation of the lysis gene but also promotes degradation of the long mRNA . However , CsrA is also recognized by two specific sRNAs , CsrB and CsrC [42] , marked blue in Fig 1 . These sRNAs can therefore sequester CsrA dimers , preventing them from binding to target mRNAs [43–45] . Thus , translation of the ColE2 lysis gene is indirectly regulated by sequestration of CsrA . This process , also known as “molecular titration” , exhibits ultrasensitive thresholds and has been extensively studied [46 , 47] . The basic interaction network that controls the ColE2 regulatory network has been studied in great detail in previous works [48–51] , and many of its functional characteristics , in particular the threshold behavior , were described for a wide range of both bacterial and eukaryotic systems [52] . However , a detailed theoretical description of the dynamics leading to the release of colicin is still missing , in particular the role of the hierarchically ordered regulation involving CsrB and CsrC . In this work , we have formulated this post-transcriptional network in a detailed mathematical model , constructed by analogy to studies of simpler sRNA-regulated systems ( for example , [33 , 34 , 36] ) . We then simplified the model by assuming fast complex equilibration , and combining the sRNAs CsrB and CsrC into a single , effective sRNA ( see S1 Text for details ) . This reduced the regulation network to three relevant components: free long mRNA , free CsrA and the effective sRNA ( see Fig 2 ) . We then analyzed this simplified network in detail . In contrast to previous work [36] , we give a general analytical solution for the three component system , and derive a precise approximation for fast and clear analysis . This analytic solution exhibits a pronounced threshold in mRNA production due to CsrA-dependent regulation , which was also confirmed using numeric simulations . We investigated , how this threshold depends on system parameters , and how it affects the actual biological system . Furthermore , we have analyzed the role of fluctuations in the post-transcriptional regulation network and how fluctuations in long mRNA expression may be dampened by sRNA . Finally , we extended our model by including the transcriptional regulation , and analyzed how the system behaves during a realistic SOS response . Previous studies have shown discrete activation peaks in LexA-repressed promoters [26] that can lead to large fluctuations close to the threshold of mRNA expression [9] . In a stochastic simulation of the complete model , we were able to reproduce this phenomenon . Comparison with experimental data on lysis time distributions [48] also shows that our model can explain the delayed and broadly distributed release times of colicin complexes . This underlines the importance of stochasticity for the heterogeneous expression of colicin E2 in E . coli populations .
For our theoretical analysis , we initially developed a detailed mathematical model for the post-transcriptional regulation of colicin E2 release . To this end , we derived a set of coupled , deterministic rate equations from the interaction network depicted in Fig 1 , with the corresponding rates for transcription , degradation , binding interactions etc . as parameters . In the following , we briefly review how we reduced the network to its core components , which comprise the theoretical model . The interaction scheme underlying the complete model is presented in S1 Fig and further explanations can be found in the Supporting Information , where we also detail how our model can account for sequestration by other targets of the global regulator CsrA . As we wished to study the post-transcriptional regulation of colicin E2 expression , we included in the model only those components that are relevant at that stage . The model therefore omits the short mRNA and its products . However , the rate of transcription of the long mRNA is a crucial parameter , which is influenced by the kinetics of activation of the SOS promoter , and thus by the processing of its repressor LexA . Upon DNA damage , RecA promotes auto-cleavage of LexA dimers , thus removing inhibition of the SOS response ( marked in red in Fig 1 ) . The LexA-RecA interaction network has recently been modeled stochastically [53] . Before including this detailed network in our final model , we focused on understanding the post-transcriptional dynamics . To this end , we initially assumed that activation of the SOS promoter occurs rapidly relative to the rates of production and degradation of the long mRNA [54] , which allowed us to approximate the transcription rate of long mRNA by an effective rate αM ( Materials and Methods ) . With respect to post-transcriptionally relevant components , we were then left with long mRNA , CsrA , and the two sRNAs CsrB and CsrC , and the mRNA-CsrA- , CsrA-CsrB- , and CsrA-CsrC-complexes . CsrB and CsrC regulate CsrA by forming complexes with it . The two sRNAs each have several ( on average: N ) CsrA binding sites , and if every occupation state of the sRNAs were to be modeled as a separate component , a large number of coupled rate equations would need to be added to the model . However , due to the fast dynamics of the CsrA-CsrB- and CsrA-CsrC-complexes , and their virtually identical biochemical behavior , we were able to reduce the sRNA interaction to a single equation for an effective sRNA , with only one binding site and transcription rate NαS ( see Materials and Methods ) . As a result , the mechanisms of complex formation , dissociation and degradation are replaced by an effective coupled degradation of complex partners . Despite the different processes that are integrated to effective ones , the effective sRNA still resembles the dynamical behavior of CsrB/CsrC . A detailed derivation of the simplified system of rate equations can be found in S1 Text . The final post-transcriptional model is thus reduced to a set of three coupled , deterministic rate equations that capture the behavior of the free long mRNA ( M ) , free CsrA dimers ( A ) , and an effective free sRNA ( S ) component with a single CsrA binding site: M ˙ = α M - δ M M - k M M A , ( 1 ) A ˙ = α A - δ A A - k M p M M A - k S p S A S , ( 2 ) S ˙ = N α S - δ S S - A k S S , ( 3 ) where ( 1 − pM ) and ( 1 − pS ) are the probabilities for CsrA to survive the coupled degradation . A graphical illustration of this differential equation system is depicted in Fig 2 . Note that in the model the quantities M , A and S represent the abundance of the corresponding free components . Once a long mRNA , sRNA , or CsrA dimer binds to some other component , it loses its function and is thus removed from the model system . For the analysis of our model , we had to determine production , degradation and binding rates . The particular values used are listed in S1 Table . As far as possible , we chose values that are measured in studies on either the same or comparable systems ( see S1 Text for details ) . In the other cases , we tried to derive plausible parameters from known factors that influence the particular rate . A detailed motivation and derivation of these rates is given in chapter 2 of S1 Text . We analyzed the reduced post-transcriptional model by first calculating its steady state . In order to obtain a cleaner and simpler result , we derived an approximation ( see Materials and Methods ) for the steady state solution , which agreed very well with the results of numerical simulations ( see S2 Fig ) . Using these simplified equations , we then investigated the impact of the rates of production of long mRNA ( αM ) and sRNA ( αS ) on the levels of the three components . The results ( see Fig 3 ) reveal a linear threshold that appears at the same position for all three components . The threshold divides the parameter space into two regimes , in which either CsrA or long mRNA and sRNA have a non-zero abundance . This is due to the coupling between the degradation of CsrA and the abundance of both long mRNA and sRNA , such that the presence of CsrA dimers excludes that of long mRNA and sRNA , and vice versa . This mechanism in turn controls the release of colicin-immunity complexes , since a sufficiency of CsrA dimers ensures reliable repression of the long mRNA and prevents synthesis of the lysis protein . From the aforementioned analytic solution we calculated the threshold position as a function of the system parameters ( S1 Text ) . We found that the threshold for non-zero levels of long mRNA lies exactly at the point where the production rate of CsrA αA is equal to the sum of transcription rates for long mRNA αM and sRNA αS ( S1 Text ) . Thus , we observed no expression of long mRNA in the regime αM + αS < αA , as shown in Figs 3A and 4 . We find the threshold to be sharp , and attribute this to the very slow degradation of CsrA compared to long mRNA and sRNA [55 , 56] . Apart from the threshold itself , we find that the levels of free CsrA and free sRNA predicted by our steady state analysis are consistent with experimental in-vivo values determined by previous studies [43 , 57] . Moreover , our results are also consistent with the total amount of CsrA as well as its ratio to sRNA ( S1 Text ) . So far , we have demonstrated that our three-component system is capable of producing a threshold behavior . However , it has been shown previously that a mutually exclusive production of sRNA and a target mRNA is possible with just two components [36] . The question thus arises why a third component is needed at all . One possible explanation is that the sRNA makes it easier to trigger lysis , as an increase in sRNA production induces an increase in the abundance of long mRNA ( Fig 3 ) . After SOS signals , the sRNA controls and accelerates the degradation of CsrA ( see section on expression dynamics below ) , eventually leading to the expression of the lysis protein . In a next step , we analyzed the stochastic dynamics of the post-transcriptional regulation network . To this end , we switched to a stochastic description , calculated the Fano factor ( VarM/〈M〉 ) for the abundance of long mRNA ( see Materials and Methods ) , and depicted it as heatmap in Fig 4 . The Fano factor measures the relative magnitude of fluctuations , and has already been applied to gene regulatory networks in previous studies [58] . It can also be understood as a quantified comparison with the pure birth process ( Poisson process ) , which has the Fano factor F = 1 . We found that fluctuations in mRNA were most pronounced close to the threshold position , with the largest fluctuations occurring slightly above the threshold ( Fig 4 ) . Moreover , Fig 4 also shows that the fluctuations became larger as sRNA production decreases . Thus , the third component ( sRNA ) in the post-transcriptional regulation network also enables significant dampening of fluctuations in long mRNA . To understand why the fluctuations are localized to the region near threshold , one must take the characteristics of this parameter regime into account . Around the threshold , molecule numbers are close to zero , which has a direct affect on the relative size of fluctuations: the lower the abundance , the larger the fluctuations ( stochastic regime ) . Moreover , the threshold is the only regime in which all three components , CsrA , mRNA and sRNA , can coexist and interact with each other: An increase in the level of CsrA will lead to a decrease in the abundance of long mRNA and sRNA , owing to increased complex formation and subsequent degradation . Analogously , an increase in long mRNA and sRNA molecule numbers leads to a decrease in CsrA abundance . Therefore , the abundance of CsrA dimers is anti-correlated with the abundance of both long mRNA and sRNA . It has been shown for a two-component system , that anti-correlated components can create anomalously large fluctuations [59] if degradation rates are small compared to turnover ( ratio of production rate to abundance ) . For long mRNA , this is exactly the case close to threshold , where the long mRNA abundance is still very low . These results show that a third component can reduce intrinsic fluctuations of a hierarchically ordered regulatory network . To study the dynamical response of the ColE2 system to an SOS signal , we extended the post-transcriptional network by including the LexA-RecA regulatory network [53] ( Fig 1 ) . LexA not only represses the SOS promoter , it is also an auto-repressor , as well as being a repressor of RecA production . As outlined in the Introduction , RecA forms filaments after DNA damage , which then induce auto-cleavage of LexA dimers . Consequently , the levels of RecA , LexA and the colicin mRNAs increase , as repression due to LexA is relaxed . A stochastic model of this network has been introduced recently [53] . In that study , promoter activity in the LexA-RecA system was found to occur in ordered bursts that result from fluctuations and the particular structure of the RecA-LexA feedback loop . In our analysis of the ColE2 post-transcriptional regulation network so far ( see above ) , we have assumed the dynamics of SOS promoter activation to be so fast that we could use an effective transcription rate αM for long mRNA . To link the LexA regulatory network to the post-transcriptional regulation network , we must drop this assumption and explicitly model the dynamics of LexA dimers , which connect the two networks . In the biological system , this involves the binding and dissociation of LexA dimers to and from the SOS promoter in the ColE2 operon . Long mRNA and short mRNA are transcribed only from the derepressed promoter at rates αMl and αMs , respectively . Thus , the transcription rates of long mRNA and short mRNA are proportional to the number of open SOS promoters in the bacterium . The majority of transcripts are short mRNAs . The mathematical implementation of the integrated regulation network is again a system of coupled rate equations , which we describe in S1 Text . The additional parameters of the LexA-RecA regulation network are to be found in S2 Table . We simulated the SOS signal by temporarily up-regulating the coupling parameter cp , which quantifies the ability of RecA to induce cleavage of LexA ( Fig 1 ) . In the uninduced state before and after the SOS signal , the auto-cleavage parameter was set to cp = 0 . Under SOS stress cp was increased to cp = 6 . This increase in cp subsequently boosts the long mRNA production , and therefore relates to a transition from a sub-threshold state ( gray area below the white line in Fig 3A ) to a super-threshold state ( green area above the white line in Fig 3A ) . Due to the stochasticity in the LexA-RecA network and the resulting stochastic promoter dynamics , the overall transcription rate αMl of long mRNA is not constant , but fluctuates about a mean value . The production rate of sRNA was held constant at αS = 57 . 5 . Fig 5 shows the dynamics of short and long mRNA levels and the abundance of CsrA dimers and sRNA in response to transient SOS signaling . When we compared a stochastic realization using Gillespie simulations ( Materials and Methods ) with a numerical solution of the deterministic rate-equation system , we observed significant qualitative and quantitative differences . First , the stochastic realization exhibited significant fluctuations that manifested themselves in abrupt , short-lived changes in the abundance of short mRNA over the whole time-course ( Fig 5A ) . Second , the average over 500 stochastic realizations deviated from the deterministically predicted value . Both phenomena arise from the intrinsic stochasticity of the LexA-RecA-regulatory network , as explained by Shimoni [53] . Fluctuations may lead to a spontaneous dip in the number of LexA dimers which releases all LexA-regulated genes , including the lexA gene itself , from repression . This consequently leads to a sudden rise in the abundance of short mRNA . The open lexA and recA promoters will then generate a burst of newly produced LexA and RecA proteins , which block and regulate the promoters for the next burst . Focusing on the dynamics of mRNA transcription , we found that , due to initial simulation parameters , only small numbers of the short mRNA are produced in the uninduced state . After up-regulation of the LexA auto-cleavage parameter cp at t = 200 min , the abundance of short mRNA rises and the aforementioned large bursts appear . The amount of long mRNA , however , follows a completely different trajectory , conditioned by post-transcriptional regulation . Before the SOS signal , expression of long mRNA is almost completely repressed by CsrA ( Fig 5B ) . Even the bursts of SOS promoter activity reflected in fluctuating amounts of the short mRNA have little or no impact on the long mRNA . This filtering effect is biologically relevant , as it ensures that noisy promoter activity does not erroneously trigger lysis . After induction of the SOS signal , the deterministic dynamics of the underlying rate equations predicted that , after a delay of about 40 min , the abundance of long mRNA should rapidly rise to a saturation value ( black dashed line in Fig 5B ) . However , a mean of 500 realizations deviated significantly from this prediction ( Fig 5B ) . In particular , the average number of long mRNA molecules increased more slowly than predicted by deterministic dynamics . Hence the abundance saturated at a much lower value . An appreciable delay between SOS signal induction and expression of long mRNA was still observed , but lasted for only 15 min . Studying the dynamics of a single stochastic realization , we observed that the number of long mRNA molecules underwent large fluctuations , which were followed by periods of no expression at all . Moreover , the timing of these bursts varied considerably between different realizations . This constitutes a significant qualitative difference compared to the average over 500 realizations and to the deterministic dynamics ( Fig 5 ) , both of which exhibit a smooth and continuous temporal behavior . Fig 5B and 5C indicates the origin of this behavior: The abundance of long mRNA can only grow if the number of free CsrA dimers is low . The same holds for the abundance of sRNA , which supports the degradation of CsrA and also can only reach non-zero abundances if there is no CsrA left ( Fig 5D ) . Thus , before any long mRNA can be expressed , the free CsrA concentration must drop to very low values due to degradation or complex formation . The delay between SOS signal induction and the first burst of long mRNA synthesis therefore depends on the amount of CsrA available . We went on to study the precise timing of the first burst in long mRNA abundance , since it is crucial for the time-point of release of colicin-immunity complexes . To this end , we calculated the probability distribution for the first peak from an ensemble of 500 stochastic realizations . The probability of a peak in long mRNA abundance rose quickly and reached its maximum approximately 60 min after induction of the SOS signal ( Fig 6A ) . This phenomenon is also seen in experimental systems: time-lapse studies with colicin-producing bacteria revealed that their lysis time is broadly distributed [48] . The distribution depicted in Fig 6A matches qualitatively with comparable datasets from these experiments . Moreover , our model is able to numerically predict average lysis times in dependence on different SOS signal strengths ( see S5 Fig ) . From the probability distribution of the timing of the initial peak in long mRNA abundance we calculated the survival function , i . e . the probability with respect to time that a cell will not release toxin . Here we assumed that this first burst provides enough long mRNA in the cell to produce the lysis protein , which then induces its lysis with concomitant release of colicin-immunity complexes into the surrounding medium . The function of lysed cells plotted in Fig 6B shows that the number of cells that release the toxin rises with the duration of the SOS signal . Incorporation of the LexA-RecA regulatory network allowed us to model the colicin E2 expression dynamics in response to a realistic SOS signal , and the results presented above highlight the importance of CsrA for colicin release .
Gene expression is a process that allows for various forms of regulation at all levels . In theoretical studies of post-transcriptional regulation of several biological systems , modulation of mRNA production by proteins or sRNA has been shown to create , for instance , temporal thresholds for mRNA translation [9 , 35 , 36] . Proteins have also been shown to regulate the expression of the toxin colicin E2 [28] in the context of an SOS response to environmental stress . Experimental studies have elucidated the detailed interaction network responsible for the production and release of the colicin [28] . However , the dynamics of this system , in particular at the post-transcriptional level , have remained elusive . In close analogy to previous two-component models , we developed a mathematical model for this hierarchically ordered post-transcriptional regulation of colicin E2 release . Interestingly , the known interaction network for this system necessitated the modeling of three , not two , components: the long mRNA which is necessary for colicin release , its negative regulator CsrA , and sRNA , which in turn negatively regulates CsrA . Contrary to previous studies [9 , 35 , 36 , 60] , the sRNAs do not regulate the mRNA directly , but control the level of the regulator protein CsrA . Thus , the sRNA acts as the “regulator’s regulator” . In our analysis of the model , we used rate constants that were determined from experimental systems ( see chapter 2 of S1 Text for details ) . Comparing the predicted CsrA levels before the SOS signal ( see Fig 5C ) with in-vivo measurements of E . coli [57] shows that our model results in a pre-SOS free CsrA abundance that agrees with actual bacterial systems ( for other abundances , see S1 Text ) . Moreover , the model is not just able to predict steady state abundances , but also reproduces the reaction to varying external stress levels as seen in experiments ( see S5 Fig ) . Investigation of the dynamics revealed that the model exhibits a time delay in the production of free long mRNAs . This delay is due to the high abundance of CsrA in the non-SOS state of the cell , which causes CsrA to quickly bind to free long mRNA and thus prevents its transcription . Only during an SOS signal , which indicates external stress for the cell , the level of CsrA gets steadily reduced . The time this process takes to get CsrA levels so low that fluctuations in long mRNA production result in free long mRNA , causes a delay in colicin release . As colicin release is coupled to cell lysis , the delay is therefore a mechanism for filtering out transient SOS signals that might erroneously lead to synthesis of the lysis protein . Moreover , also intrinsic fluctuations , for instance in sRNA production , are filtered out by this mechanism: Even if a large and sudden burst in sRNA were strong enough to drop CsrA abundance close to zero , the CsrA buffer gets restored quickly due to the large production rate of CsrA . This rate is only effectively lowered during a SOS signal , which increases the production of the CsrA-sequestering long mRNA . The fact that lysis is regulated by a threshold mechanism of a global regulator protein like CsrA might also be a guarding mechanism for the cell: only prolongued extreme situations will cause the abundance of these regulators to drop to low molecule numbers . However , delays and similar threshold behavior also emerge in two-component systems , raising the question why a third component is necessary here . Strikingly , we found that the third component ( sRNA ) in the post-transcriptional interaction network enables the cell to tune the duration of the delay by sequestering CsrA . In the case of the ColE2 system , this means that cells are able to adjust the ( average ) time between a SOS signal and the onset of cell lysis leading to colicin release . Furthermore , previous studies of systems with slow , bursting promoter kinetics have also uncovered a major limitation of two-component sRNA-based regulation compared to regulation based on transcription factors: Two-component systems are subject to significantly higher levels of intrinsic noise [9] . However , Fig 4 ( panels A , C , D ) shows that , in the post-transcriptional regulation of colicin E2 release , fluctuations become smaller at higher values of αS . The sRNA might therefore allow for significant dampening of these fluctuations . This idea is supported by the fact that the relatively high degradation rate of sRNA makes it less susceptible to induced fluctuations . In bacteria , these mechanisms could have several functions: First , a comparison of different sRNA production rates ( S4 Fig ) indicates that the sequestration of CsrA by the sRNA could indeed be crucial for fast release of the colicin , as CsrA degradation rates cannot be arbitrarily increased in bacterial systems . Second , they can tune the reaction to external stress at the population level . Experimental studies have shown that , in the absence of stress , 3% of colicin producing cells release the toxin during the stationary phase; but this fraction can be increased up to eventually 100% if an external SOS stress is applied [14 , 48] . Previous experimental studies also found that colicin systems exhibit heterogenous expression times , which originate from the stochasticity of the SOS signal [49 , 50] . Recent time-lapse experiments with colicin E2 producing bacteria showed that this lysis time distribution also depends on the strength of the SOS signal [48] . We reproduced these experiments with stochastic simulations , in which we created different stress levels by different values of the RecA degradation rate parameter cp . Our predictions for lysis time distributions ( Fig 6A and S5 Fig ) show qualitative agreement with these time-lapse experiments . Moreover , the ability of the sRNA to tune the average duration of the delay might serve as a mechanism to adjust the cell lysis to different stress levels . Altering the sRNA level could be an additional mechanism , apart from the stochastic SOS signal , by which bacterial populations can adjust the fraction of cells releasing the toxin depending on the strength and duration of the external stress . Finally , the co-option of sRNA makes the cells less susceptible to lysis due to adventitious fluctuations in promoter activity . This is particularly important considering the bursting behavior and large-scale fluctuations seen in the LexA-RecA-regulatory system , which are readily observed in experiments and reproduced by stochastic models [53] . In order to focus on the interplay between the LexA-RecA system and the hierarchical regulation of long mRNA by CsrA and sRNA , we kept the plasmid number constant . If we considered random , Poisson-distributed plasmid numbers instead , the effect would be very small , as shown in S4B Fig . This fact demonstrates that the colicin plasmid copy number only has minor influence on the lysis time distribution ( see S1 Text for details ) . In conclusion , we have provided here the first detailed theoretical description of colicin E2 production and release , and used it to study the dynamical behavior of this system . Moreover , the general three-component model described here should be applicable to many other systems of toxin production in microorganisms .
In most models of prokaryotic gene expression , it is assumed that promoter kinetics are fast compared to RNA production and degradation rates . In that case , the promoter state is well approximated by its steady state [54] . In the analysis of the post-transcriptional regulation network , the promoter status affects the transcription rate of the ( long ) mRNA . Thus , we replaced it by an effective transcription rate for ( long ) mRNA , which takes into account the probability of a gene being blocked . In the literature this procedure is referred to as “adiabatic elimination of fast variables” ( see for example [61] ) . For this effective rate we also took into account that the colicin operon is located on a plasmid [62] , of which approximately 20 copies exist in each cell [14] ( see S1 Text ) . The two sRNAs CsrB and CsrC regulate CsrA via complex formation . More specifically , each CsrB molecule has approximately 22 binding sites for CsrA , with 9 CsrA dimers being attached on average [63 , 64] . CsrC interacts in the same way , but has fewer CsrA binding sites [63] . As a first step , we therefore replaced the two sRNA types by a single effective one , which has N binding sites . However , all of the N + 1 sRNA configurations still enter the interaction network as separate components , since the binding and dissociation probabilities change with the number of free binding sites . By investigating the dynamics of the CsrA-sRNA complexes , we discovered that the probability distribution for occupied CsrA binding sites on the sRNAs reaches its stationary state on a time scale that is proportional to the rate of complex- ( un ) binding . Since binding and unbinding events are biochemically much simpler processes than transcription , translation or degradation , it is very likely that the dynamics of CsrA-sRNA complexes is much faster than all other reaction rates in the system . Following the line of Levine [36] and Legewie [34] , we therefore assumed rapid complex dynamics , and replaced the different binding site occupations by an effective sRNA , with only one binding site and transcription rate NαS ( see S1 Text for details on the calculation ) . For the calculations of the abundances of the three components ( for example , to obtain the plots of Fig 3 ) , we began by assuming the stationary state . Solving for the abundance of one component then gives a cubic equation , for which the exact , general solution is very lengthy and cumbersome to analyze . Therefore , we considered the cubic equation for the cases of very large and very small molecule numbers , and ignored terms that became negligible . This resulted in two easily solvable quadratic equations . Comparisons with numerical solutions of the cubic equations proved that the quadratic solutions approximate the general solution well in their respective abundance regime . Equating the terms omited in the approximation yields a criterion for the transition between the two approximations ( see S1 Text ) . The transition is depicted as a white line in Fig 3 . That this transition lies close to the threshold is coincidental . Comparison with the exact , numerical solutions showed that the threshold is not an approximation artifact . S2 Fig illustrates the precision of the approximation by comparing its prediction for long mRNA abundance to that from numerical simulations . We started the analysis of noise properties by reformulating the simplified three-component system as a Master equation . As Master equations are typically impossible to solve analytically , we performed a general van Kampen expansion in multiple variables ( components ) . Our analysis included all higher orders , and not only lowest order terms as is commonly found in textbooks [61 , 65] . With van Kampen’s expansion we were able to derive general formulas for the first up to the fourth moment of the random variable representing the fluctuations of the system around the stationary solution of the rate equations . The terms of each equation were classified in first order terms ( dominant terms ) and higher order terms ( second order , third order , etc ) , according to the scaling behavior of each term with the system size . We used different methods to calculate the Fano factor for long mRNA . The most reliable results were obtained by implementing only first order terms in the calculations of second moments . This reproduced the shape of the Fano factor well , but it overestimates fluctuations in the vicinity of the threshold . S3 Fig illustrates the degree of agreement between analytical calculations of the Fano factor agree with the results from Gillespie simulations . To verify how well our analytical results of the deterministic rate equations coincide with the actual mean molecule numbers , we set up a Gillespie simulation [66] . The Gillespie algorithm generates a statistically correct realization of the master equation behind the rate equations . The core of the algorithm lies in using random numbers to determine which next reaction will occur and the waiting time prior to the succeeding reaction . The reactions simulated by the Gillespie approach are listed in S1 Text . To quantify the delay between SOS signal induction and the first burst in long mRNA abundance , we defined the beginning of the first peak as the point when the number of long mRNA molecules exceeds 8 for the first time . The time of the peak itself was set to the point at which that number reached a maximum . We then calculated the probability distribution from an ensemble of 500 stochastic realizations , using the parameters defined in S1 and S2 Tables . | Gene expression is a fundamental biological process , in which living cells use genetic information to synthesize functional products like proteins . To control this process , cells make use of many different mechanisms . A well-studied example is the binding of expression intermediates by a cellular component in order to delay the synthesis . This mechanism is known to regulate the stress-induced release of the toxin colicin E2 by E . coli bacteria . However , experimental studies have shown that this system is not regulated by just one component , but the interplay of several cellular components , in which the hierarchically ordered main components interact . Here , we create a mathematical model for the interaction network of colicin E2 release , and study how the component levels evolve . We show that the system is able to delay the release of the toxin . Additional components allow to fine-tune the delay and dampen fluctuations in gene expression that would lead to premature toxin release . A comprehensive analysis of the time evolution reveals a broad distribution of toxin release times , which is also observed in experiments . This rich dynamical behavior emerges from the interplay of regulatory components , and , due to its generality , may also be transferred to similar regulatory networks , in particular toxin expression systems . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"dimers",
"(chemical",
"physics)",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"engineering",
"and",
"technology",
"gene",
"regulation",
"signal",
"processing",
"messenger",
"rna",
"signal",
"filtering",
"toxic",
"agents",
"toxicology",
"dna",
"transcription",
"physiological",
"processes",
"post-transcriptional",
"gene",
"regulation",
"tissue",
"repair",
"gene",
"expression",
"chemistry",
"physics",
"lysis",
"(medicine)",
"biochemistry",
"rna",
"nucleic",
"acids",
"physiology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"chemical",
"physics"
] | 2016 | Hierarchical Post-transcriptional Regulation of Colicin E2 Expression in Escherichia coli |
The INSIG2 rs7566605 polymorphism was identified for obesity ( BMI≥30 kg/m2 ) in one of the first genome-wide association studies , but replications were inconsistent . We collected statistics from 34 studies ( n = 74 , 345 ) , including general population ( GP ) studies , population-based studies with subjects selected for conditions related to a better health status ( ‘healthy population’ , HP ) , and obesity studies ( OB ) . We tested five hypotheses to explore potential sources of heterogeneity . The meta-analysis of 27 studies on Caucasian adults ( n = 66 , 213 ) combining the different study designs did not support overall association of the CC-genotype with obesity , yielding an odds ratio ( OR ) of 1 . 05 ( p-value = 0 . 27 ) . The I2 measure of 41% ( p-value = 0 . 015 ) indicated between-study heterogeneity . Restricting to GP studies resulted in a declined I2 measure of 11% ( p-value = 0 . 33 ) and an OR of 1 . 10 ( p-value = 0 . 015 ) . Regarding the five hypotheses , our data showed ( a ) some difference between GP and HP studies ( p-value = 0 . 012 ) and ( b ) an association in extreme comparisons ( BMI≥32 . 5 , 35 . 0 , 37 . 5 , 40 . 0 kg/m2 versus BMI<25 kg/m2 ) yielding ORs of 1 . 16 , 1 . 18 , 1 . 22 , or 1 . 27 ( p-values 0 . 001 to 0 . 003 ) , which was also underscored by significantly increased CC-genotype frequencies across BMI categories ( 10 . 4% to 12 . 5% , p-value for trend = 0 . 0002 ) . We did not find evidence for differential ORs ( c ) among studies with higher than average obesity prevalence compared to lower , ( d ) among studies with BMI assessment after the year 2000 compared to those before , or ( e ) among studies from older populations compared to younger . Analysis of non-Caucasian adults ( n = 4889 ) or children ( n = 3243 ) yielded ORs of 1 . 01 ( p-value = 0 . 94 ) or 1 . 15 ( p-value = 0 . 22 ) , respectively . There was no evidence for overall association of the rs7566605 polymorphism with obesity . Our data suggested an association with extreme degrees of obesity , and consequently heterogeneous effects from different study designs may mask an underlying association when unaccounted for . The importance of study design might be under-recognized in gene discovery and association replication so far .
One of the first genome-wide association ( GWA ) studies ever and the first on obesity identified the INSIG2 gene represented by the rs7566605 polymorphism as a novel gene for common obesity [1] . Functional evidence depicted the INSIG2 gene from the very start as an interesting candidate for obesity as being involved in the reversed cholesterol transport by an interaction with sterol regulatory element-binding proteins ( SREBPs ) [2] , which are transcription factors that activate the synthesis of cholesterol and fatty acids in the liver and other organs [3] . The observed SNP-obesity-association was replicated in some , but not in all studies [4]–[11] . A letter to Science by the authors of the initial report in response to the emerging debate of the early inconsistent results [1] raised the question of whether the association might be more pronounced in studies that were not ascertained for reasons related to better health status , when comparing more severely obese subjects with normal controls , in populations with a higher prevalence of obesity or in populations with a higher mean age . The need for a meta-analysis was stated there for the first time and re-stated by Lyon and colleagues [12] . Furthermore , a secular trend for increasing prevalence of obesity was observed in two large population-based studies from the same geographical region using the same protocols but one recruited 1994/95 ( KORA-S3 ) and the other 1999–2001 ( KORA-S4 ) [13] . The later study showed a stronger INSIG2-obesity-association compared to the earlier study: This raised the additional question whether the changes in nutritional intake and physical activity [14] , [15] believed to contribute to the increase in the prevalence of obesity during the last decades were the reason for some of the between-study heterogeneity observed for this SNP's association with obesity . The inconsistent reported associations and the many resulting debates motivated us to undertake a systematic meta-analysis of all available data to investigate potential causes of heterogeneity and to look for consistent results among subgroups . It was thus the specific aim of this meta-analysis to explore five hypotheses for heterogeneity of the rs7566605 association with obesity: ( Hypothesis 1 ) The association depends on study design . Therefore , we classified studies as general population-based ( GP ) when they were neither selected for any disease nor for not having any disease , as any selection of this type was shown to potentially induce bias for outcomes associated with the disease [16] . We classified studies as ‘healthy population’ ( HP ) when they were selected for reasons related to a better health status ( i . e . studies including subjects from working populations or studies excluding subjects for with diseases such as diagnosed type 2 diabetes ) , or obesity studies ( OB ) when they were specifically designed to investigate obesity , usually case-control or family studies . We did not include studies that were ascertained for any disease to reduce overly complexity . ( Hypothesis 2 ) The association is more pronounced when comparing more extreme cases of obesity with normal or lean subjects , or ( Hypothesis 3 ) among studies with a greater percentage of obese individuals . ( Hypothesis 4 ) The association is differentially seen in studies including subjects with a higher age compared to studies based on younger populations , or ( Hypothesis 5 ) more pronounced in studies with a more recent assessment of BMI ( after year 2000 ) assuming that these studies would reflect the changes in dietary habits and physical activity of the last decade and assuming that subjects with the INSIG2 risk genotype are more prone to gain weight in such an environment .
We have gathered data on 34 studies from across Europe , North America , and Asia that met the inclusion criteria including a total of 74 , 345 subjects . All studies were categorized a priori according to their study design ( GP , HP , OB ) , ethnicity , and whether it was an adult or children population ( Tables S1 and S2 ) . A more detailed description can be found in the original study publications [1] , . The main results are summarized in Table 1: The meta-analysis of 27 studies of Caucasian adults ( n = 66 , 213 ) showed a fixed effect odds ratio ( OR ) of 1 . 076 ( p-value = 0 . 023 ) and a random effects OR of 1 . 051 ( p-value = 0 . 268 ) with the I2 measure indicating significant between-study heterogeneity ( I2 = 41 . 0% , 95% confidence interval , CI = [6 . 6% , 62 . 8%] , Q-test p-value = 0 . 015 ) . The I2 measure declined ( I2 = 10 . 9% , 95% CI = [0 . 0% , 48 . 1%] , Q-test p-value = 0 . 329 ) by restricting to the 16 GP studies , including 48 , 844 subjects , yielding a fixed [random] effects OR of 1 . 097 [1 . 092] ( p-value = 0 . 015 [0 . 035] ) . The OR estimates were similar when excluding early published studies ( p = 0 . 054 [0 . 060] ) . The I2 measure for heterogeneity was zero among the five HP studies , including 7640 subjects , and combined estimates yielded a tendency towards a protective OR of 0 . 796 ( p-value = 0 . 028 ) without remarkable change when excluding the early published studies . The OR among the six OB studies , including 9729 subjects , of 1 . 152 using the random effect model was higher than the one for the GP studies , but not statistically significantly different from unity ( p-value = 0 . 253 ) . There was substantial heterogeneity among the OB studies ( I2 = 63 . 2% , Q-test p-value = 0 . 018 ) . The OR estimates were similar when excluding the early published studies . Figure 1A–1C shows forest plots of the Caucasian adult studies with combined estimates by study type . The p-value testing for difference between the GP and the HP combined estimates was 0 . 004 [0 . 012 when corrected for pair-wise testing of three subgroups] and 0 . 039 [0 . 089] when excluding the early published studies . Thus , there is some , but not completely conclusive evidence for Hypothesis 1 that study design might explain some of the between-study heterogeneity of this genetic association . In the pooled analysis of the four studies on non-Caucasian adults ( n = 4889 ) , we found no significant association of the CC genotype with obesity . The pooled analysis of the three pediatric studies ( n = 3243 ) was also not significant ( Table 1 , Figure 1D and 1E ) . Our data suggested an association with increased ORs among the Caucasian adults when more extremely obese subjects were compared to lean controls ( Table S3; Hypothesis 2 ) : combining over all studies , the ORs gradually increased from 1 . 156 to 1 . 183 , 1 . 221 , or 1 . 265 ( p-values ranging from 0 . 001 to 0 . 003 ) when moving the BMI cut-off for the obese cases from 32 . 5 to 35 . 0 , 37 . 5 or 40 . 0 kg/m2 , respectively , and comparing to controls with BMI<25 kg/m2 . A similar trend for increasing ORs was seen when comparing extremely obese subjects against controls with BMI<30 kg/m2 or BMI<20 . 0 kg/m2 . Among the GP studies , the ORs increased from 1 . 198 to 1 . 257 , 1 . 313 , or 1 . 414 ( p-values ranging from 0 . 001 to 0 . 003 ) , respectively . The analogous comparison for HP studies revealed that the protective influence of the CC homozygous was reversed in the more extreme comparisons with ORs from 0 . 856 to 0 . 959 , 1 . 139 , or 1 . 604 . For the OB studies , the ORs of the analogous comparisons were well above unity for all comparisons , but did not show a trend . As summarized in Table 2 , the accompanying trend analyses of the CC genotype frequencies across the various BMI categories indicated significantly increased CC genotype frequencies from 10 . 4% to 12 . 5% ( p-value testing for trend = 0 . 0002 ) , which persisted when excluding the early published studies ( p-value = 0 . 0008 ) . A similar trend was seen among GP studies . In both types of analyses , the varying cut-point ORs as well as the trend in genotype frequencies by BMI categories , suggest an association of the rs7566605 with extreme obesity compared to normal controls ( Hypothesis 2 ) . Table 3 summarizes the results of the further three hypotheses to explain heterogeneity , which were tested for the Caucasian adult studies . Hypothesis 3: there was some tendency towards higher ORs among ‘more obese’ study populations compared to the ‘less obese’ study populations ( p-value = 0 . 052 [0 . 285] testing for difference of the fixed [random] effects ORs ) , but not statistically significant . Hypothesis 4: there was no evidence for any difference between studies from older populations ( i . e . mean age of subjects above 50 years ) as compared to studies from younger populations ( i . e . mean age below 50 years ) . Hypothesis 5: there was a tendency towards more pronounced ORs for the studies with BMI assessed after the year 2000 as compared to studies with BMI assessed before 2000 ( p-value = 0 . 007 [0 . 095] ) , but not statistically significant given the various tests performed and particularly not when excluding the early published studies ( p-value = 0 . 086 [0 . 248] ) . Hypotheses 3–5 were not tested in the HP or OB stratified analyses as too few ( 3–6 ) studies were available . The sensitivity analyses ( Table S4 ) indicated robustness of estimates towards selection of gender or age and no significant difference between published and unpublished studies . Excluding the two studies with self-reported BMI from the overall GP meta-analysis resulted in a slight increase of the OR estimate . We specifically examined the association under the recessive genetic model as suggested by the original paper [1] . Our data on the raw numbers of obese or non-obese subjects with one of the three genotypes underscored a recessive model in the Caucasian adult studies combined ( ORCCversusGG = 1 . 112 [1 . 029 , 1 . 203] , p-value = 0 . 007 , and ORGCversusGG = 0 . 988 [0 . 940 , 1 . 037] , p-value = 0 . 618 ) as well as among the GP studies ( ORCCversusGG = 1 . 076 [0 . 976 , 1 . 185] , p-value = 0 . 142 , and ORGCversusGG = 0 . 995 [0 . 956 , 1 . 036] , p-value = 0 . 813 ) . The secondary analyses on BMI as a quantitative outcome were only performed in GP and HP studies . These analyses generally showed results consistent with the obesity analyses , but less , if any , significance ( Figure S1 , Tables S5 , S6 , and S7 ) .
The results of our analyses suggest an association of this SNP with extremely obese subjects compared to normal controls , but future research will need to confirm this finding . Study design can impact how many extremely obese subjects are included in the study . Study designs that sample more extremely obese subjects will have greater power to detect the association , while study designs that sample fewer of these subjects will have little power to detect the association . An association with extreme obesity might well be masked by study design , and meta-analyses which disregard study design differences . The tendency of higher OR estimates observed in the general population-based studies ( GP ) and the obesity case-control studies compared to ‘healthy population’ ( HP ) studies could possibly reflect the association for extreme obesity compared to normal controls . We have classified studies as ‘ascertained for criteria related to a better health status’ ( ‘HP’ ) when patient groups were excluded or when the sample was ascertained based on working populations , which are known to be usually more healthy . We have performed this classification blinded for the study estimate to exclude bias from informative misclassification . It could be that the common rs7566605 directly or via tagging another possibly rare and quite penetrant variant does not so much alter BMI throughout the distribution , but really puts participants into the very obese category . Thus an effect is picked up in the GP samples , but not in the HP studies with fewer extremely obese persons . This would also be in-line with ( i ) our more pronounced findings in the studies with a higher percentage of obese subjects , and ( ii ) the lack of association in the quantitative BMI-analysis , which tests for a shift in the full BMI distribution . It could also be hypothesized that the between-study heterogeneity is due to an interaction of the gene with the environment of high fat diet: INSIG2 is regulated by atherogenic diet and oxidized oil in rodents [23] , [24] and such a diet relates to higher obesity status . A gene-environment interaction was also suggested by reports that life-style interventions including physical training have less positive effects in CC genotype carriers than in CG/GG subjects [25]–[27] . A person at the brink of getting obese might either comply to exercise and avoid becoming obese or might give up and end in the extreme obesity category . This would be in-line with our pronounced findings for more extreme degrees of obesity . It might also be speculated that our more pronounced association among studies with BMI assessment after the year 2000 compared to before 2000 reflects this gene-environment interaction as well: assuming that a change in nutritional habits and physical activity contributed to the increase in obesity observed in the last decades , the studies with a more recent BMI assessment might reflect this more “modern” environment and the INSIG2-obesity association would emerge here more clearly . Also , unknown epistatic interaction of the rs7566605 with one or other ( rare ) polymorphisms could lead to association with the more extreme obesity phenotype , with the INSIG2 gene being part of a complex that functions as a biological entity ( SREBP , SCAP , INSIG2 ) . The importance of ascertainment of the study sample might be under-recognized so far . Monsees and colleagues [16] have illustrated that ascertaining for or against disease would induce a bias in genetic association estimates when the genetic marker as well as the phenotype under study ( here obesity ) are associated with the disease . As obesity is associated with many chronic diseases such as type 2 diabetes and cardiovascular disease , exclusion of such study participants opens up for bias , if association of the SNP with the disease cannot be precluded . We had specifically excluded studies ascertained for disease and had also planned on separating HP from GP studies ahead of the analyses . We would like to highlight that we have adopted a very strict definition of GP and that there might be special advantages in using either disease-ascertained studies [28] or particularly healthy samples in other instances [29] . This meta-analysis has several strengths: ( 1 ) We have conducted a systematic approach by collecting all studies published before January 1 , 2008 , including seven studies that were unpublished at that time . ( 2 ) The meta-analysis is large including a total of 74 , 365 subjects . ( 3 ) We separated working tasks , with one researcher designing the analysis plan , recruiting studies , classifying studies by study type , and deciding upon compliance to inclusion criteria , while the other cared for the incoming data and performed the analysis . Therefore , design decisions were made in a blinded way , which guarded against subtle post-hoc data-driven analysis decisions , study selection bias , and informative misclassification of study design . ( 4 ) We collected data according to a strict protocol including standardized analysis from each study partner , with strong quality control of study-specific results . ( 5 ) We performed only a limited number of pre-defined subgroup analyses with some amendment during the review process . ( 6 ) We had a strong focus on the diversity of study design , which is unique in genetic epidemiological research at the time being and an issue probably under-recognized so far . It might be considered a disadvantage that we did not include studies with subjects selected for diseases , particularly those associated with type 2 diabetes and thus a higher prevalence of obesity , as the association might be stronger in such studies . This might have been one reason for the initial investigation by Herbert and colleagues to detect this association , as mostly type 2 diabetes or asthma ascertained samples had been used . However , we excluded these samples by design in order to reduce heterogeneity and to reduce the influence of counter-regulating disease processes or medications . Furthermore , publication bias is always a threat for meta-analyses as the extent and direction of this selection cannot fully be determined; we attempted to guard against this by recruiting also unpublished studies . It might be considered a further disadvantage that our hypotheses were motivated by the early published studies , which are included in this meta-analysis; to accommodate for this fact , we repeated all analyses excluding these studies ( see Text S1D ) . Finally , it might be considered a disadvantage that we were not able to recruit enough non-Caucasian adult or children studies for a conclusive comparison with the Caucasian adult studies . This pooled analysis including all study designs does not provide evidence for overall association of the INSIG2 rs7566605 CC genotype with increased risk of obesity compared to the CG or GG genotypes . Our data suggest an association with extreme degrees of obesity and consequently heterogeneous effects from different study designs may mask an underlying association when unaccounted for . The importance of study design might be under-recognized in gene discovery and association replication so far .
We designed our meta-analysis as a pooled analysis of study-specific association estimates according to a standardized protocol ( see ‘data form’ , Text S1B , and pre-defined analysis plan , Text S1C ) with an amendment added during the review process ( Text S1D ) . Our eligibility criteria for studies were ( i ) data available on BMI , the INSIG2 rs7566605 SNP genotypes , age and sex , ( ii ) sample size of at least 200 subjects , ( iii ) ethical approval , and ( iv ) either general population-based ( GP ) , ascertained for reasons related to a better health status such as studies including only subjects in the work-force or studies excluding subjects with diseases ( ‘healthy population’ , HP ) , or designed specifically to study obesity such as obesity case-control or obesity family studies ( OB ) . We excluded all studies selecting subjects for any disease . For more information on the classification of studies by study type , see Text S1E . We did not exclude on any age or ethnicity criteria to allow exploration of potential heterogeneity . We controlled for study selection bias by separating the two main tasks between the two first authors: IMH . took care of study recruitment , compliance to inclusion criteria , and classification of studies by study type , and CH performed quality control and statistical analysis . We identified all eligible studies published before January 1 , 2008 by a systematic PubMed literature search using the search terms ‘INSIG2’ OR ‘INSIG-2’ OR ‘rs7566605’ . Additionally we identified unpublished studies through contacting researchers in the field by making a call for this meta-analysis in several consortia ( GIANT , KORA-500K , IL-6-consortium ) , in the letter to Science by Herbert and colleagues [1] , in the paper by Lyon and colleagues [12] , and in meeting presentations . We sent out and collected standardized data forms , and verified all entries for within-plausibility as well as consistency with publications , if available . We made plausibility checks by use of double information in the aggregated data . All study-specific ambiguities were clarified with the respective study investigators . All involved studies were conducted according to the principles expressed in the Declaration of Helsinki . The studies were approved by the local Review Boards . All study participants provided written informed consent for the collection of samples and subsequent analysis . For each study , OR estimates comparing the odds of obesity ( BMI≥30 kg/m2 ) for subjects with the minor-allele homozygous genotype ( CC ) with subjects of the other genotypes combined ( CG , GG ) , thus assuming a recessive model , were calculated using logistic regression adjusting for age and sex . We also collected OR estimates with standard errors ( SE ) for the odds of more extreme degrees of obesity ( i . e . subjects with BMI≥32 . 5 , 35 . 0 , 37 . 5 , or 40 . 0 kg/m2 ) compared to various degrees of leanness ( i . e . subjects with BMI<30 , 25 , or 20 kg/m2 ) . Furthermore , we collected summary statistics ( mean and SE ) on the difference in mean BMI between subjects with the CC genotype compared to subjects with the CG or GG genotypes using linear regression adjusted for age and sex . For each study , analyses stratified for sex or age ( with a cut-off at 50 years ) were performed as well . Among the six studies from non-Caucasian populations , two studies had too few ( <3 ) subjects among the obese with the CC genotype to be included into the dichotomous obesity analysis , while they were included for the quantitative BMI analysis . For the meta-analysis , we combined beta-estimates among Caucasian adult studies ( All-CA ) followed by a stratified analysis by study type ( GP , HP , OB ) and combined estimates among non-Caucasian ( All-NC ) or children studies ( All-CH ) , see ‘amendment to analysis plan’ , Text S1D . The following was only performed on Caucasian adults as the number of available non-Caucasians or children was too low . We tested for differential association between the GP , HP , or OB studies applying a t-test on the combined beta-estimates and correcting p-values for testing three subgroups . We tested for a trend in CC genotype frequencies across the different BMI categories and tested for differential associations separating the studies for higher or lower obesity prevalence , higher or lower mean age of study subjects , or for a more or less recent BMI assessment using a t-test on the combined beta-estimates . This was complemented by sensitivity analyses stratifying on sex , age , publication status , and type of BMI assessment . As the hypotheses were motivated by the early published studies mentioned in the letter to Science by Herbert and colleagues [1] , we repeated all analyses with exclusion of these hypotheses-generating studies . In all analyses , between-study heterogeneity was tested by the χ2-based Q-statistic and quantified by I2 as a measure of the proportion of variance between the study-specific estimates that is attributable to between-study difference rather than random variation . We pooled study-specific estimates according to the inverse-variance weighted fixed effect or the DerSimonian and Laird random effects model [30] . Heterogeneity was considered to be significant at the 10% level . All statistical analyses were performed with SAS ( statistical analysis software , SAS institute , Inc . ) . Forest plots were prepared using Review Manager software ( Cochrane Collaboration , Copenhagen , DK ) . | A polymorphism of the INSIG2 gene was identified as being associated with obesity in one of the first genome-wide association studies . However , this association has since then been highly debated upon inconsistent subsequent reports . We collected association information from 34 studies including a total of 74 , 000 participants . In a meta-analysis of the 27 studies including 66 , 000 Caucasian adults , we found no overall association of this polymorphism rs7566605 with obesity , comparing subjects with a body-mass-index ( BMI ) ≥30 kg/m2 with normal BMI subjects ( BMI<30 kg/m2 ) . Our data suggested an association of this polymorphism with extreme obesity ( e . g . , BMI≥37 . 5 kg/m2 ) compared to normal controls . Such an association with extreme obesity might induce heterogeneous effects from different study designs depending on the proportion of extreme obesity included by the design . However , further studies would be required to substantiate this finding . The importance of study design might be under-recognized in gene discovery and association replication so far . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"public",
"health",
"and",
"epidemiology"
] | 2009 | Meta-Analysis of the INSIG2 Association with Obesity Including 74,345 Individuals: Does Heterogeneity of Estimates Relate to Study Design? |
Mouse taste receptor cells survive from 3–24 days , necessitating their regeneration throughout adulthood . In anterior tongue , sonic hedgehog ( SHH ) , released by a subpopulation of basal taste cells , regulates transcription factors Gli2 and Gli3 in stem cells to control taste cell regeneration . Using single-cell RNA-Seq we found that Gli3 is highly expressed in Tas1r3-expressing taste receptor cells and Lgr5+ taste stem cells in posterior tongue . By PCR and immunohistochemistry we found that Gli3 was expressed in taste buds in all taste fields . Conditional knockout mice lacking Gli3 in the posterior tongue ( Gli3CKO ) had larger taste buds containing more taste cells than did control wild-type ( Gli3WT ) mice . In comparison to wild-type mice , Gli3CKO mice had more Lgr5+ and Tas1r3+ cells , but fewer type III cells . Similar changes were observed ex vivo in Gli3CKO taste organoids cultured from Lgr5+ taste stem cells . Further , the expression of several taste marker and Gli3 target genes was altered in Gli3CKO mice and/or organoids . Mirroring these changes , Gli3CKO mice had increased lick responses to sweet and umami stimuli , decreased lick responses to bitter and sour taste stimuli , and increased glossopharyngeal taste nerve responses to sweet and bitter compounds . Our results indicate that Gli3 is a suppressor of stem cell proliferation that affects the number and function of mature taste cells , especially Tas1r3+ cells , in adult posterior tongue . Our findings shed light on the role of the Shh pathway in adult taste cell regeneration and may help devise strategies for treating taste distortions from chemotherapy and aging .
In mouse tongue taste buds are found in three types of papillae: anterior fungiform ( FF ) , lateral foliate ( FO ) , and posterior circumvallate ( CV ) . The numerous FF papillae each contain a single taste bud , while the two FO and single CV papillae each contain hundreds of taste buds [1 , 2] . Each taste bud contains ~50–100 mature receptor cells classified as type I , type II , or type III cells based on morphology and markers . These cells are further classified into functional subtypes that respond to basic taste qualities of sweet , bitter , umami , sour , and salt [3–5] . Embryonic taste papillae development , especially for anterior tongue , has been well-studied [6–8] . Canonical developmental pathways such as Wnt , sonic hedgehog ( Shh ) , Notch , and fibroblast growth factor ( Fgf ) pathways drive embryonic taste papillae development [reviewed in: 2 , 6 , 9] . In adult mice , taste cells survive only 3–24 days , necessitating their regeneration throughout life [10] . However , much less is known about the regulation of adult taste cell regeneration , even though it is essential for maintaining the sense of taste throughout life . Adult taste cell regeneration is affected by aging , radiation treatment and chemotherapy , infection , and autoimmune diseases [11–18] . The role of Shh signaling in regulating taste papillae development and taste cell differentiation at the embryonic and adult stages has been well studied [19–27] . In embryos , SHH is a suppressor of taste placode formation in the FF papillae [17 , 28] while it promotes development of taste buds in CV papillae [29] . Yet , SHH overexpression in adult taste epithelium induces numerous ectopic FF taste buds but has no major effect in other taste fields [1 , 27] . SHH signals through the membrane-bound receptors PTCH1 and SMO to regulate the bi-functional transcription factors GLI2 and GLI3 , the principal effectors of the pathway in adults [30–34] . In the absence of SHH signaling , GLI2 and GLI3 are C-terminally truncated to generate transcriptional repressors that are mostly sequestered in the cytoplasm [30 , 33 , 34] . SHH signaling prevents the proteolysis of GLI2 and GLI3 and promotes their localization to the nucleus , where they regulate the expression of numerous target genes [35–37] . The role of GLI3 in adult taste cell turnover has not been investigated to date . We generated single-cell RNA-Seq data from multiple taste cell subtypes from mouse CV papillae and found that Gli3 is highly expressed in green fluorescent protein ( GFP ) marked cells positive for Lgr5 ( Lgr5-GFP marked taste stem cells ) and in Tas1r3-GFP marked type II taste cells . These results were confirmed using reverse transcription PCR ( RT-PCR ) , in situ hybridization , and immunohistochemistry . Conditional knockout of Gli3 ( Gli3CKO ) in CV and FO papillae increased taste bud size and the numbers of Tas1r3- and Lgr5-expressing taste cells relative to wild-type animals ( Gli3WT ) . Similar changes were observed in Gli3CKO taste organoids derived from Lgr5-GFP+ taste stem cells . In posterior tongue in vivo and in organoids these alterations were accompanied by changes in the expression of Tas1r3 , Trpm5 , Gnat3 , and multiple Gli3 target genes . In line with changes in taste cell number and gene expression , Gli3CKO mice showed increased lick and glossopharyngeal ( GL ) nerve responses to sweet and umami taste stimuli and decreased lick responses to bitter and sour taste stimuli . Our data indicate that Gli3 is a negative regulator of differentiation and/or survival of taste stem cells and Tas1r3+ type II taste cells that influences taste receptor cell composition and function .
To identify Gli-family transcription factors selectively expressed in subsets of adult taste cells , we analyzed single-cell RNA-Seq data generated from Lgr5-GFP+ stem , Tas1r3-GFP+ type II and Gad1-GFP+ type III taste cells isolated from respective GFP-transgenic mouse strains . The transcription factors Gli1 and Gli2 were expressed in all three types of cells , while Gli3 ( and its upstream regulators Ptch1 and Smo ) was highly expressed in both Lgr5-GFP+ and Tas1r3-GFP+ but not in Gad1-GFP+ taste cells ( S1 Table ) . RT-PCR showed that Gli3 was expressed in FF , FO and CV papillae taste tissue , as well as in lingual epithelium devoid of taste buds ( Fig 1A ) . Using GFP+ taste cells purified by fluorescence-activated cell sorting ( FACS ) Gli3 was found in Lgr5-GFP+ and Tas1r3-GFP+ but not Gad1-GFP+ taste cells ( Fig 1A ) . Quantification of Gli3 mRNA expression by quantitative PCR ( qPCR ) showed that it is expressed at high levels in Lgr5-GFP+ and Tas1r3-GFP+ , but at only low levels in Gad1-GFP+ taste cells ( Fig 1B ) . In situ hybridization using Gli3 antisense probes confirmed that it is expressed in taste cells in all three taste papillae ( Fig 1C–1E ) , while the control sense probe produced only minimal background signal in taste cells ( Fig 1F–1H ) . Indirect immunohistochemistry using an antibody generated against the N-terminus of GLI3 capable of detecting both the truncated and full-length forms of the protein revealed that it too is expressed in taste cells in all three taste papillae ( Fig 1I–1K ) . The specificity of RNA probes ( S1B and S1C Fig ) and antibody ( S1E Fig ) were validated with the positive control jejunum tissue . Further , both reagents produced weak or no signal in CV papillae from Skn-1a knockout mice that lack all type II taste cells ( S1D and S1F Fig ) ; pre-incubation of GLI3 antibody with its immunogenic peptide confirmed specificity of the reagent ( S1G Fig ) . To confirm these results and identify other taste cell subtypes that express GLI3 we double-labeled taste cells with the GLI3 antibody along with other antibodies or with GFP transgenes that mark specific types of taste cells . In both anterior and posterior tongue fields GLI3 was frequently co-expressed with Tas1r3-GFP , a marker for sweet and umami receptor-expressing type II cells ( Fig 2A , 2D and 2G ) ; less frequently with TRPM5 , a marker for all type II cells ( Fig 2B , 2E and 2H ) ; and at even lower frequency with Gnat3-GFP , a marker for another subset of type II cells ( Fig 2C , 2F and 2I ) . Double-labeled immunohistochemistry with anti-GLI3 antibody plus an antibody against the type III taste cell markers CAR4 ( S2A–S2C Fig ) and 5-hydroxytryptamine ( 5-HT ) ( S2D–S2F Fig ) or with intrinsic GFP fluorescence of the type I marker Glast1-GFP ( S2G–S2I Fig ) revealed that GLI3 is generally not expressed in type I or III taste cells . Quantification showed that GLI3 was frequently expressed with TAS1R3 and TRPM5 , but rarely or not at all with CAR4 , 5HT or GLAST ( S2 Table ) . Among TAS1R3+ cells , ~92% expressed GLI3 in either CV or FO papillae . With TRPM5+ cells , 45–64% expressed GLI3 , with a higher percentage in the CV papillae . For GNAT3+ cells 31–51% expressed GLI3 , with a lower percentage in the FO vs . CV papillae taste cells . Only about 3% of CAR4+ and 1% of 5HT+ type III cells also expressed GLI3 . While for GLAST+ type I cells , no GLI3+ cells were found among 88 CV and 64 FO papillae taste cells examined . Among the GLI3+ cells , nearly all ( 97–99% ) also expressed TAS1R3 and/or TRPM5 , but only 35–63% expressed GNAT3 . In sum , GLI3 is expressed in type II cells , most often in the Tas1r3-GFP+ subset and less frequently in TRPM5+ or GNAT3+ subsets . As a key mediator of the Shh pathway that regulates the expression of a large number of genes [35 , 37] , Gli3 may play a significant role in taste cell regeneration and survival . To test this , we generated a double-knockin mouse strain homozygous for the floxed Gli3 allele and also carrying the Lgr5-EGFP-IRES-CreERT2 allele . In this strain , administering the ERT2 ligand tamoxifen would ablate Gli3 in Lgr5+ stem cells in posterior tongue . Immunostaining with an antibody for KCNQ1 , a marker for all taste cells , showed that in Gli3 conditional knockout ( Gli3CKO ) mice , the CV papillae taste buds were larger in size and contained more taste cells than did those of Gli3WT mice ( Fig 3A , 3B and 3I ) . Most of this may be accounted for by an increase in the number of TAS1R3+ type II cells , as the numbers of TRPM5+ and TAS1R3+ but not GNAT3+ and PKD2L1+ ( a type III cell marker ) cells increased dramatically in Gli3CKO mice ( Figs 3C–3H , 3J , S3A , S3E and S3I ) . Conversely , the numbers of CAR4+ type III cells decreased significantly in Gli3CKO mice ( S3B , S3F and S3I Fig ) . In agreement with this , qPCR showed that mRNAs expressing Tas1r3 , Gna14 and Trpm5 , but not Gnat3 , Pkd2l1 or Snap25 , increased in Gli3CKO mice ( Figs 3K and S3J ) . As expected , the number of GLI3+ cells and the amount of Gli3 mRNA decreased drastically , indicating that Gli3 deletion was successful . At the same time , GFP expression from the Gli3 locus was turned on in CV papillae tissue from Gli3CKO mice , which supports this conclusion ( S3C , S3D , S3G–S3J Fig ) . Lgr5-GFP is also expressed in FO papillae ( S4A Fig ) and changes similar to that in CV papillae were observed in taste buds from FO papillae from Gli3CKO mice; the number of taste cells per taste bud and the size of taste buds were higher , as were the numbers of TRPM5- and T1R3-expressing type II taste cells ( S4B–S4G , S4P & S4Q Fig ) . Conversely , the number of GNAT3-expressing type II cells and PKD2L1- and CAR4-expressing type III cells did not change; as expected , the number of GLI3-expressing cells were reduced drastically ( S4H–S4O , S4P & S4Q Fig ) . Analysis of FACS-purified cell populations revealed that the proportion of Lgr5+ taste stem cells increased in Gli3CKO mice ( S5A Fig ) . Consistent with this , qPCR showed that expression of Lgr5 mRNA in FACS-purified Lgr5-GFP+ cells ( S5B Fig ) and in CV papillae ( S5C Fig ) increased in Gli3CKO mice . As expected , the level of Gli3 mRNA in Lgr5-GFP+ cells from Gli3CKO mice was markedly reduced ( S5B Fig ) . Examining CV papillae taste cells from Gli3CKO mice showed that the expression of mRNA encoding the Gli3 target gene Jag2 decreased , while that for its upstream regulator Ptch1 increased ( S5D Fig ) . Based on results from selectively ablating Gli3 from Lgr5+ cells in posterior tongue , we infer that Gli3 suppresses the generation or survival of certain subsets of taste cells , in particular the Lgr5+ stem and Tas1r3+ type II cells in CV and FO papillae , but also the CAR4+ type III cells in CV papillae in vivo . Taste organoids cultured from single Lgr5-GFP+ cells faithfully recapitulate many features of taste cell development and function [38] . The effect of Gli3 ablation on the regenerative potential of Lgr5-GFP+ taste stem cells was tested in taste organoids derived from the double-knockin ( floxed Gli3 Lgr5-EGFP-IRES-CreERT2 ) mice by adding tamoxifen to the culture medium . Immunostaining of Gli3CKO organoids showed that the proportion of TAS1R3+ cells increased significantly while that of CAR4+ cells decreased ( the proportion of GNAT3+ cells remained unchanged ) ( Figs 4A–4E , S6A , S6B and S6I ) . Although not quantified , the proportion of NTPDase-expressing cells appeared largely unchanged ( S6C–S6D Fig ) . Consistent with these results , qPCR of Gli3CKO organoids showed that mRNAs expressing Tas1r3 , Lgr5 , Gna14 and Gnat3 increased , while those of Pkd2l1 and Snap25 decreased , and that of Trpm5 and NTPDase2 remained unchanged ( Figs 4F and S6J ) . Further , the expression of Shh target genes Gli1 , Mycn , Jag2 , and Ccnd2 decreased in Gli3CKO organoids , while that of the upstream regulator Ptch1 increased dramatically ( S6K Fig ) . As expected , in tamoxifen-treated Gli3CKO organoids the numbers of GLI3-immunoreactive cells and the level of Gli3 mRNA decreased , while GFP expression from the Gli3 locus , indicative of successful Gli3 deletion , was turned on ( S6E–S6J Fig ) . Collectively , these data suggest that Gli3 ablation in Lgr5+ taste stem/progenitor cells promote expansion or survival of Tas1r3+ cells and suppresses the differentiation of CAR4+ type III taste cells . In light of the profound changes in the proportion of taste cell subtypes and taste gene expression in tamoxifen-treated Gli3CKO mice , we investigated the effect of Gli3 ablation on taste responses . In brief-access taste tests , Gli3CKO mice showed altered behavioral responses to multiple taste qualities . Compared to Gli3WT , the Gli3CKO mice displayed increased preference for sucrose , sucralose and monosodium glutamate ( Fig 5A–5C ) , increased aversion to denatonium benzoate and citric acid ( Fig 5D and 5F ) , but no change in response to salt ( Fig 5E ) . Glossopharyngeal ( GL ) nerve recording revealed that compared to Gli3WT the Gli3CKO mice had increased nerve responses to sucrose , sucralose , and denatonium ( Figs 6A , 6B , 6D and S7 ) . However , the GL nerve responses to monosodium glutamate , citric acid , and NaCl were unchanged in Gli3CKO vs . Gli3WT mice ( Figs 6E and S7 ) . Furthermore , there were no significant differences in chorda tympani ( CT ) nerve responses of Gli3CKO vs . Gli3WT mice to most of the taste stimuli tested ( S8 Fig ) , consistent with the mosaic expression of Lgr5-Cre in the FF papillae taste buds that are innervated by the CT nerve [39] .
We used single cell transcriptomics to identify transcription factors selectively expressed in Tas1r3+ taste receptor cells , reasoning that they might play a role in the development and/or maintenance of these taste cells . Although the transcription factor Skn-1a is critical for the development of Tas1r3+ taste receptor cells , it plays this role in all type II taste cells [40] . We anticipated that other transcription factors would be expressed selectively in particular subsets of type II cells , e . g . Tas1r3+ sweet/umami cells or Tas2r+ bitter cells . By single cell transcriptomics we found that the transcription factor Gli3 and its upstream regulators Ptch1 and Smo were more highly expressed in Tas1r3+ taste cells and Lgr5+ stem cells than in Gad1+ type III cells . Conventional expression studies using PCR , in situ hybridization and immunohistochemistry confirmed that Gli3 was indeed expressed in taste cells . Using Skn-1a null mice lacking all type II cells we showed that Gli3 was expressed selectively in type II cells . By double immunohistochemistry we found that Gli3 was most highly expressed in Tas1r3+ taste cells vs . other types of type II cells ( e . g . Trpm5+ or Gnat3+ type II cells ) , and not expressed in type I or III taste cells . Gli1 , 2 , 3 are zinc finger-containing transcription factors that act via the Shh pathway to regulate organogenesis and self-renewal [41 , 42] . Gli2 and Gli3 are the main effectors of the Shh pathway in adults , with Gli2 acting mainly as a transcriptional activator and Gli3 as a repressor [33 , 34] . Overexpression of Gli2 leads to malformation of taste buds in FF papillae , while overexpression of a dominant negative Gli2 transgene or deletion of Gli2 in taste cell precursors results in loss of taste buds in both FF and CV papillae [19 , 43] . However , prior to our work the effects of manipulating Gli3 on adult taste cell regeneration were not known . To determine what role Gli3 might play in taste cells we turned to knockout mice . Conventional Gli3 null mice are embryonic/perinatal lethal [44 , 45]; therefore we generated conditional null mice in which Gli3 was selectively eliminated from taste stem cells using a transgene in which CRE-ERT2 was driven from the Lgr5 promoter . Conditional ablation of Gli3 from taste stem cells and their progeny in the posterior taste field led to altered taste bud morphology with numbers of Tas1r3+ taste cells , but not of Gnat3+ cells . These changes may be cell-autonomous and cause only an increase in taste bud size at the expense of the epithelial tissue within the taste papillae or cause an overall increase in the size of the taste papillae by affecting the fate of the neighboring non-taste epithelium by non-cell-autonomous mechanisms . We have not tested which of these two possibilities account for the changes in Gli3CKO mice . The Gli3CKO mice showed altered short-term lick test responses to sweet , umami , and bitter tastants and diminished glossopharyngeal nerve responses to sweet and bitter . Although Lgr5-CRE-ERT2 is only expressed in a weak , mosaic pattern in FF papillae , we observed modest changes in CT nerve responses to sucralose and citric acid , indicating that Gli3 could play a role in the anterior taste field also . Definitively determining this will require experiments using a Cre driver that is strongly expressed in FF papillae . The Shh pathway is active in all taste papillae [19 , 24 , 46 , 47] , but its effect is context dependent . In the embryonic stage , Shh signaling suppresses the development of FF papillae , while it promotes taste bud development in CV papillae [20 , 28 , 29 , 48] . In adults , Shh expressing cells give rise to all subtypes of taste cells; pharmacological inhibition of Shh signaling inhibits taste cell turnover [21 , 24 , 47 , 49] . Further , overexpression of Shh in the lingual epithelium triggers the development of multiple ectopic FF taste buds [27] . SHH is secreted by a subpopulation of post-mitotic cells in the base of the taste buds , and SHH-responsive , putative stem cells are located around and outside the base of taste buds . Indeed , current evidence suggests that the Shh pathway is active in stem cells [19 , 23]; and is critical for the development of taste cells in all taste fields , as noted above . Lgr5 is a marker for posterior taste stem cells , but also a co-receptor in the Wnt signaling pathway [38 , 39] . Because we ablated Gli3 in posterior tongue using the Lgr5-CreERT2 driver and because the Shh pathway is downstream of Wnt [28 , 50] it is likely that we ablated Gli3 in taste stem cells before Shh signaling was turned on . Using the Lgr5-CreERT2 driver and tamoxifen , Gli3 was ablated from most but not all posterior field taste cells . The remaining Gli3+ cells may be progeny of the Lgr5+ cells where Gli3 deletion failed or long-lived taste cells generated prior to tamoxifen treatment . In Gli3CKO mice the numbers of type III cells did not change overall , but the Car4+ subset of type III cells decreased markedly . Notably , Gli3 is not expressed in type III cells , including Car4+ cells , so its effect on this cell type is most likely a consequence of Gli3 activity in the Lgr5+ stem cells themselves or in lineage-specific precursor cells that gave rise to Car4+ cells . It is possible that the lack of Gli3 inhibits the differentiation of Car4+ cells as it does not affect the expression of Car4 . CAR4 is thought to be necessary for amiloride-insensitive salt taste perception [51] , but Gli3CKO mice retained normal salt taste sensitivity . Conversely , Gli3CKO mice had heightened sensitivity to all other primary taste qualities in brief-access tests and to bitter and sweet tastants in taste nerve responses . The magnitude of the changes in lick responses in particular are somewhat surprising because the anterior taste field is not affected in Gli3CKO mice , and may mask the effect of changes in the posterior taste field . But it is possible that taste buds in the soft palate , which also are endoderm-derived and in the pharynx could show changes similar to those in CV and FO papillae in Gli3CKO mice , although we have not tested this . Consistent with these observations , only those taste qualities that elicit stronger responses in the posterior taste field , namely sweet and bitter , show robust changes in Gli3CKO mice . On the other hand , the behavioral and GL nerve responses to umami tastants did not change dramatically , although Gli3CKO mice had a higher number of Tas1r3+ cells and expressed more Tas1r3 mRNA than did wild-type mice in CV papillae . This may reflect the low baseline umami taste sensitivity and expression level of the Tas1r1 subunit of the umami taste receptor in CV papillae [52 , 53] . In Gli3CKO mice the number of bitter ( Gnat3+ ) and sour ( Pkd2l1+ ) receptor cells did not change , but the sensitivity to these tastants , especially to bitter , increased . This may be attributed at least in part to changes in innervation density or selective innervation of particular types of taste cell types , although we have not tested this . Another possibility is that the expression level of taste receptors or their downstream signaling/regulatory machinery changed in Gli3CKO mice . Indeed , the expression of many taste marker genes , such as Tas1r3 , Trpm5 , Gnat3 , Gna14 , Snap25 , Pkd2l1 and NTPDase2 , is affected in Gli3CKO mice and/or organoids . In CV papillae and/or organoids derived from Gli3CKO mice , we observed changes in mRNA expression of the Gli3 target genes Ccnd2 , Mycn , and Jag2 and of the upstream regulator Ptch1 . While these changes confirm that Gli3 deletion had the expected effects , they represent only a small subset of the thousands of Gli3 target genes . RNA-Seq analysis of Gli3CKO taste cells may help identify many more genes affected by Gli3 deletion and help delineate the developmental pathways regulated by Gli3 in taste cells . In this study we demonstrate the utility of organoids cultured from purified taste stem cells for studying taste system development . Being an ex vivo system , taste organoids are not influenced by signals from other tissues . Hence , the results of genetic or other manipulations can be interpreted in a more straightforward manner . Also , the role in the taste system of key genes and pathways can be readily studied in taste organoids without concern for lethal effects from knockouts in vivo . Further , large numbers of cultured taste cells can be obtained from organoids which will be useful for protein expression and biochemical studies . Indeed , the effect of Gli3 knockout in taste organoids largely parallels that observed in vivo , underlining the utility of this system . What could be responsible for the increases in Lgr5+ and Tas1r3+ cells in Gli3CKO mice ? In other tissues , Shh signaling can drive either differentiation or maintenance of stem cells [54 , 55] . It is possible that Gli3 enhances taste stem cell maintenance and acts as a negative regulator of taste cell differentiation . Another possibility is that Gli3 promotes apoptosis of Tas1r3+ and/or Lgr5+ cells . In either case , our data support a critical role of Gli3 activity in both stem and type II sweet taste receptor cells . The continued expression of Gli3 in Tas1r3+ cells and the profound changes in the number of these cells and in sweet and bitter taste sensitivity in Gli3 conditional knockout mice are evidence for an additional role for Shh signaling and Gli3 in these mature taste cells . One way to tease apart the role of Shh signaling in stem and Tas1r3+ cells is by conditional ablation of Gli3 or other Shh pathway components using Tas1r3- or type II-specific Cre drivers ( e . g . Skn-1a [40] ) . The role of other signaling pathways in taste development can also be context dependent . The Wnt and Bmp signaling pathways are critical for the development of embryonic FF papillae , but play relatively minor roles in the CV papillae [17 , 28 , 56–58] , while the Fgf signaling pathway , much like the Shh pathway , plays opposite roles in embryonic CV and FF papillae development [2] . Such differences are not surprising given that developmentally the FF papillae originate from the ectoderm while the CV and FO papillae are derived from the endoderm [59] . Many of these pathways may play relatively subtle but significant roles in taste fields where they seem dispensable ( similar to what we have shown for Gli3 , and by extension the Shh pathway ) . In summary , our results indicate that Gli3 is a suppressor of taste stem cell proliferation and affects the number and function of mature taste cells , especially of the Tas1r3+ subtype in posterior tongue . Our findings shed more light on adult taste cell regeneration and may help devise strategies for treating taste distortions caused by conditions such as chemotherapy and aging .
All animal experiments were performed in accordance with the National Institutes of Health guidelines for the care and use of animals in research and approved by the Institutional Animal Care and Use Committee at Monell Chemical Senses Center ( protocols: #1163 , #1151 ) . 6-12-week old were used for all experiments . Animals were housed with a 12-h light/dark cycle and ad libitum access to food and water . The double-knockin mouse strain carrying a floxed Gli3 allele was a kind gift from Dr . Rolf Zeller , University of Basel ( Basel , Switzerland ) [60] . Lgr5-EGFP-IRES-CreERT2 knockin mice and Tas1r3-GFP and Gnat3-GFP transgenic mice were as previously described [61 , 62] . Glast1-EMTB-GFP was a kind gift from Dr . Eva Anton , University of North Carolina School of Medicine ( Chapel Hill , NC ) [63] . Skn-1a knockout mice were a kind gift from Dr . Ichiro Matsumoto , Monell Chemical Senses Center ( Philadelphia , PA ) [40] . For Cre activation , tamoxifen ( Sigma-Aldrich , St . Louis , MO; cat . no . T-5648 ) was dissolved in corn oil ( Sigma-Aldrich cat . no . C8267 ) to a stock concentration of 20 mg/ml and administrated by oral gavage for three weeks at a dose of 2 mg/20 g body weight . Mice were given 2-day breaks each week during treatment to recover from the drug . Tissue was harvested 4 weeks after completion of tamoxifen treatment . Mice were sacrificed by CO2 asphyxiation , and the tongues excised . An enzyme mixture ( 0 . 5 ml ) consisting of dispase II ( 2 mg/ml; Roche , Mannheim , Germany; cat . no . 04942078001 ) and collagenase A ( 1 mg/ml; Roche cat . no . 10103578001 ) in Ca2+-free Tyrode’s solution ( 145 mM NaCl , 5 mM KCl , 10 mM HEPES , 5 mM NaHCO3 , 10 mM pyruvate , 10 mM glucose ) was injected under the lingual epithelium , which was then incubated for 15 min at 37°C . Lingual epithelia were peeled gently from the underlying muscle tissue and used for single-cell RNA-Seq , FACS sorting , or RNA isolation . Single cell RNA-Seq was done as described [64] . GFP-expressing cells that were not physically attached to any other cell or cell fragment were picked irrespective of their shape individually from single cell preparations of CV papillae of Tas1r3-GFP ( type II , sweet and umami receptor cells , n = 9 ) , Lgr5-GFP ( stem cells , n = 5 ) , and Gad1-GFP type III , sour and high salt receptor cells , n = 11 ) transgenic mice . Two rounds of single-cell mRNA amplification were done using the TargetAmp 2-Round aRNA Amplification Kit 2 . 0 ( Epicentre , Madison , WI ) . The antisense RNA generated from single cells was converted to Illumina sequencing libraries using the NEBNext Ultra Directional RNA Library Prep Kit for Illumina ( New England Biolabs , Ipswitch , MA ) and sequenced using the Illumina HiSeq 2000 platform . Sequencing reads were mapped to the mouse genome ( version mm10 , p4 ) using the STAR aligner [65] using Gencode M7 as splice junction database ( https://www . gencodegenes . org/mouse_releases/7 . html ) . The reads mapping to genes were counted using the featureCounts package [66] with Gencode M7 as reference . Data normalization and differential expression analysis were done using the DESeq2 package in R [67] . We obtained 30–70 million reads per library , of which 70–90% could be aligned to the mouse genome . On average , 10 , 184 genes were expressed per cell above an arbitrary cutoff of 10 reads per gene after normalization . GFP-fluorescent Tas1r3+ , Gad1+ , and Lgr5+ taste cells were isolated by FACS from male mice of these respective genotypes . The region of the lingual epithelium containing the CV papillae from four to five mice was excised and pooled , minced into small pieces , incubated with trypsin ( 0 . 25% in PBS ) for 10–25 min at 37°C , and mechanically dissociated into single cells using heat-pulled Pasteur pipettes . Cell suspensions were filtered using 70-μm cell strainers ( BD Biosciences , Bedford , MA; cat . no . 352350 ) and then with 35-μm cell strainers ( BD Biosciences cat . no . 352235 ) . Cells were sorted into culture medium for organoid culture or Trizol LS ( Thermo Fisher cat . no . 10296010 ) for RNA isolation using a BD FACS Aria II SORP FACS machine ( Flow Cytometry and Cells Sorting Resource Laboratory , University of Pennsylvania ) , according to the enhanced green fluorescent GFP ( EGFP ) or GFP signal ( excitation , 488 nm; emission , 530 nm ) . Total RNA was isolated from freshly dissected taste papillae , nontaste control epithelium from the ventral surface of the tongue , and taste organoids using the PureLink mini kit with on-column DNA digestion using PureLink DNase ( Thermo Fisher cat . no . 12185010 ) and converted into cDNA using Super Script VILO kit ( Thermo Fisher cat . no . 11755050 ) . RNA from FACS-sorted cells was isolated using the Trizol LS kit , and cDNA was synthesized using Ovation qPCR System ( NuGEN , San Carlos , CA; cat . no . 2210–24 ) . End-point PCR and qPCR were done as described [68] . Initially the expression of Gli3 was plotted as the logarithm of the ratio between its cycle threshold value and that of Gapdh . Subsequently , all qPCR results were normalized using the ΔΔCt method with Bact as reference . Taste organoids were prepared as described [38] . Briefly , GFP fluorescent cells sorted from double-knockin mice were mixed with 4% chilled Matrigel ( v/v; BD Biosciences , San Jose , CA; cat . no . 354234 ) and cultured in DMEM/F12 ( Thermo Fisher cat . no . 11320–033 ) supplemented with Wnt3a-conditioned medium ( 50% , v/v ) , R-spondin-conditioned medium ( 20% , v/v ) , Noggin-conditioned medium ( 10% , v/v ) , N2 ( 1% , v/v; Thermo Fisher cat . no . 17502–048 ) , B27 ( 2% , v/v; Thermo Fisher cat . no . 12587–010 ) , Y27632 ( 10 μM; Sigma-Aldrich cat . no . Y0503 ) , and epidermal growth factor ( 50 ng/mL; Thermo Fisher ) . Wnt3a- and R-spondin-conditioned medium are generated from Wnt3a and R-spondin stable cell lines as described [69] . Noggin conditioned medium was made in house . The culture medium was changed first at day 5–7 and once every 2–3 days thereafter . For passage , single-cell preparations were made from taste organoids by digestion with 0 . 25% trypsin for 10 min at 37°Cat day 14 before seeding again onto culture plates . 4-Hydroxytamoxifen ( 10 μg/ml; Sigma-Aldrich cat . no . H7904 ) was added into the fresh sorted cells for 5 consecutive days for Cre activation . Adult male mice were euthanized by CO2 asphyxiation , and taste-papillae-containing portions of the tongue were quickly removed and briefly rinsed in ice-cold PBS . For in situ hybridization , tissues were freshly frozen in Tissue-Tek O . C . T . mounting media ( Sakura Finetek , Torrance , CA; cat no . 4583 ) using a 100% ethanol dry ice bath and sectioned within 1 h after dissection . For immunohistochemistry , tissues were fixed for 1 h at 4°C in 4% paraformaldehyde in 1× PBS and cryoprotected in 20% sucrose in 1× PBS overnight at 4°C before embedding in O . C . T . Sections ( 10 μm thickness , coronal for FF and CV papillae , horizontal for FOL papillae ) were prepared using a CM3050S cryostat ( Leica Microsystems ) and applied on precoated Fisherbrand Superfrost microscope slides ( Fisher Scientific , Hampton , NH , Cat no 12-550-123 ) . Sections were dried at 40°C for 20 min and immediately used for in situ hybridization or stored at −80°C for immunostaining . Standard in situ hybridization methods were used as described . Fresh tissue sections with taste papillae were incubated with hybridization of 0 . 3 μg/ml Gli3 probe ( GenBank NM_00813 , 1809–2427 bp ) . Antisense and sense RNA probes were used at equivalent concentrations and run in parallel in the same experiment to ensure equivalent conditions . For each experiment , a positive control hybridization using Tas1r3 probe was done . In addition , in situ hybridization experiments were done on positive control tissues to confirm the quality and specificity of the RNA probes . Immunostaining of taste buds was done as described . The antibodies used in this study and their concentrations are listed in S4 Table . For serotonin detection , mice were injected with 5-HT ( Sigma-Aldrich cat . no . H9523 ) and sacrificed after 2 h . Species-specific secondary antibodies ( S4 Table ) were used to visualize specific taste cell markers and GLI3 . For antibody staining of organoids , cultured organoids were collected in 1 . 5 mL Eppendorf tubes and fixed for 15 min in fresh 4% paraformaldehyde in 1× PBS supplemented with MgCl2 ( 5 mM ) , EGTA ( 10 mM ) , and sucrose ( 4% , wt/v ) , washed three times for 5 min with 1× PBS , and blocked for 45 min with SuperBlock blocking buffer ( Thermo Fisher cat . no . 37515 ) supplemented with 0 . 3% ( v/v ) Triton X-100 and 2% ( v/v ) donkey serum . They were then incubated at 4°C overnight with the desired primary antibodies ( S4 Table ) . They were washed 3× for 5 min with 1× PBS and incubated for 1 h with species-specific secondary antibodies ( 1:500 ) . 6-Diamidino-2-phenylindole ( DAPI , 1:1000 ) in deionized water was used to visualize the nuclei following secondary antibody . Bright-field images were generated using a Nikon DXM 1200C digital camera attached to a Nikon Eclipse 80i microscope and captured using Nikon NIS-Element F 3 . 00 software . Acquisition parameters were held constant for images with both antisense and sense probes . Fluorescent images were captured with the TCS SP2 Spectral Confocal Microscope ( Leica Microsystems Wetzlar , Germany ) using UV , Ar , GeNe , and HeNe lasers and appropriate excitation spectra . Scanware software ( Leica Microsystems ) was used to acquire z-series stacks captured at a step size of 2–3 μm . Acquisition parameters ( i . e . , gain , offset , PMT settings ) were held constant for experiments with antibodies and for controls without antibodies . Digital images were cropped and arranged using Photoshop CS ( Adobe Systems ) . Fluorescence images within a figure were adjusted for brightness and contrast for background standardization . Quantitative measurements were carried out to determine the percentage of singly and doubly labeled type II and type III taste cells that co-expressed GLI3 and taste marker proteins . Confocal images from two to four sections from CV and FO papillae in each mouse were used for counting . To avoid counting the same cells more than once , sections separated from each other by at least 40 μm were chosen . Nuclear staining with DAPI was used to help distinguish individual taste cells . Only cells with entire cell bodies and nuclei visible were used for counting . GLI3-positive and taste-marker-labeled taste cells were counted in respective single-channel images , and the double-positive cells were counted using overlaid images . KCNQ1 antibody staining was used to visualize taste buds for determination of taste bud size and taste cell number . Measurement of taste bud size was conducted as previously described [70] . Five Gli3CKO and Gli3WT mice were used for counting taste cell number . We found , on average , 10 taste buds per trench and 20 cells in each taste bud section . Only taste buds with typical morphology ( with clear taste pore and the base of taste bud reaching the basement membrane ) were used for analysis . Serial sections from similar regions of the tissue from each mouse were used to minimize location difference in taste bud number and size . The average number of nuclei in each taste bud was used as a proxy for the number of taste cells . For measuring the number of taste buds , all KCNQ1+ taste buds were counted , regardless of the morphology of taste bud . Double immunostaining was conducted using KCNQ1 antibody and respective taste cell marker antibodies to quantify the number of T1R3+ , TRPM5+ , GNAT3+ , PKD2L1+ , and CAR4+ taste cells per total taste cells per section . To quantify the percentage of taste cell subtypes in taste organoids with a clear single organoid profile from Gli3CKO mice , single and double immunostaining was performed using specific taste cell markers or Gli3 antibody . Nuclear staining with DAPI was used to help distinguish individual taste cells . Only cells with entire cell bodies and nuclei visible were used for counting . Brief access tests were conducted using the Davis MS-160 mouse gustometer ( Dilog Instruments , Tallahassee , FL ) as described[71] . The following taste compounds were tested: sucrose ( 30 , 100 , 300 , 1000 mM ) , sucralose ( 1 , 3 , 10 , 30 mM ) , monosodium glutamate ( MSG; 30 , 100 , 300 , 100 mM ) , denatonium ( 0 . 3 , 1 , 3 , 10 mM ) , citric acid ( 1 , 3 , 10 , 30 mM ) , and NaCl ( 30 , 100 , 300 , 600 mM ) . Mice were water- and food-restricted ( 1 g food and 1 . 5 mL water ) for 23 . 5 h before test sessions for appetitive taste compounds ( sucrose , sucralose , and MSG ) . For the aversive taste compounds ( citric acid , denatonium , and NaCl ) , mice were water-deprived for 22 . 5 h before testing . In each test session , four different concentrations of each taste compound and water control were presented in a random order for 5 s after first lick , and the shutter reopened after a 7 . 5-s interval . The total test session time was 20 min . An additional 1-s “washout period” with water was interposed between each trial in sessions testing aversive tastants . Gli3CKO and Gli3WT mice were tested at the same time in parallel . Each mouse was tested with all the compounds . After each session mice were allowed to recover for 48 h with free access to food and water . Body weight of the mice was monitored daily , and only mice at or over 85% their initial body weight were used . The ratio of taste stimulus to water licks was calculated by dividing the number of licks for taste compounds by the number of licks for water presented in the parallel test session . Lick ratios > 1 indicate preference behavior to the taste compound , and lick ratios < 1 indicate avoidance behavior to the taste compound . With bitter and sour stimuli it appears that lick responses show a ceiling effect ( maximum aversion ) at higher concentrations . Thus , strain differences could only be seen at the lower concentrations . The same sets of mice used for behavioral tests were used for electrophysiological recording of taste responses . Whole-nerve responses to tastants were recorded from the chorda tympani ( CT ) or the glossopharyngeal ( GL ) nerves as described [72] . Mice were anesthetized by an intraperitoneal injection ( 10 ml/kg , with 2 . 5 ml/kg further doses as necessary ) of a mixture of ketamine ( 4 . 28 mg/ml ) , xylazine ( 0 . 86 mg/ml ) , and acepromazine ( 0 . 14 mg/ml ) . Under anesthesia , the trachea of each mouse was cannulated , and the mouse was then fixed in the supine position with a head holder to allow dissection of the CT or the GL nerve . The right CT nerve was dissected free from surrounding tissues after removal of the pterygoid muscle and cut at the point of its entry to the tympanic bulla . The right GL nerve from a different animal was exposed , dissected free from underlying tissues and cut near its entrance to the posterior lacerated foramen . All chemicals were used at ~24°C . The entire nerve was placed on the Ag-AgCl electrode . An indifferent electrode was placed in nearby tissue . For taste stimulation of fungiform papillae ( FP ) , the anterior half of the tongue was enclosed in a flow chamber made of silicone rubber . For taste stimulation of the CV , an incision was made on each side of the animal's face from the corner of the mouth to just above the angle of the jaw , and the papillae were exposed and their trenches opened by slight tension applied through a small suture sewn in the tip of the tongue . For taste stimulation of fungiform papillae ( FP ) , the anterior half of the tongue was enclosed in a flow chamber made of silicone rubber . Taste solutions were delivered to each part of the tongue by gravity flow for 30 s ( CT ) or 60 s ( GL ) at the same flow rate as the distilled water used for rinse ( ~0 . 1 ml/s ) . The following taste compounds were tested: sucrose ( 100 , 300 , 1000 mM ) , sucralose ( 3 , 10 , 30 mM ) , MSG ( 30 , 100 , 300 mM ) , denatonium ( 1 , 3 , 10 mM ) , citric acid ( 3 , 10 , 30 mM ) , and NaCl ( 30 , 100 , 300 mM ) . Neural responses resulting from chemical stimulations of the tongue were fed into an amplifier ( K-1; Iyodenshikagaku , Nagoya , Japan ) and monitored on an oscilloscope and an audio monitor . The whole-nerve responses were integrated with a time constant of 1 . 0 s , recorded using software ( PowerLab 4/30; AD Instruments , Bella Vista , Australia ) , and analyzed using LabChart Pro software ( AD Instruments ) . Nerve response magnitudes were measured at 5 , 10 , 15 , 20 , and 25 s after stimulus onset for the CT nerve and at 5 , 10 , 20 , 30 , and 40 s for the GL nerve . The stability of each preparation was monitored by the periodic application of 0 . 1 M NH4Cl . A recording was considered to be stable when the 0 . 1 M NH4Cl response magnitudes at the beginning and end of each stimulation series deviated by no more than 15% . Only responses from stable recordings were used for data analysis . At the end of the experiment , animals were killed by injecting an overdose of the anesthetic . The response values were averaged and normalized to responses to 100 mM NH4Cl to account for mouse-to-mouse variations in absolute responses . In Glossopharyngeal nerve recordings ( S7 Fig ) , the responses appear not to return to baseline for some of the highest concentrations of stimuli because it is difficult and takes much time to wash them out completely from the CV cleft . Subsequent recordings were only done after repeated washings to make sure the previous response return to baseline . All data were compared as normalized units . Prism ( GraphPad Software ) was used for statistical analyses , including calculation of mean values , standard errors , and unpaired t-tests of cell counts and qPCR data . Data from taste behavioral tests and gustatory nerve recording were compiled using Microsoft Excel . For statistical analyses of behavioral and nerve responses , two-way ANOVA and post hoc t-tests were used to evaluate the difference between genotype ( Gli3CKO and Gli3WT mice ) and concentration using OriginPro ( OriginLab ) . p-Values < 0 . 05 were considered significant . | Adult taste cell regeneration is essential for maintaining peripheral taste cells throughout life . The Shh pathway is an important regulator of taste bud development and regeneration in both embryonic and adult stages . We show that the transcription factor Gli3 , an important effector of the Shh pathway , is expressed in Tas1r3-expressing sweet/umami taste receptor cells and Lgr5-expressing taste stem cells . Conditional deletion of the Gli3 gene led to increased numbers of Tas1r3-expressing taste cells and Lgr5-expressing stem cells along with altered responses to bitter and sweet tastants . Our findings shed light on adult taste cell regeneration and may lead to new treatments of taste disorders associated with aging and chemotherapy . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"taste",
"buds",
"organoids",
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"biological",
"cultures",
"social",
"sciences",
"vertebrates",
"mice",
"neuroscience",
"animals",
"mammals",
"tongue",
"organ",
"cultures",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"citric",
"acid",
"taste",
"gene",
"expression",
"acids",
"chemistry",
"sensory",
"receptors",
"signal",
"transduction",
"psychology",
"immunostaining",
"eukaryota",
"mouth",
"rodents",
"cell",
"biology",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"sensory",
"perception",
"physical",
"sciences",
"amniotes",
"organisms"
] | 2018 | Gli3 is a negative regulator of Tas1r3-expressing taste cells |
There is growing evidence that molecular subtypes ( e . g . luminal and basal subtypes ) affect the prognosis and treatment response in patients with muscle invasive urinary bladder cancer ( invasive urothelial carcinoma , iUC ) . Modeling these subtypes in pre-clinical animal studies is essential , but it is challenging to produce these subtypes , along with other critical host and tumor features , in experimentally-induced animal models . This study was conducted to determine if luminal and basal molecular subtypes are present in naturally-occurring canine iUC , a cancer that mimics the human condition in other key aspects . RNA sequencing was performed on 29 canine treatment naive iUC tissue samples and on four normal canine bladder mucosal samples . Data were aligned to CanFam 3 . 1 , and differentially expressed genes were identified . Unsupervised hierarchical clustering of these genes revealed two distinct groups ( n = 13 , n = 16 ) . When genes that distinguish basal and luminal subtypes in human cancer ( n = 2015 ) were used to probe genes differentially expressed between normal canine bladder and iUC , 829 enriched signature genes were identified . Unsupervised hierarchical clustering of these genes revealed two distinct groups comprised of 18 luminal subtype tumors and 11 basal subtype tumors . The enriched genes included MMP9 , SERPINE2 , CAV1 , KRT14 , and RASA3 in basal tumors , and PPARG , LY6E , CTSE , CDK3 , and TBX2 in luminal tumors . In supervised clustering , additional genes of importance in human iUC were identified in canine iUC associated with claudin-low and infiltrated tumors . A smaller panel of genes ( n = 60 ) was identified that distinguished canine luminal and basal iUC with overall 93 . 1% accuracy . Immune signature patterns similar to those in human iUC were also identified with the greatest enrichment of immune genes being in the basal subtype tumors . These findings provide additional compelling evidence that naturally-occurring canine iUC is a highly relevant and much needed model of human iUC for translational research .
Muscle invasive bladder cancer or more specifically , invasive urothelial carcinoma ( iUC ) , is lethal in approximately 50% of patients , with most deaths being due to drug-resistant metastatic disease [1] . Clearly better therapies for iUC are needed . One of the key advances that could improve iUC therapy has been the identification of molecular subtypes ( luminal , basal ) which are linked to iUC behavior and response to chemotherapy , targeted agents , and immunotherapies [2–10] . Basal iUC is inherently aggressive , is associated with more advanced local stage and metastasis at diagnosis , and is enriched for STAT3 , TP63 , KRT5/6A , CD44 , HIF-1 , and EGFR [2–10] . Luminal iUC is thought to be associated with better clinical outcomes , and is enriched for ER , TRIM24 , FOXA1 , GATA3 , FGFR3 , and PPARG [2–10] . Subclassifications within these major subtypes have also been identified [9 , 10] . New drugs and drug combinations delivered in the context of molecular subtypes could greatly improve the outlook for patients with iUC . Relevant pre-clinical animal models that accurately predict drug effects in humans are needed to optimize new treatment protocols for iUC [11 , 12] . This is especially the case now as the number of bladder cancer patients is insufficient to test even part of the potential new drug and combination therapy protocols of interest . Studies in relevant pre-clinical models could be used to screen multiple approaches with the most promising ones taken into human trials . These animal models must recapitulate the molecular subtypes of iUC found in humans , as well as mirroring other aspects of the human condition ( heterogeneity , innate and acquired drug resistance , metastasis , etc . ) . In addition , with the resurgence of immunotherapies , and the recognition that the immune response plays an important role in the efficacy of other cancer therapies as well , it is critical that the animal model also have a level of immunocompetence similar to that in human cancer patients [13 , 14] . While it is difficult to create an experimental animal model that has these collective features , there is growing evidence that naturally-occurring iUC in pet dogs can serve as this essential relevant model to complement traditional experimental models of bladder cancer [11 , 12] . Invasive UC in dogs closely mimics the cancer in humans in pathology , molecular features , biological behavior including sites and frequency of distant metastasis , and response to chemotherapy [11 , 12] . Proteomic and genomic analyses have defined further intriguing similarities between iUC in dogs and humans [15 , 16] . Differentially expressed genes ( between iUC and normal bladder ) that are shared between dogs and humans include >450 genes identified by microarray , and >800 identified by RNA-seq ( P<0 . 05; 2FC ) [15 , 16] . Relevant to the importance of molecular subtypes , early microarray studies of canine iUC have indicated the presence of luminal and basal subtypes similar to those in human iUC [15] . Though vast similarities exist between the human and dog disease , there are also expected differences . Activating mutations in the MAPK pathway occur in iUC in both species , but BRAF mutations which predominate in dogs , are rare in humans [11] . Clinical trials in dogs with iUC are considered a win-win scenario with each dog having the chance to benefit , and new knowledge gained that will help humans and dogs . With canine iUC comprising approximately 2% of the estimated 6 million new canine cancer cases each year , there are ample numbers of dogs for translational research . In addition to treatment trials , canine iUC can be used to elucidate environmental exposures and gene-environment interactions involved in the cancer development . Exquisitely high breed-associated risk , such as the 20X increased risk in Scottish Terriers compared to mixed breed dogs , offers unparalleled opportunities to investigate the causes of iUC , and to perform early detection / early intervention trials [12] . Chemical exposures ( lawn chemicals , older flea control products ) have been associated with increased iUC risk in dogs , and consumption of vegetables in addition to commercial dog food has been associated with a lower risk of iUC in dogs [12] . RNA-seq analysis of canine iUC was performed to confirm and extend earlier findings . Specifically , the aims of the work were to: ( 1 ) determine the presence of the major luminal and basal transcriptional subtypes in canine iUC , ( 2 ) explore the presence of additional subclassifications within subtypes as reported in humans , ( 3 ) define a smaller set of genes which could be used to subtype canine iUC , and ( 4 ) investigate the immune signatures present in canine iUC . The study was successfully completed , and the findings provide further compelling evidence that naturally-occurring iUC in dogs can serve as a highly relevant model of human invasive bladder cancer .
Unsupervised clustering of canine iUC samples segregated canine iUC into two distinct groups ( n = 13 and n = 16 ) ( Fig 1 ) . Hierarchical clustering of genes that are differentially expressed in basal and luminal subtypes in human breast cancer ( n = 2015; of which 829 were present in the canine dataset ) , and confirmed to be of relevance in human iUC [9] , revealed two distinct groups within the canine tumor samples ( Fig 2 ) . The group with the larger number of tumors ( n = 18 , 62% of tumors ) was enriched for genes reported in human luminal iUC with examples including PPARG , FOXA1 , CTSE , CDK6 , TBX2 , and SNCG [1 , 4 , 5 , 8] . The second group ( n = 11 , 38% of tumors ) was enriched for genes reported in human basal iUC including MMP9 , SERPINE2 , RASA3 , and PLAUR [5 , 6 , 8] . It is pertinent to state that the tumors identified as basal ( n = 11 ) , were the same tumors as present in the smaller group using K-means clustering ( n = 13 ) . Differentially expressed genes commonly identified in the canine and human iUC samples in The Cancer Genome Atlas ( TCGA ) included , but were not limited to , PIGR , AGR2 , CDK3 , ACTG2 , DES , TPM2 , TAGLN , and MYH11 [4 , 17] . Supervised clustering of genes reported to be of importance in subclassifying subtypes in human iUC confirmed the presence of the gene signatures in canine iUC , and included genes in addition to those enriched in the pairwise comparison . Genes of particular interest because of their relevance to human luminal iUC included PPARG , VGLL1 , and LY6E [5 , 8] . The expression of these genes was heterogenous across canine luminal iUC samples ( Fig 3A ) . Similar heterogeneity was observed when examining PI3 , DSG3 , and KRT14 in canine basal iUC samples ( Fig 3B ) [5 , 8] . Most of the genes reported to be associated with assigning “infiltrated” human iUC were found to be upregulated in the canine basal iUC samples [8] , although SRGN , SERPINE2 , and RGS2 were enriched in some of the canine luminal iUC samples ( Fig 3C ) . Hierarchical clustering of bladder tumors using a list of genes that was previously utilized to define claudin-low tumors in human breast cancer patients , confirmed basal tumors to be enriched for claudin-low genes [18] . Two luminal tumors also exhibited enrichment of genes associated with claudin-low subtype ( Fig 4A ) . In the panel of genes relevant to the claudin-low transcriptional subtype ( Fig 4B ) , CLDN7 , CLDN3 , and CDH1 were upregulated in canine luminal iUC , and SNAI2 , ZEB1 , and ZEB2 were upregulated in canine basal iUC [8 , 17 , 18] . The enriched immune-suppressive signatures associated with claudin-low ( in human breast cancer ) and the enriched chemokines and cytokines enriched in human iUC ( Fig 4D ) further confirmed the claudin-low subtype in the majority of canine basal ( and one luminal ) tumors [18 , 19] . The p53 signature genes [3 , 5] were enriched in some of the luminal and basal tumors , as reported in human iUC ( Fig 5A ) . For example , IGF-1 , SERPING-1 , MFAP4 , LMO3 , and PGM5 were enriched in canine basal tumors , whereas CCNE2 , CX3CL1 , FEN1 , and UHRF1 were found to be enriched in canine luminal tumors . P63 pathway genes such as TNC , PI3 , SERPINE-1 , S100A8 , GFI1 , and RAC2 , were enriched in canine basal tumors ( Fig 5B ) [3 , 5] . Some of the genes upregulated in the PPARG pathway in basal tumors in human iUC , i . e . , COL1A1 , COL1A2 , CAV1 , and ACTA2 , were also upregulated in canine basal tumors ( Fig 5C ) [5] . A panel of genes ( n = 60 ) was developed as a class prediction model that could segregate canine iUC into basal or luminal subtypes ( Fig 6A ) . The validation algorithm output predicted tumors with basal and luminal transcriptional subtype with an overall accuracy of 93 . 1% . Basal tumors were predicted with an accuracy of 100% ( n = 11/11 ) , and luminal tumors were predicted with 88 . 9% accuracy ( n = 16/18 ) . Of particular interest are genes also implicated in human iUC including PPARG , CTSF , LY6E , VGLL1 , SERPINE2 , SULT1A1 , and CAV1 ( Fig 6 ) . The PCA plot shows clear segregation of basal and luminal subtypes ( Fig 6 ) [3 , 5 , 6 , 8 , 10 , 17] . When analyzing a list of genes ( n = 595 ) associated with immune profiling of human cancers in the canine iUC samples , immune signatures were enriched in 9 of 11 basal tumors as compared to luminal tumors ( Fig 7A ) [20] . Interferon-γ ( IFN-γ ) inducible genes ( n = 694 ) were also found to be enriched in canine basal tumors ( Fig 7B ) [21] . Genes that are involved in immunosuppressive functions , i . e . , those regulating the function of myeloid derived suppressor cells ( MDSCs ) and regulatory T-cells , and genes identified using GO:0002376 ( immune system processes ) were also found to be enriched in basal tumors . It was noted that two tumors identified as having the basal subtype did exhibit immune cold signatures . Variable number of luminal tumors were found to be enriched for immune signatures ( 1/18 as visualized in 7A , 7B and 7C , and 4/18 as depicted in 7D ) .
The most important outcome of the study was the finding of distinct luminal and basal subtypes in canine iUC . It is becoming widely accepted that these subtypes have critical bearing on the clinical behavior of iUC and response to therapy in humans , and thus must be modeled in pre-clinical animal models [10 , 19 , 22] . Luminal iUC commonly has FGFR3 , ERBB2 , and ERBB3 activating mutations , and is thought to generally be associated with a better prognosis [10 , 23] . Basal iUC is enriched with EGFR and HIF-1 expression , is often metastatic at presentation , and can possess squamous and sarcomatoid histological features and epithelial-to-mesenchymal transition cell biomarkers [10 , 23] . In canine iUC , 62% of tumors were luminal , and 38% were basal . These percentages are similar to those reported in human iUC [10] . It is noted that different reports in the literature have classified human iUC into two to six subtypes including some subclassifications of the two major ( luminal , basal ) subtypes [2 , 5 , 6 , 10 , 17 , 24] . Multiple studies have focused on TCGA data [4 , 10 , 17] . In a 2014 report of TCGA data from the first 131 samples , tumors were classified into four subtypes , referred to as clusters I-IV , with clusters I and II being luminal , and clusters III and IV being basal [4] . In the 2017 report of TCGA data from the larger set of 412 samples , iUCs were categorized into five subtypes including luminal papillary ( similar to previous cluster I ) , luminal infiltrated ( similar to previous cluster II ) , luminal , basal/squamous ( similar to previous clusters III and IV ) , and neuronal subtypes [10] . Within the basal subtype , earlier reports also identified a subset referred to as claudin-low tumors characterized by RB mutations , EGFR amplification , and low expression of FGFR3 and PPARG [6 , 17 , 19] . These tumors had enhanced immune signatures , but more immunosuppressive features ( e . g . immune checkpoints ) than immune enhancing features . Although our study of canine iUC did not reveal subtypes beyond luminal and basal subtypes , the finding of claudin-low and infiltrated features in some of the canine tumors suggest that these subtypes could exist in canine bladder cancer and may be elucidated with larger numbers of cases in follow-up studies . Regardless of the numbers of subtypes in human iUC , there is consensus that the most critical distinction is between luminal and basal subtypes , as occurs in canine iUC [10 , 22 , 24] . The characterization of subtypes is important because the prognosis and treatment response are thought to be impacted by the subtype of the cancer [4 , 10 , 19 , 22] . It is anticipated that future therapies could be tailored to the individual patient based in part , on the subtype of their cancer . It is recognized that basal tumors are inherently more aggressive and are associated with shorter overall survival than luminal tumors , yet basal tumors appear more sensitive to cisplatin-based therapies than luminal tumors . [3 , 4 , 5] . Additional subtypes also could affect treatment outcome . Luminal papillary tumors ( previously included in TCGA cluster I ) which have FGFR3 mutations , fusions and amplifications , but an underactive immune environment may be best treated with FGFR3 inhibitors [10] . Luminal infiltrated tumors ( previously included in TCGA cluster II ) would be expected to respond to immune checkpoint inhibitors , but less so to cisplatin-based regimens [10] . There is evidence for this in that patients with locally advanced and metastatic iUC refractory to platinum chemotherapy were more likely to respond to the anti-PD-L1 immune checkpoint inhibitor , atezolizumab , if they had luminal cluster II tumors [25] . Similarly , in a study in which iUC were assigned to subtypes using the “UNC” classification scheme , luminal tumors had the best overall survival with or without neoadjuvant chemotherapy [8] . Patients with basal tumors appeared to gain the most improvement in overall survival from neoadjuvant cisplatin-based chemotherapy [8] . In human iUC , claudin-low tumors have been reported to have poor overall survival regardless of treatment [8] . Claudin-low tumors are characterized by a stromal phenotype , enrichment of epithelial-to-mesenchymal transition ( EMT ) markers , immune response genes , and lack of luminal differentiation markers [18 , 19] . It has been proposed that claudin-low tumors express high levels of cytokines and chemokines normally repressed by PPARG [19] . The enrichment of immune signatures , in particular PD-CD1 in claudin-low tumors , reinforces the need to stratify these patients to potentially receive immune checkpoint inhibitor treatment as opposed to basal tumors that could respond better to chemotherapy [19] . Canine basal tumors were found to be enriched for p63 pathway genes as reported in human basal transcriptional subtype [3] . Both canine luminal and basal iUC tumors expressed patterns associated with p53 target genes , as previously reported in humans [3 , 5] . Although the p53 phenotype was originally suggested to be a third subtype , this is now being reconsidered [3] . It is recognized that the contribution of stromal cells could have been driving the p53 signature in some of the tumors in earlier analyses [3] . An intriguing finding in the canine iUC samples , was the presence of different RNA-seq immune signatures between the luminal and basal subtypes . Identifying transcriptional immune signatures in tumors offers a tool to complement other methods ( microscopy , immunohistochemistry , flow cytometry , and others ) in analyzing the immune response to cancer and to cancer therapy [22] . There is renewed and growing recognition that the immune response impacts patient outcome , response to immunotherapies , and response to other types of treatments [26] . In melanoma and non-small cell lung cancer , for example , the tumor immune state is recognized for prognostic value and for predictive value in the response to immune checkpoint blockade therapies [27 , 28] . Immunotherapies are rapidly shifting the treatment priorities for bladder cancer , as well as other human cancers [29] . Within the last two years , there has been a dramatic shift in the treatment of metastatic iUC with the approval of five therapies targeting the programmed cell death protein ( PD-1 ) /programmed cell death ligand 1 ( PD-L1 ) axis [30] . As described in other cancers types , it is expected that the pre-existing tumor immune landscape in iUC will impact the response to chemotherapy as well as to immunotherapies . Molecular subtyping of the cancer could allow more precise prognostication , and the selection of therapies most likely to help each individual patient [31] . With the dramatic effects of immune checkpoint inhibitors only occurring in approximately 20% of patients , tools to assist in predicting and monitoring outcomes and in selecting patients most likely to benefit from immunotherapy are critical [30] . In human iUC , immune transcriptomic analyses have revealed different immune signatures of prognostic relevance in the molecular subtypes [22] . These immune signatures were also found in canine iUC . RNA-seq data from the canine iUC samples were interrogated using four panels of immune genes . The nCounter PanCancer Immune Profiling Panel and GO analyses ( GO:0002376 immune system processes ) provide a broad picture of the immune landscape by incorporating several hundred genes from different immune cell types , common checkpoint inhibitors , tumor antigens , and other genes covering both the innate and adaptive immune response [20] . The third panel of genes used to interrogate the immune landscape in canine iUC consisted of 694 IFN-γ inducible genes from the Interferome database [21] . Interferons have a central role in anti-tumor immune responses , and are emerging as prognostic and predictive biomarkers of chemotherapy , as well as immunotherapy [32] . In analyses of human iUC samples , some of the most enriched biological processes included responses to IFN-γ , especially in cluster IV tumors [21] . In the canine samples , notable enrichment in IFN-γ inducible genes was also observed , especially in basal tumors . The fourth panel used consisted of immunosuppressive genes regulating the function of myeloid derived suppressor cells and regulatory T-cells ( compiled from GSEA database ) . Immunosuppressive genes were also found enriched in basal tumors , further supporting the use of immunotherapy in basal iUC . Future work will focus on specific immune features in the canine tumors . Another key finding from the study was the identification of a smaller subset of genes ( 60 genes ) that can be used to categorize canine iUC as luminal or basal . This is similar to the finding of a small subset of 47 genes which could be used to distinguish human luminal and basal subtypes [6] . As larger canine iUC data sets become available , this panel of 60 genes can be further validated , and can expedite further studies . The findings in this study further increase the value of the naturally-occurring canine model of iUC . Dogs with naturally-occurring iUC have previously been appreciated as a model for human iUC because of similarities in histopathology , molecular features , biological behavior ( local invasion , distant metastases in 50% of cases ) , prognostic factors , and response to chemotherapy [11 , 12] . Clinical trials in pet dogs are quite feasible as most dog owners appreciate the access to new therapies for their pet , the opportunity to help generate new knowledge that will benefit other dogs and humans , and the financial assistance that often accompanies canine clinical trials [12] . New therapies and new combinations of therapies can often be evaluated in a frontline setting because there is not a highly effective , well defined or regimented standard of care that must be used in dogs [11 , 12] . In addition , most dog owners are open to allowing cystoscopy before and during treatment to obtain tissue samples for biological endpoints , and many dog owners will allow an autopsy and tissue collection if the dog is euthanized due to declining quality of life from the cancer or comorbid conditions . The work reported here adds to the value of the canine model by demonstrating the presence of luminal and basal subtypes , and suggesting a different immune landscape between the two subtypes . Furthermore , the identification of a canine 60-gene expression signature capable of distinguishing luminal versus basal tumor subtypes provides the opportunity , with further validation , to design a feasible point-of-care cross-species assay suitable for simultaneous parallel comparative oncology investigations in both canine and human bladder cancer patients to further optimize outcomes in both species . In conclusion , this study provides compelling evidence for the presence of luminal and basal subtypes in naturally-occurring canine iUC , a finding that further increases the value and utility of this disease as a highly relevant model for translational research to improve the outcome of humans with iUC . Future work in larger iUC datasets is warranted to investigate additional subtypes , validate the 60-gene panel for distinguishing luminal and basal iUC , correlate subtypes to clinical outcome , and apply the canine iUC model in high impact translational research .
With consent from the owners of dogs with invasive urothelial carcinoma who were undergoing cystoscopy as part of their diagnostic evaluation , a small tissue sample collected during cystoscopy was saved for RNA sequencing analysis . This was approved by the Purdue Animal Care and Use Committee ( Approval Number 1111000169 ) . | Approximately 50% of patients with invasive urinary bladder cancer ( invasive urothelial carcinoma , “iUC” ) die from their cancer . Better therapeutic strategies are essential . A relatively recent important finding in iUC is the presence of molecular subtypes ( luminal and basal subtypes ) . These subtypes affect the cancer aggressiveness and response to particular treatments . To make progress against iUC , these subtypes must be modeled in animal studies which are used to select the most promising new treatments for human trials . It is , however , challenging to replicate these subtypes combined with other key features in experimentally-induced tumors . We performed RNA sequence analyses of naturally-occurring iUC in pet dogs ( with comparison to normal canine bladder and sequencing data from humans ) , and demonstrated that the luminal and basal subtypes are clearly present in canine iUC . This builds on previous findings of the similarities between iUC in dogs and humans in pathology , metastatic behavior , and response to chemotherapies . In conclusion , the findings of our study provide further compelling evidence that canine iUC is a highly relevant model of human iUC for translational research . Canine clinical trials can be used to select the most promising strategies to take into human trials , thus benefiting humans as well as pet dogs . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"sequencing",
"techniques",
"urology",
"medicine",
"and",
"health",
"sciences",
"cancer",
"treatment",
"immunology",
"clinical",
"oncology",
"cancers",
"and",
"neoplasms",
"vertebrates",
"genitourinary",
"tract",
"tumors",
"cancer",
"immunotherapy",
"dogs",
"animals",
"mammals",
"oncology",
"clinical",
"medicine",
"bladder",
"cancer",
"pharmaceutics",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"immunotherapy",
"research",
"and",
"analysis",
"methods",
"cancer",
"chemotherapy",
"gene",
"expression",
"molecular",
"biology",
"immune",
"response",
"chemotherapy",
"eukaryota",
"clinical",
"immunology",
"genetics",
"biology",
"and",
"life",
"sciences",
"drug",
"therapy",
"amniotes",
"organisms"
] | 2018 | Naturally-occurring canine invasive urothelial carcinoma harbors luminal and basal transcriptional subtypes found in human muscle invasive bladder cancer |
The BXD genetic reference population is a recombinant inbred panel descended from crosses between the C57BL/6 ( B6 ) and DBA/2 ( D2 ) strains of mice , which segregate for about 5 million sequence variants . Recently , some of these variants have been established with effects on general metabolic phenotypes such as glucose response and bone strength . Here we phenotype 43 BXD strains and observe they have large variation ( ∼5-fold ) in their spontaneous activity during waking hours . QTL analyses indicate that ∼40% of this variance is attributable to a narrow locus containing the aryl hydrocarbon receptor ( Ahr ) , a basic helix-loop-helix transcription factor with well-established roles in development and xenobiotic metabolism . Strains with the D2 allele of Ahr have reduced gene expression compared to those with the B6 allele , and have significantly higher spontaneous activity . This effect was also observed in B6 mice with a congenic D2 Ahr interval , and in B6 mice with a humanized AHR allele which , like the D2 allele , is expressed much less and has less enzymatic activity than the B6 allele . Ahr is highly conserved in invertebrates , and strikingly inhibition of its orthologs in D . melanogaster and C . elegans ( spineless and ahr-1 ) leads to marked increases in basal activity . In mammals , Ahr has numerous ligands , but most are either non-selective ( e . g . resveratrol ) or highly toxic ( e . g . , 2 , 3 , 7 , 8-tetrachlorodibenzo-p-dioxin ( TCDD ) ) . Thus , we chose to examine a major environmental influence—long term feeding with high fat diet ( HFD ) —to see if the effects of Ahr are dependent on major metabolic differences . Interestingly , while HFD robustly halved movement across all strains , the QTL position and effects of Ahr remained unchanged , indicating that the effects are independent . The highly consistent effects of Ahr on movement indicate that changes in its constitutive activity have a role on spontaneous movement and may influence human behavior .
Recent studies have highlighted the utility of the BXD murine genetic reference population in the study of metabolism [1] . The BXD population consists of ∼100 related strains descended from C57BL/6J ( B6 ) and DBA/2J ( D2 ) [2] , and has wide phenotypic variance in key traits such as blood pressure [3] , body weight , and glucose response [1] caused by ∼5 million sequence variants segregating across the population . Many specific variants have been established as causal of overt phenotypic changes , including SNPs in the aryl hydrocarbon receptor ( Ahr ) mediating TCDD response [4] , missense SNPs in alkaline phosphatase causing impaired vitamin B6 metabolism and bone weakness [1] , and a CNV of glyoxalase 1 causing increased anxiety [5] . However , the full phenotypic consequences of such variants in the BXDs are only partly understood . Ahr is a basic helix-loop-helix ( bHLH ) transcription factor that has been established over the past decades as a key regulator of a variety of processes , including embryonic development [6] , xenobiotic metabolism [7] , immune response and inflammation [8] , and tumorigenesis [9] . In its inactive form , the AHR protein is localized in the cytoplasm with a variety of chaperones , such as HSP90 . When activated , AHR dissociates from its chaperones , dimerizes with the aryl hydrocarbon nuclear translocator ( ARNT ) [10] , and translocates to the nucleus . This complex then induces the transcription of a multitude of target genes [11] . In vertebrates , AHR is constitutively active but can also be activated by endogenous ligands such as kynurenine and dietary compounds such as indirubin [8] , [12] . These ligands can induce diverse transcriptional networks leading to distinct phenotypic outcomes . For example , indirubin and 2 , 3 , 7 , 8-tetrachlorodibenzo-p-dioxin ( TCDD ) are both high-affinity AHR ligands [13] , yet the former is a anti-cancer compound , while the latter is a potent toxin [14] , [15] . Natural variants in Ahr are also known to strongly affect the activity of AHR within and across species [16] , [17] , such as in mice , where C57BL/6J ( B6 ) and DBA/2J ( D2 ) , the parents of the BXDs , have a ∼20-fold difference in AHR activity due to several missense SNPs [18] . Through the systematic measurement of a large population of BXD mice , we observed large variations in spontaneous activity , a large portion of which was attributable to a single genetic locus on chromosome 12 . With the aid of genomic and transcriptomic data , we inferred Ahr as the most likely candidate gene responsible for this effect . Through a cross-species approach , from humanized AHR mice , to heterozygotic Drosophila melanogaster mutants and Caenorhabditis elegans exposed to RNAi , we were able to conclude both that Ahr is the quantitative trait gene ( QTG ) and that in each case , reducing its expression increases movement . Subsequently , we examined the effects of high fat diet ( HFD ) in the BXDs and establish that , while it robustly decreases movement and increases weight , the effects of Ahr remain strong and entirely independent . Together , these data indicate a striking new moonlighting phenotypic role for Ahr in the control of locomotor activity .
We established a colony of 43 BXD strains and designed a basic phenotyping program to examine how two basic metabolic parameters—activity and weight—vary , and how much of this may be attributed to genetic variants in the population . In a previous study , we examined how many metabolic traits vary due to sex in the BXDs [1] , but did not examine basal movement . We thus examined 68 female and 68 male retired breeders from 22 BXDs strains at approximately 20 weeks of age . Animals were placed individually in normal housing cages and spontaneous movement recording over a 48 hour period . Females were slightly more active ( Figure 1A ) , and movement was highly variable for both sexes ( ∼3-fold for ambulations and ∼4-fold for rearing ) , but overall , movement was strongly correlated by strain between males and females ( Figure 1B ) , with the range of across genotypes ( ∼5-fold range ) dramatically outweighing the range across the sexes ( ∼1 . 3-fold ) . To search for the genetic drivers of movement , we designed an enlarged phenotyping program to examine activity in male BXDs using a larger population sample: 43 strains with 5 animals per cohort phenotyped at precisely 23 weeks of age , using the same diet and recording setup . In the expanded data , both rearing and ambulatory activity again varied dramatically across the population—5-fold and 8-fold respectively ( Figure 1C ) . The two aspects of movement were tightly correlated by strain ( Figure 1D ) and highly consistent for all five biological replicates within each strain , yielding high estimates of heritability ( h2 = 0 . 59 for ambulations , and h2 = 0 . 68 for rearing , shown in Figure 1E ) . The BXD population likewise had highly variable body weights across the population ( ∼2-fold range ) with a high estimate of heritability ( h2 = 0 . 74 ) . Surprisingly however , spontaneous activity had no association with body weight or food intake ( Figure 1F ) , indicating the strains' movement is driven primarily by internal motivating factors , rather than by access to food or water , both of which require rearing to reach . Due to the strong heritability and wide and consistent cross-strain variance , we suspected that the movement variation may be linked to quantitative trait loci ( QTLs ) , which could indicate the region ( s ) of the genome causing the genotypically-driven effects . For both measurements , rearing and ambulation , we detected overlapping QTLs , with the narrow-sense peak located on chromosome 12 from 35 . 5 to 37 . 6 Mb , and the broad-sense peak QTL from 30 . 3 to 37 . 6 Mb ( Figure 1G ) . For ambulatory activity , this locus explains 25% of the overall variance , or 1300 counts/hr , and for rearing the same locus explains 41% of the variance , or 400 counts/hr ( Figure 1H ) . While the movement parameters mapped to several suggestive and two significant loci , the significant locus on chromosome 12 was the most striking and consistent ( Figure 2A ) , thus we prioritized it for validation . To establish the effect of the Chr12 locus , we examined a congenic strain of B6 with the D2 locus at the region of interest ( B6 . D2N-Ahrd ) [16] . We sequenced this line and observed it carries a 6 Mb segment of D2 genome on Chr 12 between 34 . 60 and 40 . 48 Mb , while the rest of the genome is B6 ( though several dozen individual SNPs—i . e . spontaneous mutations—are observed elsewhere in the genome , see Materials & Methods ) . Ten males from all three cohorts were then entered into the same phenotyping platform until 23 weeks of age , at which point the movement experiment was performed . As predicted , the congenic line and D2 moved significantly more than the B6 animals , while the congenic line and D2 moved the same amount ( Figure 2B ) , validating the QTL as causative of movement variance . Moreover , these increases matched the calculated effect size from the QTL: ambulatory activity increased by 1500 counts/hr , while rearing activity increased by 380 counts/hr . This analysis validated the QTL as influencing movement , though did not indicate which candidate gene ( s ) cause the effect ( Figure 2C ) . The broad-sense QTL , from 30 . 3 to 37 . 6 Mb , contains 38 genes , including 9 which are under the narrow-sense QTL ( 35 . 5 to 37 . 6 Mb ) . The congenic region , from 34 . 6–40 . 5 Mb , contains 17 genes , 13 of which are within the QTL bounds . We retained all 42 genes ( 38 within the QTL , 4 exclusively in the congenic region ) for subsequent bioinformatic analyses , though with a particular eye for the 9 genes overlapping in the congenic and significant narrow-sense QTL . To select candidate genes for validation experiments , we used several established methods to prioritize candidate gene ( s ) , which are most likely to influence movement [19] . First , the sequence variants were examined for all candidate genes , including within 5 kb of the 3′ or 5′ untranslated regions . 10 of the candidates are identical by descent across all of the BXD strains , making these genes unlikely to be causal for the QTL [20] , including 3 of the 9 priority candidates . For the 32 genes with sequence variants , seven have protein-coding changes: peroxidasin homolog ( Pxdn ) , thyroid peroxidase ( Tpo ) , histone deacetylase 9 ( Hdac9 ) , mesenchyme homeobox 2 ( Meox2 ) , transmembrane proteins 18 and 195 ( Tmem18 and Tmem195 ) , and Ahr . To further rank the candidate genes , we examined the transcriptional variance and regulation in eight diverse microarray datasets . Candidate genes with higher transcript variability , which have strong genetic variants at the same locus as the phenotype , and/or which associate with the phenotype are more likely to be the QTG under the QTL [19] . First , we examined the 42 genes in 8 datasets from 6 tissues . In three tissues—liver , brown adipose ( BAT ) , and quadriceps—measurements were performed in the same BXD animals , which mapped to the movement QTL . Quadriceps and liver were also sampled in the same BXD strains on a high fat diet . The other three tissues—hypothalamus , pituitary , and adrenal—were collected previously and published by other research groups in the same BXD strains in similar conditions ( i . e . age , sex , diet ) [21] . All transcripts were detected in at least one tissue except Slc26a3 ( in the congenic interval ) and Prps1l1 ( in the narrow QTL region ) . We then mapped all transcripts to identify the existence and location of significant expression QTLs ( eQTLs; LRS≥20 , Figure 2D ) . No trans-eQTLs were consistent across more than one dataset , while four genes gave consistent significant cis-eQTLs: acid phosphatase 1 ( Acp1 , in 5 datasets ) , Tmem18 ( in 4 ) , Ahr ( in 3 ) and syntrophin gamma 2 ( Sntg2 , in 3 ) . We then examined the transcript variance of all candidate genes in the eight datasets . Most genes had at significant variability across the strains in each tissue ( 50% have variance ≥1 . 75 fold; Figure 2E ) , with three genes , Sh3yl1 , Prkar2b and Acp1 , being particularly highly variable ( range ≥3 . 0 fold ) in multiple tissues . However , most genes did not covary across the tissues , with only two having particularly consistent expression: Ahr and Acp1 ( Figure 2F ) . We last examined how the expression of each gene associated with movement phenotypes in the BXDs , with particular focus on candidates under the significant QTL and congenic locus ( e . g . Ahr , Sostdc1 , Ispd ) and those with major or consistent transcript variance ( e . g . Acp1 , Prkar2b , and again Ahr ) . Only two genes , Ahr and Acp1 had significant correlations after multiple testing correction ( Figure 2G ) , though several other genes yielded consistent but non-significant correlations ( e . g . Tmem18 ) . Together , these bioinformatic analyses indicated several genes as potentially causative of movement variance , and with one top candidate: Ahr , which was prioritized as the first gene for validation as the quantitative trait gene ( QTG ) , as the other strong candidate gene , Acp1 , was not in the congenic region or under the peak QTL . Ahr is strongly conserved throughout evolution ( Figure 3A ) , and acts as a bHLH-type transcription factor with impact on development and homeostasis in all species [22] . Given this consistency , we examined whether movement regulation may be another conserved physiological process regulated by the gene . In the BXDs , three particular SNPs have been established as causal for differences in Ahr by affecting its enzymatic activity ( A375V ) , ligand binding and the cis-regulatory mechanism ( L471P ) , and protein length ( *805R , which adds 43 amino acids to the C terminus ) [18] , [23] . Strikingly , these three particular variants are conserved in humans , with the most common human allele ( hAHR ) humans matching the D2 allele at all three ( Figure 3B ) [24] . Correspondingly , hAHR enzymatic activity is similar to that of D2 mice [24] , [25] . Moreover , the valine at position 375 ( or 381 in humans ) is unique among mammalian reference genomes to D2 and humans , and is not even found in macaques , which likewise have much higher Ahr activity than humans and D2-type mice [26] , [27] . To examine a potential link between hAHR and activity , we phenotyped transgenic B6 mice with the hAHR allele replacing the murine Ahr [28] . In the same home cage monitoring experiments as those used before , we observed that the humanized animals were significantly more active than their wildtype counterparts ( Figure 3C ) , and again with an increase equal to that of the congenic line ( Figure 2A ) . This finding indicates both that Ahr was properly selected as the QTG , and that the effect may well be conserved cross species . To validate this hypothesis , we looked to lower organisms . The key transcription factor motifs in Ahr ( bHLH and PAS ) are highly conserved in the D . melanogaster ortholog called spineless ( ss ) [29] , and the C . elegans homolog called ahr-1 [30] , thus we hypothesized the regulation of movement may be further conserved to these simpler model organisms . We first examined movement in D . melanogaster , where we crossed the w strain with a loss of function allele ssD115 . 7 [29] , and examined this line as a heterozygous knockout , both in males and females . In both models , ss reduction resulted in a robust ∼25% increase in movement ( Figure 3D ) . Given the conservation to D . melanogaster , we hypothesized that this connection may manifest also in C . elegans . As before , inhibition of ahr-1 by RNAi resulted in a marked and robust increase in activity ( Figure 3E ) . Moreover , the data in C . elegans indicates that the effect of Ahr inhibition on activity is approximately linear , at least within the expression variation tested . Full knockouts in mice , while viable , have poor postnatal survival rates [31] and have dramatically smaller livers ( <50% size [32] ) ; likewise , full knockouts in Drosophila of ss have notable morphological problems [22] . Ahr inhibition robustly increased movement in all models examined , and the effect was observed in the absence of any clearly established Ahr ligand . In Drosophila and C . elegans , the Ahr orthologs are suspected to be exclusively constitutively active [33] . However , this does not indicate whether it is constitutive activity operating in the mice , or if it is an unknown dietary component of the chow diet common to all cohorts , which could influence movement in a ligand-dependent manner . As known Ahr ligands are either known to be non-selective and activate multiple signaling pathways ( e . g . resveratrol or quercetin [34] , [35] ) or are highly toxic and poorly suited for normal physiological studies ( i . e . TCDD ) , we chose to expose BXD cohorts to an environmental component that is known to influence movement: a high fat diet ( HFD ) . We raised males from the same BXD strains , but now on a HFD from 8 until 23 weeks of age at which point we again measured spontaneous activity ( Figure 4A ) . HFD robustly increases body weight by this age , by about 12 grams or ∼35% of body weight in each strain ( Figure 4A ) . Movement is similarly affected , with rearing decreasing by ∼50% ( Figure 4B ) and ambulatory movement by ∼25% . Strikingly , both movement parameters again map to the same locus on Chromosome 12 , indicating the genetic effect of this locus is independent of the dietary influence on movement ( Figure 4C ) . Moreover , the dietary effect is consistent across all strains ( Figure 4D , left ) and is not directly due to the increase in body weight ( Figure 4D , right ) . Thus , while there may be a gene-by-environment effect of diet on movement in the BXDs , it is independent of Ahr , which is equally expressed in both dietary cohorts . As in the CD cohorts , Ahr expression has a strong negative correlation with spontaneous activity in the HFD population ( Figure 4E ) . This independent phenotype both confirms the locus identified using CD cohorts , but also indicates that Ahr can influence movement in mammals independently of incidental dietary or environmental effects .
In this study we characterized 43 strains in the BXDs genetic reference population to assess basal physiological parameters: movement and body weight . Both phenotypes are driven by numerous and complex interactions between genes and environment , and also by neurological/motivational states and by physiological limitations . In the BXDs , body weight and activity are sexually dimorphic , highly variable , and heritable . Surprisingly , body weight and food intake have no impact on standard spontaneous activity in either CD or HFD-fed cohorts despite a major decrease in movement in HFD cohorts . In both dietary groups , we identified a single common QTL causal of ∼25–40% of the variance of movement across the population . Using a congenic line , we confirmed the effect of this locus and set out to establish the causal gene through a bioinformatics approach . By analyzing nine diverse tissues , five in the cohorts phenotyped and three from other published BXD studies , we were able to establish the aryl hydrocarbon receptor ( Ahr ) as the single best candidate gene for mechanistic validation . Ahr is an evolutionary conserved transcription factor involved in development , signal transduction , and metabolism [36] , [37] . Ahr has constitutive activity , but can also be activated by a variety of ligands such as the endogenous metabolite , kynurenine , and a wide variety of environmental chemicals . Moderate to strong ( ∼95% ) reductions in Ahr activity appear to have little negative effect an organism's health or viability , whether in D2 or humanized hAHR mice , or in heterozygotic fly mutants . Through cross-species analysis to Drosophila and C . elegans , we were able to confirm that reducing the expression of this gene consistently leads to an evolutionarily consistent increase in spontaneous movement . After examining the same BXD strains on a HFD , we observed that despite a major decrease in movement in this population , the Ahr QTL is consistent and with a similar effect size , independently of this major environmental perturbation . This conserved effect in BXDs across different environmental conditions indicates a constitutive role for Ahr in the regulation of movement in vertebrates as well . The observation that reduction of Ahr orthologs in invertebrates has a consistent effect on movement furthers this hypothesis , as the Drosophila ortholog ( spineless ) is constitutively active [38] and does not appear to be affected by any exogenous ligands [33] . A large number of AHR polymorphisms have been identified in large and diverse human population studies [39] , [40] , though it remains to be seen if these variants lead to variation in locomotion and/or disposition to exercise in humans as in mice . However , as the movement link is conserved in mice with a humanized Ahr allele , it seems likely that natural variation in Ahr or of its ligands may explain part of the natural variation in human proclivity for activity . Furthermore , while our data indicate constitutive Ahr activation as a regulator of movement , it is conceivable that this role may be further modulated by specific ligands in mice and humans . In combination , our study expands the phenotypic roles of AHR , endowing it with a commanding role in the control of movement that is conserved across evolution .
All animals were communally housed by strain until phenotyping and fed a chow diet ( CD; ( Harlan 2018; 6% kCal/fat , 20% kCal/protein , 74% kCal/carbohydrate ) throughout life after weaning . All BXD strains ( BXD43–103 ) were originally sourced from the vivarium at the University of Tennessee Health Science Center ( Memphis , TN , USA ) then bred for two or more generations until progeny entered the phenotyping colony . Male versus Female Phenotyping: 136 retired breeders ( 68 female , 68 male ) from 22 strains ( male ) or 19 strains ( female; all 19 overlap ) were taken at 20±4 weeks of age from a breeding colony at the EPFL facility and transferred to the phenotyping unit . Males and females were separated for 3+ weeks to ensure pregnant females were not phenotyped . Males and females were phenotyped on separate days , with 10–16 animals entered into the phenotyping program every 2 days . CD Male Phenotyping: 196 male mice from 43 strains of the BXD family were bred at the EPFL facility and transferred to the phenotyping unit at 8 weeks of age . Each cohort was communally housed ( 3–5 animals per cage ) under 12 h light , 12 h dark cycle with ad libitum access to food and water at all times . Animals were in solitary cages only for the movement phenotyping test ( 48 hours ) and were sacrificed 5 weeks after at ∼28 weeks of age . HFD Male Phenotyping: 186 animals from 42 strains ( all but 1 overlapping ) were entered into the colony as before , with HFD starting at 8 weeks of age ( Harlan 06414; 60% kCal/fat , 20% kCal/protein , 20% kCal/carbohydrate ) . Animals were in solitary cages only for the movement phenotyping test ( 48 hours ) and were sacrificed 5 weeks after at ∼28 weeks of age . Congenic AHR mice were purchased from The Jackson Laboratory ( stock number 002921 ) , with animals delivered at 8 weeks of age along with control B6 and control D2 mice . The congenic mice were generated by crossing B6 with D2 , followed by successive backcrossing ( N13 ) to return the B6 genome except in the region of Ahr [16] . The congenic region was genotyped independently to confirm the size of the interval . We then sequenced the DNA of the congenic strain using an Ion Proton PI Chip at 1 . 5× depth ( i . e . ∼15 million 200 bp reads ) and aligned it against the C57BL/6J reference . We confirmed the reported congenic interval is the only region that retains a D2 background . Several dozen SNPs were observed throughout the genome outside the reported Chr12 interval ( ∼34 . 6 to 40 . 5 ) , but were distributed evenly across the chromosomes and consequently represent spontaneous mutations or sequencing and assembly errors , rather than residual D2 genotype . 10 humanized AHR mice were ordered from Taconic ( model 9165 ) , delivered at 8 weeks of age along with B6 controls [25] . The transgenic mice have exons 3–11 ( of 11 ) replaced with hAHR , while exons 1–2 retain the B6 sequence , retaining 2 amino acids unique to B6 not present in hAHR . The key mutations ( 375 , 471 , and the lost stop codon ) are present in the transgenic animal . For tissue collection on CD and HFD BXD cohorts , animals were sacrificed under isoflurane anesthesia and cardiac perfusion after an overnight fast . High fat diet treatment and two day isolation for the recording experiment were considered as having low impact on the animals' welfare , while all other measurements and conditions were considered as having no negative impact . All research was approved by the Swiss cantonal veterinary authorities of Vaud under licenses 2257 . 0 and 2257 . 1 . Home cage monitoring was performed at 23±1 weeks of age for all mice except retired breeders ( 23±4 weeks ) , using a laser detection grid developed by TSE Systems ( Bad Homburg , Germany ) and used in the animals' standard housing cages . The detection grid has two layers: one for detecting X-Y movement ( “ambulations” ) the other for Z movement ( “rearings” ) . Both measurements are technically independent , though the measurements of movement are strongly correlated ( r∼0 . 70 , see Figure 1B ) . Animals were housed individually for the 48-hour experiment starting at about 10am , with the night cycles ( 7pm–7am with 30 minutes of both dawn and dusk ) used for movement calculations [41] . D . melanogaster lines containing a null mutation in the spineless gene , the D . melanogaster ortholog of Ahr , w; ssD115 . 7/TM3 , were obtained courtesy of Ian Duncan's laboratory and passaged for two generations in a standard incubator . This line was crossed with w− and the progeny segregated accordingly ( i . e . w vs . w;ssD115 . 7 and w;TM3 , hb-LacZ vs . wssD115 . 7/TM3 , hb-LacZ ) . Movement was recorded by placing flies in a sealed chamber , tapping the chamber , and recording their movement as they naturally climb towards the top . For the tapping test , 1–2 day old flies were recorded using a standard SLR camera with a Leica macro objective . The experiment was performed four times for each cohort with one minute recordings each , with a “tap” sending the flies to the bottom of the chamber every 10 seconds . The speed with which flies reached the top of the chamber was measured using the Parallel Worm Tracker for MATLAB , which we modified slightly to work with D . melanogaster [42] . This speed was converted into distance by taking the area-under-the-curve ( AUC ) integral of their velocity . C . elegans movement was recorded for 45 seconds at days 2 , 3 , and 4 of adulthood using a Nikon DS-L2/DS-Fi1 camera and controller setup , attached to a computerized Nikon bright field microscope . Seven plates of worms , with 10 worms per plate , were measured in each condition . The movement of worms during this time was calculated by following the worm centroids using the same modified version of the freely-available for the Parallel Worm Tracker as above . R was used for basic analysis of phenotypic data . GeneNetwork ( www . genenetwork . org ) was used for correlation and genetic analyses . The original phenotypes published in this paper and all microarray data generated in these cohorts are available for public analysis or download using the GeneNetwork database ( Species: Mouse , Group: BXD , Type: Adipose mRNA , Liver mRNA , or Muscle mRNA , then select the EPFL datasets ) . The three historical BXD mRNA datasets , for adrenals , pituitary , and hypothalamus , are also available here [43] . Phenotype data were checked for normality using the Shapiro-Wilk test , with a W-value ≥0 . 80 accepted as approximately normal . Heritability was calculated by one-way ANOVA—the aov ( ) function in R—taking the sum of squares of within-strain variance divided by the total sum of squares variance . Dot plots are represented as individual measurements , or mean+SEM depending on the figure panel . Dot plots with error bars ( e . g . Figure 1D ) indicate each dot is a strain average of ∼5 individuals . Individual QTL plots consider a suggestive LRS≥12 and significant LRS≥18 . Large scale QTL plots ( Figure 2D ) use LRS≥20 for significance due to multiple testing . Welch's t-tests were performed for two-way comparisons between phenotype data , as variances were typically unequal in these comparison groups . Student's t-tests were performed for array data , as all data are normally distributed with equal variance . Pearson's r is calculated for correlation plots as no outliers were observed . A p-value of less than 0 . 05 was considered the significance threshold for all analyses , except in QTL mapping when correction for multiple testing was used . All BXD phenotype data can be found on GeneNetwork . org under the “Type: Phenotype” entry then by searching for “Lisp3” . | Using 43 strains from the BXD mouse reference population , we observed a 5-fold difference in spontaneous activity . QTL analysis indicated that ∼40% of this variance is due to the aryl hydrocarbon receptor ( Ahr ) . Ahr is a conserved transcription factor found in nearly all multicellular organisms and implicated in a multitude of functions , ranging across development , liver metabolism , and neuronal health . This gene is highly variant in the BXDs , and strains with the low-active Ahr allele have significantly higher voluntary locomotion . This increase is also observed in independent mouse models , which have reduced Ahr activity , including in transgenic mice with humanized AHR . Furthermore , decreasing Ahr expression in C . elegans and Drosophila causes similar , robust increases in spontaneous movement . This link is independent of major environmental perturbations as well: BXD strains fed high fat diet long-term move only half as much as their chow-fed brethren , yet the effects of Ahr were consistent and equally strong in both dietary cohorts . While Ahr is a highly liganded transcription factor in mammals , these data indicate that modifications to its constitutive activity are sufficient to control movement . However , certain ligands may be able to specifically act on this phenotypic aspect of the gene . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"genome",
"expression",
"analysis",
"animal",
"genetics",
"functional",
"genomics",
"quantitative",
"trait",
"loci",
"population",
"genetics",
"quantitative",
"traits",
"diet",
"mutation",
"nutrition",
"genome",
"analysis",
"trait",
"locus",
"analysis",
"missense",
"mutation",
"mutant",
"strains",
"genetic",
"polymorphism",
"gene",
"expression",
"mammalian",
"genetics",
"genetic",
"loci",
"invertebrate",
"genetics",
"phenotypes",
"heredity",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"computational",
"biology",
"complex",
"traits"
] | 2014 | An Evolutionarily Conserved Role for the Aryl Hydrocarbon Receptor in the Regulation of Movement |
Worldwide the mosquito Aedes aegypti ( L . ) is the principal urban vector of dengue viruses . Currently 2 . 5 billion people are at risk for infection and reduction of Ae . aegypti populations is the most effective means to reduce the risk of transmission . Pyrethroids are used extensively for adult mosquito control , especially during dengue outbreaks . Pyrethroids promote activation and prolong the activation of the voltage gated sodium channel protein ( VGSC ) by interacting with two distinct pyrethroid receptor sites [1] , formed by the interfaces of the transmembrane helix subunit 6 ( S6 ) of domains II and III . Mutations of S6 in domains II and III synergize so that double mutants have higher pyrethroid resistance than mutants in either domain alone . Computer models predict an allosteric interaction between mutations in the two domains . In Ae . aegypti , a Ile1 , 016 mutation in the S6 of domain II was discovered in 2006 and found to be associated with pyrethroid resistance in field populations in Mexico . In 2010 a second mutation , Cys1 , 534 in the S6 of domain III was discovered and also found to be associated with pyrethroid resistance and correlated with the frequency of Ile1 , 016 . A linkage disequilibrium analysis was performed on Ile1 , 016 and Cys1 , 534 in Ae . aegypti collected in Mexico from 2000–2012 to test for statistical associations between S6 in domains II and III in natural populations . We estimated the frequency of the four dilocus haplotypes in 1 , 016 and 1 , 534: Val1 , 016/Phe1 , 534 ( susceptible ) , Val1 , 016/Cys1 , 534 , Ile1 , 016/Phe1 , 534 , and Ile1 , 016/Cys1 , 534 ( resistant ) . The susceptible Val1 , 016/Phe1 , 534 haplotype went from near fixation to extinction and the resistant Ile1 , 016/Cys1 , 534 haplotype increased in all collections from a frequency close to zero to frequencies ranging from 0 . 5–0 . 9 . The Val1 , 016/Cys1 , 534 haplotype increased in all collections until 2008 after which it began to decline as Ile1 , 016/Cys1 , 534 increased . However , the Ile1 , 016/Phe1 , 534 haplotype was rarely detected; it reached a frequency of only 0 . 09 in one collection and subsequently declined . Pyrethroid resistance in the vgsc gene requires the sequential evolution of two mutations . The Ile1 , 016/Phe1 , 534 haplotype appears to have low fitness suggesting that Ile1 , 016 was unlikely to have evolved independently . Instead the Cys1 , 534 mutation evolved first but conferred only a low level of resistance . Ile1 , 016 in S6 of domain II then arose from the Val1 , 016/Cys1 , 534 haplotype and was rapidly selected because double mutants confer higher pyrethroid resistance . This pattern suggests that knowledge of the frequencies of mutations in both S6 in domains II and III are important to predict the potential of a population to evolve kdr . Susceptible populations with high Val1 , 016/Cys1 , 534 frequencies are at high risk for kdr evolution , whereas susceptible populations without either mutation are less likely to evolve high levels of kdr , at least over a 10 year period .
Worldwide Aedes aegypti ( L . ) mosquitoes are the principal urban vectors of dengue , chikungunya , and yellow fever viruses . Approximately 2 . 5 billion people ( 40% of the human population ) currently live with the risk of dengue transmission . In Mexico , Ae . aegypti is the primary vector of the four dengue virus serotypes ( DENV1-4 ) , the causative agents of dengue fever ( DF ) , dengue hemorrhagic fever ( DHF ) and dengue shock syndrome ( DSS ) . Mexico is severely affected by DF , DSS , and DHF because all four dengue serotypes co-occur in most states of Mexico . A recent review of dengue disease in Mexico [2] reported an increase in incidences from 1 . 72 per 100 , 000 in 2000 to 14 . 12 per 100 , 000 in 2011 . Currently the most effective means to reduce dengue transmission by Ae . aegypti is through reduction of larval and adult populations . In Mexico larval reduction is accomplished chiefly through the application of the organophosphate temephos to peridomestic larval breeding sites and through physical source reduction or alteration of potential water-holding containers . Following recommendations of the official Mexican policy for vector control , ( NOM-032-SSA2-2002 ) , pyrethroids were almost exclusively used to control adults in and around homes from 1999 to 2010 . Pyrethroid insecticides prolong the opening of the voltage gated sodium channel protein ( VGSC ) in insect nerves to produce instant paralysis and ‘‘knock-down . ” The α-subunit of VGSC has four repeat domains , labeled I-IV , each of which contains six transmembrane helix segments , S1-S6 . Pyrethroids preferentially bind to the open state of vgsc by interacting with two distinct receptor sites formed by the interfaces of the transmembrane helix S6 of domains II and III , respectively [1] . The original computer modeling studies [3] suggest that simultaneous binding of pyrethroids to S6 in both domains II and III is necessary to efficiently lock sodium channels in the open state . These models also predict that mutations in the S6 of domain III allosterically alter S6 in domain II via a small shift of IIS6 thus establishing a molecular basis for the coevolution of S6 mutations in domains II and III in conditioning pyrethroid resistance . In 2006 we described a mutation , Ile1 , 016 , in the S6 of domain II in Ae . aegypti that is associated with very high knock-down resistance ( kdr ) to the pyrethroid insecticide permethrin in mosquitoes homozygous for this mutation . We examined collections of Ae . aegypti from Mexico during 1996–2009 [4] and found that the overall Ile1 , 016 frequency increased from 0 . 1% in 1996–2000 , to 2%–5% in 2003–2006 , to 38 . 3%–88 . 3% in 2007–2009 depending upon collection location . In 2010 another vgsc mutation was described in the S6 of domain III in Ae . aegypti that was also strongly correlated with kdr and involved a cysteine replacement ( Cys1 , 534Phe ) [5–7] . A general trend in these studies was that Cys1 , 534 frequencies were generally higher and increased more rapidly than Ile1 , 016 frequencies in natural populations . Based upon these observations and on the dual binding model [3] , we analyzed fresly collected DNA from Ae . aegypti for Ile1 , 016 and Cys1 , 534 while DNA previously analyzed for Ile1 , 016 [4] were tested for the presence of Cys1 , 534 . The purpose of this study was to test the hypothesis that mutations in the S6 of domains II and III coevolve in a dependent manner through various allosteric interactions as suggested by computer models [3 , 8] . An analysis of linkage disequilibrium was performed on the two alleles in 1 , 016 ( Val 1 , 016 ( susceptible ) , Ile 1 , 016 ( resistant ) ) and on the two alleles in 1 , 534 ( Phe 1 , 534 ( susceptible ) , Cys1 , 534 ( resistant ) ) to assess whether alleles at 1 , 534 and 1 , 016 evolve independently or in a correlated fashion through epistasis .
Larval mosquitoes were collected from the locations mapped in Fig 1 and listed in Table 1 . At each collection site , we collected immatures from at least 30 different containers in each of three different areas located at least 100 m apart . This included water storage containers and discarded trash containers such as plastic pails , tires , and cans . Larvae were returned to the laboratory where they were reared to adults and then identified to species . The Viva Caucel collection was west of the city of Merida in Yucatán State ( 20 . 9979639° , 089 . 7174611° ) . The Vergel collection was from eastern Merida ( Fig 1 ) ( 20 . 9575694° , -89 . 5886889° ) . Both were collected in 2011 by Universidad Autónoma de Yucatán . DNA was isolated from individual adult mosquitoes by the salt extraction method [9] and suspended in 150 mL of TE buffer ( 10 mM Tris-HCl , 1 mM EDTA , pH 8 . 0 ) . The SNP identification , allele-specific polymerase chain reaction ( PCR ) , melting curve conditions , and genotype readings followed published procedures [6 , 10–12] . The F3 generation of the Viva Caucel and Vergel strains were exposed to 25 μg permethrin ( Chem Service , West Chester , PA ) coated 250 mL Wheaton bottles . In each bottle approximately fifty 3–4 days old mosquitoes were exposed for one hour . Active mosquitoes were transferred to cardboard cups and frozen at -80°C and formed the ‘alive’ group . Knocked down mosquitoes were transferred to a second cardboard cup and placed into an incubator at 28°C and 70% humidity . After four hours , newly recovered mosquitoes were aspirated , frozen and labeled as ‘recovered’ . The mosquitoes that remained inactive were scored as ‘dead’ . There are four potential 1 , 016/1 , 534 dilocus haplotypes: Val1 , 016/Phe1 , 534 ( VF ) , Val1 , 016/Cys1 , 534 ( VC ) , Ile1 , 016/Phe1 , 534 ( IF ) , Ile1 , 016/Cys1 , 534 ( IC ) . The number of times ( Tij ) that an allele at locus i = 1 , 016 appears with an allele at locus j = 1 , 534 was determined by the program LINKDIS [13] . The program then calculated composite disequilibrium frequencies [14] because the phase of alleles at 1 , 016 and 1 , 534 are unknown in double heterozygotes . An unbiased estimate of the composite disequilibrium coefficient Δij [14 , 15] was calculated as: Δij= ( N/ ( N-1 ) ) ( ( Tij/N ) −2pipj ) Where N is the sample size and pi and pj are the frequencies of alleles at locus i = 1 , 016 and locus j = 1 , 534 respectively . Bayesian 95% Highest Density Intervals ( HDI ) around pi and pj were calculated in WinBUGS[16] . A χ2 test was performed to determine if significant disequilibrium exists among all alleles at 1 , 016 and 1 , 534 . The statistic was calculated and summed over all two-allele-interactions [15]: χ[1d . f . ]2=N∑i∑j ( Δij2pipj ) The linkage disequilibrium correlation coefficient Rij [15] is distributed from -1 ( both mutations trans ) to 0 ( 1 , 534 and 1 , 016 mutations occur independently ) , to 1 ( both mutations cis ) and therefore provides a standardized measure of disequilibrium: Rij=Δij/ ( pi ( 1−pi ) +Ci ) ( pj ( 1−pj ) +Cj ) Where the Ci term corrects for departures from Hardy-Weinberg expectations: Ci=Hobs ( i ) −pi2 where Hobs ( i ) is the observed frequency of i homozygotes . Departures from Hardy-Weinberg expectations were also expressed as Wright’s inbreeding coefficient ( FIS ) and calculated as 1- ( Hexp/2p ( 1- p ) ) where Hexp is the observed frequency of heterozygotes . A χ2 test of the hypothesis FIS = 0 with one degree of freedom is: χ[1d . f . ]2=N ( Hexp−Hobs ) 2∑ipi2+ ( ∑ipi2 ) 2−2∑ipi3
The locations of all sampling sites are shown in Fig 1 and the latitude and longitude coordinates are provided in Table 1 . The sample sizes and numbers of the nine dilocus genotypes ( Three 1 , 534 genotypes x Three 1 , 016 genotypes ) are listed in Table 1 . From a total of 615 treated mosquitoes in Viva Caucel , 17 . 6% ( n = 108 ) were scored as alive , 15 . 6% ( n = 96 ) as recovered and 66 . 8% ( n = 411 ) as dead ( Table 2 ) . Genotypes at 1 , 016 and 1 , 534 were identified in 95 randomly chosen individuals from each group . From a total of 337 treated Vergel mosquitoes , 48 . 1% ( n = 162 ) were scored as alive , 20 . 5% ( n = 68 ) as recovered and 31 . 5% ( n = 106 ) as dead . We randomly chose 95 , 68 and 95 Vergel individuals from each group , respectively to obtain the genotypes at 1 , 016 and 1 , 534 ( Table 2 ) . In Viva Caucel , the frequency of the Ile1 , 106 allele was 0 . 746 and the frequency of the Cys1 , 534 allele was 0 . 926 ( Table 3 ) , while in Vergel Ile1 , 016 was at a slightly higher frequency of 0 . 80 while the Cys1 , 534 allele was close to fixation at 0 . 988 . The Ile1 , 106 and Cys1 , 534 alleles were in positive disequilibrium in Viva Caucel , but were only marginally significant in Vergel . Genotypes at the 1 , 016 and 1 , 534 loci were not independent , in agreement with the linkage disequilibrium analysis in Table 4 . Table 5 is a three-way contingency analysis of genotypes at loci 1 , 016 , 1 , 534 and numbers alive or dead individuals in Viva Caucel . Numbers of alive were not independent of genotypes at the 1 , 016 locus; specifically , numbers of alive were significantly greater in Ile1 , 016 homozygous mosquitoes than in heterozygotes or in Val1 , 016 homozygotes . Numbers of alive were also not independent of genotypes at the 1 , 534 locus; specifically , numbers of alive were significantly greater in Cys1 , 534 homozygous mosquitoes than in heterozygotes or in Phe1 , 534 homozygotes . In general , numbers of alive in the Viva Caucel strain were not independent of genotypes at either locus . However , a problem with this analysis is that genotypes at the two loci are not independent . In this and previous studies [10 , 11] , Ile1 , 016 homozygous mosquitoes have the greatest survival , while few , if any heterozygotes or Val1 , 016 homozygotes survive . To evaluate Cys1 , 534 genotypes independently of Ile1 , 016 homozygous mosquitoes , we only compared the three Cys1 , 534 genotypes among Ile1 , 016 heterozygotes and Val1 , 016 homozygotes . A significantly larger proportion of Cys1 , 534 homozygotes survived . Table 5 also shows the contingency analyses of Vergel mosquitoes . Genotypes at the 1 , 016 and 1 , 534 loci were not independent , while they were marginally significant in the linkage disequilibrium analysis in Table 4 . Numbers of alive were not independent of genotypes at the 1 , 016 locus again because numbers of alive were significantly greater in Ile1 , 016 homozygous mosquitoes than in heterozygotes or in Val1 , 016 homozygotes . Numbers of alive were however independent of genotypes at the 1 , 534 locus; specifically because Cys1 , 534 was almost fixed in the Vergel strain . Table 5 also shows the three-way contingency analysis between genotypes at loci 1 , 016 and 1 , 534 and the numbers recovered or dead in Viva Caucel . As in Table 4 , genotypes at the 1 , 016 and 1 , 534 loci were not independent . The numbers of recovered mosquitoes were not independent of genotypes at the 1 , 016 locus; specifically-numbers recovered were significantly greater in Ile1 , 016 homozygous mosquitoes than in heterozygotes or in Val1 , 016 homozygotes . Numbers of recovered were also not independent of genotypes at the 1 , 534 locus; specifically , numbers of alive were significantly greater in Cys1 , 534 homozygous mosquitoes than in heterozygotes or in Phe1 , 534 homozygotes . In general , numbers of recovered in the Viva Caucel strain were heavily dependent on genotypes at both loci . An interesting difference between the two loci is that 32% ( 28/88 ) of Ile1 , 016 heterozygotes recovered while only 3 . 6% ( 1/28 ) of Cys1 , 534 heterozygotes recovered . This difference was significant ( χ2 = 7 . 59 , df = 1 , p-value = 0 . 006 ) . Table 5 also shows the same analysis of recovery but in Vergel mosquitoes . Genotypes at the 1 , 016 and 1 , 534 loci were not independent , while they were marginally significant in the linkage disequilibrium analysis in Table 4 . Numbers of recovered were not independent of genotypes at the 1 , 016 locus , again because numbers of recovered were significantly greater in Ile1 , 016 homozygous mosquitoes than in heterozygotes or in Val1 , 016 homozygotes . However , numbers of recovered were independent of genotypes at the 1 , 534 locus; specifically because Cys1 , 534 was approaching fixation in the Vergel strain . Table 6 contains the frequencies of Ile1 , 016 and Cys1 , 534 and their Bayesian 95% HDI . FIS was significantly greater than zero ( heterozygote deficiency ) in two of the 36 collections where Ile1 , 016 and Val1 , 016 alleles were segregating . In contrast , a significant heterozygote deficiency occurred in eight of the 53 collections where Cys1 , 534 and Phe1 , 534 were segregating and an heterozygote excess occurred in two collections . The frequencies of the Ile1 , 016 and Cys1 , 534 alleles from 1999 to 2012 are plotted in Fig 2 . The Cys1 , 534 allele appears sooner and increases more rapidly than Ile1 , 016 . Only the states of Veracruz and Chiapas had sufficient samples over the years to compare the spatial distributions of Ile1 , 016 and Cys1 , 534 ( Fig 3 ) . It is very clear that Ile1 , 016 and Cys1 , 534 were increasing in frequency much earlier in Veracruz state in eastern Mexico than in Chiapas state in southwestern Mexico . It is also clear that in both states Cys1 , 534 was increasing in frequency much earlier than in Ile1 , 016 . Starting in 2002 , the frequency of Cys1 , 534 was greater than or equal to that of Ile1 , 016 . In a yearly comparison of Ae . aegypti collection sites , 80 out of 87 sites ( Table 6 ) had a frequency of Cys 1 , 534 being greater than the frequency of Ile1 , 016 . In 6 of the 7 cases where the frequency of Ile1 , 016 exceeded that of Cys1 , 534 , the difference was only from 1–2% and values were not different ( overlapping 95% HDI ) . Only in Martınez de la Torre in 2002 was there a credible difference of 9% . Linkage disequilibrium analysis can only be performed in datasets where alleles are segregating at both loci . There were 34 datasets that met this criteria of the 87 collections listed in Table 1 . Table 7 lists the state , city and year of the 34 datasets along with linkage disequilibrium correlation coefficient Rij and its associated χ2 values and the probability of a greater χ2 . Ile1 , 016 and Cys1 , 534 were in disequilibrium in the majority ( 21/34 = 62% ) of datasets . For the most part , alleles in 1 , 534 and 1 , 016 were evolving in a correlated , dependent fashion . However , this analysis does not provide specific information about the four haplotypes . The frequencies of the four potential dilocus haplotypes are plotted by year in Fig 4 . The frequency of the susceptible Val1 , 016/Phe1 , 534 ( VF ) haplotype remained high from 1999–2003 ( Fig 4A ) . No collections were made again until 2008 , by which time frequencies had dropped to 0–0 . 6 . Four years later , VF was approaching extinction in all collections . Fig 4B plots the frequency of the Val1 , 016/Cys1 , 534 ( VC ) haplotype . From 1999–2003 , VC frequencies remained low ( 0–0 . 10 ) . By 2008 , frequencies had increased to 0 . 1–0 . 75 . Four years later , VC was declining in frequency in two collections and was increasing in four collections . A very different trajectory occurred for Ile1 , 016/Phe1 , 534 ( IF ) ( Fig 4C ) . From 1999–2002 , the IF frequency remained low and only reached as high as 0 . 1 in two collections . By 2008 frequencies were approaching extinction and four years later similar trends were seen , even though VC and IC frequencies had increased dramatically . Fig 4D is a plot of the frequency of the resistant Ile1 , 016/Cys1 , 534 ( IC ) haplotype . From 1999–2002 , the IC frequency was low and only reached 0 . 1 in one collection . By 2008 frequencies had increased dramatically in all collections and continued to increase in all collections up to 2012 when frequencies ranged from 0 . 5–0 . 9 .
The frequency of Cys1 , 534 has increased dramatically in the last decade in several states in Mexico including Nuevo Leon in the north , Veracruz on the central Atlantic Coast , and Chiapas , Quintana Roo and Yucatan in the south . The linkage disequilibrium analysis on the Ile1 , 016 and Cys1 , 534 alleles in Ae . aegypti collected in Mexico from 2000–2012 ( Table 7 ) strongly supports statistical associations between 1 , 534 and 1 , 016 mutations in natural populations . Furthermore , the dynamics of haplotype frequencies during that time suggest pyrethroid resistance in the vgsc gene requires the sequential evolution of 1 , 534 and 1 , 016 mutations . Fig 4C suggests that the Ile1 , 016/Phe1 , 534 haplotype has a low fitness , even when pyrethroids are being released . For this reason Ile1 , 016 is unlikely to have evolved independently . Instead it is much more likely that the Cys1 , 534 mutation evolved first but conferred only a low level of resistance . This conjecture is strongly supported by the fact that in 80 of 87 collections ( 92% ) , the frequency of Cys1 , 534 was greater than the frequency of Ile1 , 016 . The findings of this study are different in many respects from those in a study of a Tyr1 , 575 substitution in Anopheles gambiae that occurs just beyond the S6 of domain III , within the linker between domains III and IV [17] . This linker contains a sequence of three amino acids ( IFM ) that close the sodium channel pore following activation , block the influx of sodium into the cell and restore the membrane resting potential . In contrast , Cys1 , 534 in Ae . aegypti occurs in the S6 of domain III . This is close to a Met1 , 524Ile substitution that has been associated with knockdown resistance in Drosophila melanogaster [18] and a Phe1 , 538Ile mutation that reduces sensitivity to deltamethrin in arthropods and mammals [19 , 20] . Mutations in S6 of domain II , such as Phe1 , 014 , Ser1 , 014 in An . gambie and Ile1 , 016 and Gly1 , 016 in Ae . aegypti are not directly in the binding pocket , but affect the resistance phenotype by preventing binding of insecticides and changing the conformation of the VGSC [3 , 21] . In contrast , a binding site located in a hydrophobic cavity delimited by the IIS4-S5 linker and the IIS5/IIIS6 helices has recently been proposed [22] that it is accessible to the lipid bilayer and lipid-soluble insecticides . The methyl-cyclopropane ( or equivalent structure ) of pyrethroids and the trichloromethyl group of DDT appear to be critical features for the action of both pyrethroids and DDT . Both insecticides fit into a slot in a small pocket in the main hydrophobic cavity , flanked by Val1 , 529 and Phe1 , 530 on IIIS6 . The binding site is formed upon opening of the sodium channel and is consistent with observations that pyrethroids bind preferentially to open channels . This binding pocket includes several known mutations in the S6 of domain III that reduce sensitivity to pyrethroids . Two nearby residues ( Gly1 , 535 and Phe1 , 538 ) have been previously implicated in resistance in other insect species ( 23 ) . Study in which An . gambiae mosquitos were collected from a range of approximately 2000 km throughout West/Central Africa and had Tyr1 , 575 occurring at frequencies up to 30% in both M and S forms . Even though this mutation is seen over a large range of the continent , only a single Tyr1 , 575 haplotype occurred with a Phe1 , 014 haplotype background ( possibly analogous in function to Ile , 1016 ) , which infers strong positive selection acting on a recent mutant [17] . In contrast to the present study , Phe1 , 014 is almost fixed in West Africa and the Tyr1 , 575 allele is increasing in frequency in M form but not in S form . Thus in contrast to the apparent evolution of Ile1 , 016 on a Cys1 , 534 background as reported here in An . gambiae , Tyr1 , 575 appears to have evolved on a Phe1 , 014 background . There are many potential reasons for this difference including the possibility that mutations within the S6 of domain III may produce a different resistance mechanism and have a different impact on fitness than mutations in the linker between domains III and IV . It is also possible that the specific changes of amino acids at these sites are unique and may confer different resistance phenotypes . In either case it seems likely that one of the mutations compensates for deleterious fitness effects of the other mutation and/or confers additional resistance to insecticides . An interesting difference between the two mutations in the present study is that 32% of Ile1 , 016 heterozygotes recover from pyrethroid exposure but only 3 . 6% of Cys1 , 534 heterozygotes recover . Thus while Cys1 , 534 in synergy with Ile1 , 016 may confer greater survival following pyrethroid exposure , Ile1 , 016 may confer a greater ability to recover following knockdown in heterozygotes . There was evidence of heterozygote deficiency in eight of the 53 collections and the average FIS among these eight collections was large and positive ( 0 . 580 ) while the average among all collections was 0 . 052 . This suggests that the fitness of Phe1 , 534 and Cys1 , 534 homozygotes may be greater than the fitness of G/T heterozygotes ( i . e . underdominance ) . While these parameters have been estimated at the 1 , 016 locus [23] , no similar studies have involved the 1 , 534 locus and so the stability point beyond which the frequency of either allele would increase has not been determined . Since the Cys1 , 534 confers some degree of pyrethroid resistance ( Tables 2–5 ) , directional selection could increase the frequency of Cys1 , 534 beyond the underdominance stability point , at which stage the frequency of Cys1 , 534 would rapidly increase towards fixation . Little is known of other mutations in the Ae . aegypti vgsc that may affect pyrethroid resistance . Codon 989 in the “super-kdr” region of domain II was assessed and no mutations were found [11] . Ile , Met and Val alleles occur at codon 1 , 011 [11] but these alleles were not associated with resistance in our initial survey of 1 , 318 mosquitoes from the 32 strains throughout Latin America [11] . The recombination dynamics of the Ae . aegypti vgsc are also poorly understood . Analysis of segregation between alleles at the 1 , 011 and 1 , 016 codons in F3 showed a high rate of recombination even though the two codons are only separated by a approximately 250 bp intron [11] . A maximum parsimony phylogeny of the intron spanning exons 20 and 21 in 88 mosquitoes with different genotypes in exons 1 , 011 and 1 , 016 indicated the presence of three clades with bootstrap support > 80% . These were arbitrarily labelled clades 1–3 . The frequencies of Ile1 , 011 , Met1 , 011 , Val1 , 011 , Val1 , 016 , Ile1 , 016 and Gly1 , 016 ( from Thailand only ) were then compared among the three clades . The frequency of Ile1 , 011 was distributed independently among the three clades , as was Val1 , 011 and Met1 , 011 . However , there was a very evident excess of Val1 , 016 alleles in clade 1 and an excess of Ile1 , 016 alleles in clade 2 . Ile1 , 016 alleles occurred in disequilibrium with a large number of segregating sites in the intron and a large excess of Ile1 , 016 alleles were found to be associated with clade 2 in the phylogenetic analysis . This pattern is consistent with a hypothesis that a genetic sweep of the Ile1 , 016 allele and proximate intron sequences has occurred through DDT exposure and subsequently pyrethroid selection . Furthermore , the genetic sweep was recent enough that there has been insufficient time for recombination to disrupt the disequilibrium between the Ile1 , 016 allele and proximate intron sequences . Recent work on the dual binding model may shed some light on the next steps in the evolution of pyrethroid resistance in the vgsc [8] . The Tyr1 , 575 mutation in An . gambiae was introduced alone into an Ae . aegypti sodium channel ( AaNav1-1 ) [8] and then in combination with Phe1 , 014 . Both substitutions were then functionally examined in Xenopus oocytes [8] . Tyr1 , 575 alone did not alter AaNav1-1 sensitivity to pyrethroids . However , the Tyr1575- Phe1014 double mutant was more resistant to pyrethroids than the Phe1014 mutant channel alone . Further mutational analysis showed that Tyr1 , 575 could also synergize the effect of Ser1 , 014 and Trp1 , 014 , but not Gly1 , 014 , or other pyrethroid-resistant mutations in subunit 6 of domains I or II . Computer modeling predicted that Tyr1 , 575 allosterically alters pyrethroid binding via a small shift of the subunit 6 of domain II . This establishes a molecular basis for the coexistence of Tyr1 , 575 with Phe1 , 014 in pyrethroid resistance , and suggests an allosteric interaction between IIS6 and Loop III/IV in the sodium channel . The rapid increase in Cys1 , 534 ( Fig 4B and 4D ) cannot be the result of neutral forces such as genetic drift or founder’s effects . Parallel increases in Cys1 , 534 frequency occurred throughout Mexico . Even though the forces that caused an increase in the frequency of Cys1 , 534 are unclear , our results suggest that Ile1 , 016 in domain IIS6 arose from a Val1 , 016/Cys1 , 534 haplotype and was rapidly selected possibly because double mutants confer higher pyrethroid resistance . When combined with Phe1014 , the Tyr1 , 575 mutation in An . gambiae increased resistance to permethrin and deltamethrin by 9 . 8- and 3 . 4-fold , respectively [8] . Fig 5 illustrates two models for the evolution of 1 , 534 and 1 , 016 mutations . Model 1 proposes that the 1 , 534 and 1 , 016 mutations occurred independently and became cis by crossing over . Model 2 instead proposes that 1 , 534 mutations occurred first because 1 , 016 mutations confer low fitness . Ile1 , 016 mutations then arose on a Val1 , 016/Cys1 , 534 background . These results suggest that knowledge of the frequencies of both 1 , 534 and 1 , 016 mutations are important to predict the potential of a population to evolve kdr . Obviously , the frequency of Ile1 , 016 by itself is a poor predictor ( Fig 4C ) . Populations that are pyrethroid susceptible , but have high Val1 , 016/Cys1 , 534 frequencies , are at high risk for rapid kdr evolution . If our experience in tracking the frequencies of Ile 1 , 016 and Cys1 , 534 mutations over the past 15 years can be extended to other Ae . aegypti populations , then populations with intermediate to high frequencies of Cys1 , 534 might only be susceptible for 5–10 years . Conversely , pyrethroid susceptible populations without either mutation are unlikely to develop kdr quickly and might be susceptible for up to 10–15 years . | Constant use of pyrethroid insecticides has driven mosquito populations to develop resistance . In Aedes aegypti , the primary mosquito vector of dengue , yellow Fever , and chikungunya viruses , pyrethroid resistance is primarily associated with mutations in the voltage-gated sodium channel protein . One mutation occurs in codon 1 , 016 and involves a replacement of valine with isoleucine ( Ile1 , 016 ) , and a second located in subunit 6 of domain III in codon 1 , 534 , replaces phenylalanine with cysteine ( Cys1 , 534 ) . In Mexico , we found that Cys1 , 534 was present in the same mosquito collections that were previously analyzed for Ile1 , 016 . In this study , we performed a linkage disequilibrium analysis on both Ile1 , 016 and Cys1 , 534 in Mexican collections from 2000–2012 . Our analysis suggests that pyrethroid resistance requires the sequential evolution of the two mutations and that Cys1 , 534 must occur first and appears to enable the Ile1 , 016 mutation to survive . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Coevolution of the Ile1,016 and Cys1,534 Mutations in the Voltage Gated Sodium Channel Gene of Aedes aegypti in Mexico |
Malaria parasites must undergo a round of sexual reproduction in the blood meal of a mosquito vector to be transmitted between hosts . Developing a transmission-blocking intervention to prevent parasites from mating is a major goal of biomedicine , but its effectiveness could be compromised if parasites can compensate by simply adjusting their sex allocation strategies . Recently , the application of evolutionary theory for sex allocation has been supported by experiments demonstrating that malaria parasites adjust their sex ratios in response to infection genetic diversity , precisely as predicted . Theory also predicts that parasites should adjust sex allocation in response to host immunity . Whilst data are supportive , the assumptions underlying this prediction – that host immune responses have differential effects on the mating ability of males and females – have not yet been tested . Here , we combine experimental work with theoretical models in order to investigate whether the development and fertility of male and female parasites is affected by innate immune factors and develop new theory to predict how parasites' sex allocation strategies should evolve in response to the observed effects . Specifically , we demonstrate that reactive nitrogen species impair gametogenesis of males only , but reduce the fertility of both male and female gametes . In contrast , tumour necrosis factor-α does not influence gametogenesis in either sex but impairs zygote development . Therefore , our experiments demonstrate that immune factors have complex effects on each sex , ranging from reducing the ability of gametocytes to develop into gametes , to affecting the viability of offspring . We incorporate these results into theory to predict how the evolutionary trajectories of parasite sex ratio strategies are shaped by sex differences in gamete production , fertility and offspring development . We show that medical interventions targeting offspring development are more likely to be ‘evolution-proof’ than interventions directed at killing males or females . Given the drive to develop medical interventions that interfere with parasite mating , our data and theoretical models have important implications .
Malaria parasites are obliged to undertake a single round of sexual reproduction in the mosquito vector before they can transmit to new hosts , making this stage of their life-cycle a potential target for medical interventions [1] , [2] . The success of interventions aiming to disrupt mating success will depend upon a variety of epidemiological parameters ( e . g . transmission intensity/seasonality ) , but will also be strongly determined by the parasites' behavioural and evolutionary responses [1]–[3] . Current candidates for transmission-blocking vaccines ( TBV ) involve targeting proteins , expressed on the surface of sexual stages , that are essential for the fertility of males ( e . g . P48/45 and P230 ) [4]–[8] . However , theory predicts that the efficacy of a vaccine that reduces the fertility of one sex may be eroded if parasites respond by adjusting their sex ratios in favour of the targeted sex . The study of sex allocation has been one of the most successful areas of evolutionary biology , with empirical data matching clear theoretical predictions across a variety of taxa [9] . Before describing evolutionary theory for sex allocation strategies we outline the relevant aspects of Plasmodium mating biology . Every asexual replication cycle , a small proportion of parasites differentiate into male and female sexual stages – termed gametocytes – which are developmentally arrested gamete precursors [10] , [11] . Gametogenesis of both sexes begins as soon as gametocytes are taken up in a mosquito blood meal , fertilization occurs within 30 minutes , and zygotes develop into the stages infective to vectors ( ookinetes ) after 18–20 hours [12] , [13] . To differentiate into gametes , gametocytes must leave the relative safety of their red blood cells ( RBCs ) , becoming exposed to host- and mosquito-derived factors that can block mating [12] . Males are expected to be more vulnerable than females to transmission-blocking factors due to their more complex gametogenesis and mating activities [14] , [15] . Whereas female gametocytes only have to leave their RBCs to become gametes , male gametogenesis also includes three rounds of mitosis and flagellum construction to produce a ( rarely achieved ) maximum of eight ‘sperm-like’ gametes [16]–[20] . Mature male and female gametocytes are easily distinguished by their phenotypes as their reproductive roles result in different cellular contents [21] , [22] . Mature males are terminally differentiated , only having pre-synthesized proteins and machinery for gamete production ( e . g . α-tubulin II , cell cycle proteins , dynein ) [11] , [22] , [23] . In contrast , mature female gametocytes are prepared for continued development after fertilization , having high levels of ribosomal proteins , mitochondria ( which are absent in mature males ) and pools of translationally repressed messenger RNAs ( mRNAs; similar to P bodies in metazoan oocytes ) [11] , [22] , [24] . Therefore , male and female gametocytes are primed for gametogenesis and zygote development , respectively [25] . Sex allocation is an important fitness-related trait in Plasmodium and could play an important role in the response of malaria parasites to medical interventions that aim to reduce mating success [19] , [26]–[28] . Parasites could respond to transmission-blocking interventions by adjusting their sex allocation strategies via two evolutionary processes . First , if conditions within hosts are unpredictable , invariant , or if variation in within-host conditions is not a good proxy for variation in the mating conditions experienced within vectors , parasites evolve fixed ( i . e . canalised ) sex allocation strategies that reflect the average environment . Second , if in-host conditions reliably predict in-vector conditions , parasites will evolve to facultatively adjust their sex ratios ( proportion of male gametocytes ) through phenotypic plasticity . In this scenario , if asexual stage parasites detect an increase in a factor ( or correlate of ) that reduces mating ability in a sex-specific way , parasites will benefit from adjusting the production of male and female gametocytes in response . Given that once parasites are taken up by a vector , no further gametocyte production can occur and gametogenesis and fertilization are completed within 30 minutes , the mating environment within the blood meal is ‘imported’ from the host . Therefore , the within-host conditions will be good predictors for mating conditions and so facultative sex ratio adjustment is both predicted and observed [14] . Currently , two complementary evolutionary theories predict how and why parasites should adjust their investment into male and female gametocytes to maximise fertilization success . These theories – Fertility Insurance and Local Mate Competition – predict that parasites adjust sex ratios in response to environmental ( e . g . transmission-blocking immunity ) and social factors ( inbreeding rate ) , respectively [14] , [15] , [29]–[34] . The ability of parasites to facultatively adjust their sex ratios in response to variation in the inbreeding rate has recently been verified [19] , [27] . Additionally , data also suggest that sex ratios are altered in response to the development of immunity [19] . Host-derived immune factors make mating challenging for parasites because they can reduce and even block fertilization [35] , [36] . This phenomenon , called ‘transmission-blocking immunity’ ( TBI ) , has been extensively observed and documented across a variety of malaria parasite species [35]–[41] . The mechanisms of TBI are varied and include damaging gametocytes , preventing successful gametogenesis [36] , [37] , [41] , [42] , decreasing the ability of gametes to interact [35] , [43] and preventing post-fertilization development [39] , [44] . Fertility Insurance predicts that when hosts mount an immune response , the fertility of male gametocytes and/or gametes is most affected , therefore parasites should produce more males to compensate [14] , [15] . Two lines of empirical data support this prediction . First , Paul et al . [26] showed that P . gallinaceum and P . vinckei increase their sex ratio in response to erythropoiesis , which is thought to act as a cue for the appearance of TBI factors . Second , Reece et al . [19] provided indirect support by suggesting that sex ratio variation observed during infections of different P . chabaudi genotypes is a mechanism to ensure fertility in face of within-host competition , host anaemia and TBI factors . Fertility Insurance currently provides the best explanation for the observed within-infection variation in the sex ratios of malaria parasites . However , the theory is based upon the untested assumption that TBI factors reduce the fertility of males more than females . Here we provide the first direct test of this key assumption by investigating whether reactive nitrogen species and pro-inflammatory cytokines , influence gametogenesis , gamete fertility and ookinete production . Levels of reactive nitrogen species ( RNS ) and pro-inflammatory cytokines vary during malaria infections . These immune factors , which are ubiquitous components of the innate immune system , have been specifically implied in the sudden loss of infectivity to vectors that occurs during paroxysms and infection crisis [37] , [41] . Specifically , tumour necrosis factor-α ( TNF-α ) is a potent pro-inflammatory cytokine and several studies have revealed a role for this cytokine in mediating the killing of Plasmodium gametocytes , across a variety of host-parasite systems [36] , [41] , [45] . This could occur through the stimulation of phagocytosis and nitric oxide ( NO ) production by white blood cells [37] , [46] , [47] , as these are capable of phagocytosing opsonized gametes in the mosquito midgut [48] and the inhibition of NO synthesis by white blood cells reduces in 60% the inactivation of P . falciparum and P . vivax gametocytes [37] , [49] . NO is produced by the enzyme inducible nitric oxide synthase in response to infection , in both hosts and vectors , and is extremely toxic at high doses . NO is a highly reactive molecule , thus a significant extent of the damage it causes is indirect , through the production of RNS ( such as peroxynitrite , nitrates , nitrites or S-nitrosothiols ) that frequently function as the ultimate effectors [50] . Hereafter , unless otherwise stated , we use the term ‘RNS’ to refer to NO and its reaction products . During Plasmodium infections , RNS appears to impair asexual replication , gametogenesis and zygote development [37] , [42] , [44] , [51] . Levels of RNS increase during P . yoelii infections and reduce ookinete production when either gametocytes or gametes are exposed [42] . Furthermore , RNS have been shown to induce the programmed cell death of P . berghei ookinetes [52] and to extensively reduce P . berghei oocyst burdens in Anopheles mosquitoes [44] . This is , at least in part , the result of a pro-inflammatory response , in which host cytokines induce the mosquito to increase NO ( and therefore RNS ) production [53] . Here , we use the rodent malaria parasite Plasmodium berghei to conduct a series of experiments to investigate how RNS and TNF-α influence mating success and ookinete production and develop theoretical models that predict the evolution of sex allocation strategies , given the effects observed in our experiments . Therefore , we use these immune manipulations as ‘proof-of-principle’ for other factors with similar effects on the sexual reproduction and transmission of malaria parasites . Specifically , we test whether: ( 1 ) RNS and TNF-α have dose dependent effects on male gametogenesis ( exflagellation ) and ookinete production; ( 2 ) exposure of male and female gametocytes to both RNS and TNF-α influences their sexual development; ( 3 ) the greater effect of RNS we observe on male gametogenesis results in sex-specific fertility effects; and ( 4 ) the observed effects of RNS depend on the developmental stage at which parasites are exposed . Our results reveal that RNS reduces male but not female gametogenesis and impairs the fertility of both sexes , whereas TNF-α only affects zygote development . The relative importance of reduced gametogenesis , impaired mating ability and reduced post-mating development have not been explicitly considered by Fertility Insurance theory . Therefore we develop a new mathematical model to derive predictions for how the effects of immune factors generated naturally or by a medical intervention are likely to impact upon parasite sex ratio evolution ( a schematic of the biological effects included in the model is presented in Figure 1 ) .
We first tested whether RNS and TNF-α influence sexual reproduction by exposing parasites to different concentrations of these factors and assaying exflagellation and ookinete production . We incubated parasites in vector mimicking media across seven concentrations of SIN-1 ( ranging from 0 to 1 mg/ml ) [55] and five concentrations of recombinant mouse TNF-α ( from 0 to 1 µg/ml; see Methods ) . Increasing concentrations of SIN-1 caused a significant linear decrease in the densities of exflagellating males ( F ( 1 , 35 ) = 16 . 28 , P<0 . 0001; transformed y = 0 . 16-0 . 10x ) and ookinetes ( F ( 1 , 35 ) = 25 . 86 , P<0 . 0001; transformed y = 0 . 17-0 . 18x ) . Similarly , TNF-α also caused a significant linear decrease in the densities of exflagellating males ( F ( 1 , 15 ) = 6 . 83 , P = 0 . 012; y = 0 . 23-0 . 09x ) and ookinetes ( F ( 1 , 15 ) = 17 . 53 , P<0 . 0001 ; transformed y = 0 . 54-0 . 37x ) . Having found significant negative effects of RNS and TNF-α on exflagellation and ookinete production we investigated whether these factors interacted with each other to further reduce parasite mating success and if these effects depended on the developmental stage at which parasites were exposed ( i . e . in host or vector conditions ) . For this set of experiments we used a fully cross-factored design , consisting of two RNS and two TNF-α levels ( see Methods ) . First , we investigated the effects of RNS and TNF-α on gametocytes by incubating parasites for 60 minutes in host mimicking media . We then replaced treatment media with vector mimicking media ( without RNS or TNF-α manipulations ) to stimulate gametogenesis and quantified the development of male and female gametocytes into gametes using the following classifications: ( a ) mature gametocytes still inside their RBC , ( b ) gametocytes that had emerged from the RBC and ( c ) exflagellating male gametes ( see Methods for criteria ) . We present the proportion of a given developmental stage relative to the total number of observed gametocytes/gametes of the same sex ( Figure 2 ) . The proportion of emerged female gametocytes was not significantly influenced by either RNS ( χ21 = 2 . 72 , P = 0 . 099 ) or TNF-α ( χ21 = 0 . 12 , P = 0 . 731; or their interaction χ21 = 3 . 38 , P = 0 . 066 ) . In contrast , the proportion of male gametocytes that emerged from RBCs was significantly reduced by RNS ( F ( 1 , 59 ) = 81 . 29; P<0 . 0001; mean ‘RNS−’ 0 . 55±0 . 02; ‘RNS+’ 0 . 32±0 . 02 ) but not by TNF-α ( χ21 = 0 . 16 , P = 0 . 689; or their interaction χ21<0 . 01 , P = 0 . 982 ) . Similarly , the ability of males to exflagellate was significantly reduced by RNS ( F ( 1 , 59 ) = 33 . 40; P<0 . 0001; mean ‘RNS−’ 0 . 15±0 . 01; ‘RNS+’ 0 . 09±0 . 01 ) but not by TNF-α ( χ21 = 0 . 85 , P = 0 . 36; or their interaction χ21 = 0 . 02 , P = 0 . 885 ) . Second , we investigated the effects of RNS and TNF-α on exflagellation and ookinete production by incubating parasites in culture media mimicking the vector environment ( Figure 3 ) . In line with the results from our previous experiments , the proportion of exflagellating males was significantly reduced by RNS ( F ( 1 , 45 ) = 11 . 24 , P = 0 . 002; mean ‘RNS−’ 0 . 32±0 . 06; ‘RNS+’ 0 . 12±0 . 03 ) . This effect was enhanced by TNF-α ( interaction: F ( 1 , 45 ) = 6 . 67 , P = 0 . 014 ) but in the absence of RNS , TNF-α had no significant effect ( F ( 1 , 45 ) = 1 . 90 , P = 0 . 175 ) . Conversely , the effect of RNS and TNF-α on ookinete production depended on each others presence ( interaction F ( 1 , 24 ) = 14 . 91 , P = 0 . 001 ) . Specifically , ookinete production was reduced by TNF-α but only in the absence of RNS ( mean ‘TNF-α−’ 0 . 41±0 . 06; ‘TNF-α+’ 0 . 17±0 . 07 ) , whereas RNS reduced ookinete production but only when TNF-α was absent ( mean ‘RNS−’ 0 . 41±0 . 06; ‘RNS+’ 0 . 09±0 . 05 ) . Experiment 2 revealed that only RNS had a significant effect on gametogenesis , in which male but not female development was impaired . Therefore , we tested whether these effects translated into sex-specific differences in fertility ( i . e . whether matings with RNS exposed gametocytes/gametes resulted in fewer ookinetes ) , when parasites were exposed as gametocytes ( in host-mimicking media ) or during gametogenesis ( in vector-mimicking media ) . We separately exposed each sex to RNS using two genetically transformed ( knock-out; KO ) P . berghei lines: Pbs48/45ko and Pbs47ko [4] , [6] , [22] , which produce unviable male and female gametes , respectively . This allowed us to assay the fertility consequences of exposing one sex to RNS by providing exposed parasites with a surplus of unexposed mates from the opposite sex and assaying ookinete production ( Figure 4 ) . We observed that RNS exposure significantly reduced fertility of both males and females regardless of whether parasites were exposed as gametocytes or during gametogenesis ( F ( 1 , 131 ) = 15 . 87 , P = 0 . 0001; mean ‘RNS−’ 0 . 30±0 . 02; ‘RNS+’ 0 . 20±0 . 02 ) . In contrast to our predictions , RNS did not have sex-specific effects ( treatment:sex interaction: χ21 = 0 . 023 , P = 0 . 88 ) , nor was this effect influenced by exposing parasites to RNS in host- or vector-mimicking environments ( treatment:environment interaction: χ21 = 0 . 366 , P = 0 . 55 ) . However , across all treatments , parasites exposed in host conditions produced significantly more ookinetes than those exposed in vector conditions ( F ( 1 , 131 ) = 10 . 19 , P = 0 . 0018; mean ‘Host’ 0 . 29±0 . 02; ‘Vector’ 0 . 21±0 . 02 ) . We incorporate our experimental results into Fertility Insurance theory by developing a mathematical model to explore the impact of transmission-blocking factors on the evolution of parasite sex allocation strategies . Specifically , we examine whether sex ratio adjustment could compensate for transmission-blocking factors with the following effects on males or females: preventing male or female gametocytes from undergoing gametogenesis ( as each female gametocyte only produces one gamete , killing of these stages is mathematically equivalent ) ; blocking the mating ability of male gametes; and causing damage to gametocytes or gametes such that mating can occur but zygotes are not viable . We term the latter phenomenon , of cryptic damage to gametocytes or gametes that results in a dead zygote , as dysfunction . Note that , although we do not observe all of the effects on all stages and all sexes , we incorporate them all in the model ( illustrated in Figure 1 ) , as they are theoretical possibilities . Also , our model makes no assumptions about whether parasites evolve fixed ( i . e . canalised ) or facultative ( i . e . plastic ) sex allocation strategies . First , we show that all zygote mortality effects ( i . e . treatments leading to 0<p<1 ) have no impact on the evolutionarily stable ( ES ) sex ratio [56] , [57] . We write W = ζ ( z ) p , i . e . fitness is the product of zygote production and zygote viability , where zygote production depends upon sex ratio but zygote viability does not . The direction of selection is given by the derivative of fitness with respect to sex ratio [58] , and this ‘marginal fitness’ is dW/dz = ( dζ/dz ) p . The ES sex ratio z* satisfies dW/dz|z = z* = 0 , i . e . selection does not favour an increase or decrease in sex ratio when the population is at the ES sex ratio , and this is equivalent to the condition dζ/dz|z = z* = 0 for all p>0 . Since ζ is not a function of p , it follows that z* is not a function of p ( and hence is not a function of ΩZ , ΩM , ΩF , ϖM or ϖF; see Methods and Figure 1 for symbol definitions ) . Therefore , treatments that simply impact upon the viability of zygotes ( e . g . cause gametocyte/gamete dysfunction ) are not expected to have an evolutionary impact upon parasite sex ratios . Second , to investigate the impact of model parameters arising from gametocyte or gamete killing on the ES sex ratio , we write an explicit expression for expected fitness: ( 1 ) The condition dW/dz|z = z* = 0 can be solved numerically for z* for any numerical parameter set ( q , dM , dF , δM ) . An exploration of the ES sex ratio z* across this parameter space is presented in Figures 5 and S1 , S2 , S3 . Specifically , we recover the prediction that the gametocyte ES sex ratio will be biased towards the more limiting sex when factors prevent male or female gametocytes from undergoing gametogenesis or block the mating ability of male gametes .
In our experiments , RNS reduced male but not female gametogenesis while impairing the fertility of both sexes . How can these results be explained ? In parasitic infections , high levels of RNS may cause: oxidative damage of DNA ( leading to mutations and strand brakes ) ; inhibition of DNA repair and synthesis; inhibition of protein synthesis; inhibition of mitochondrial activity; down- or up-regulation of cytokine ( e . g . TNF-α ) levels [50] , [59] . As described in the introduction , male and female gametocytes are prepared for gametogenesis and zygote development respectively [25] . If RNS can impair DNA synthesis and/or microtubule assembly , males would not be able to produce gametes . In contrast , female gametogenesis does not involve these activities and females ‘simply’ need to leave their RBCs , for which they use the contents of pre-synthesized secretory organelles called osmiophilic bodies [60] . Therefore , whilst female gametogenesis and mating per se is unlikely to be influenced by RNS , the development of fertilized females into zygotes and ookinetes is likely to be affected . For example , damage to stored mRNA and inhibition of protein synthesis or mitochondrial activity ( e . g . cytochrome oxidases ) would impair meiosis ( at ∼3 h after fertilization ) and zygote development , but not impair fertilization [18] , [50] , [59] . These effects could explain the observed results , because instead of reducing the ability of females to differentiate into gametes , the effects of RNS would be expressed after fertilization ( which we term dysfunction ) and lead to female-dependent zygote death , resulting in fewer ookinetes . Here we did not identify the causal RNS and their relative contributions . However , this will be important if transmission-blocking interventions cause or mimic the activities of RNS . Our experiments show that TNF-α consistently reduces ookinete production and whilst we observed a reduction in exflagellation in some experiments , this effect was inconsistent . Why does TNF-α reduce ookinete production ? As TNF-α functions are mainly modulatory and need time to develop , it is possible that gametogenesis and mating occur before the effects of TNF-α manifest . Ookinete development takes about 18–20 hours from fertilization and during this time TNF-α could exert its effects , which could also involve the activation of apoptotic-like death [61] , [62] . Recent experiments provide support for our interpretations , as the deletion of genes coding for proteins essential for the storage and stabilization of translationally repressed mRNAs , in female gametocytes/gametes , do not reduce fertilization success , but substantially reduce the differentiation of zygotes into ookinetes [24] , [63] . Interestingly , deletion of different genes can affect zygotes throughout development , suggesting that damage to stored mRNA could abort zygote development at multiple stages ( e . g . before or after meiosis ) [24] . The results of our experiments show that TBI factors can affect the sexual development and fertility of male and female parasites and that the stage at which this occurs is sex-specific . As illustrated in Figure 1 , we incorporated the observed and potential effects of transmission-blocking factors on males and females , at all stages of development , into Fertility Insurance theory and generated new predictions for the evolution of parasite sex allocation strategies . Our model predicts that the ES gametocyte sex ratio will be insensitive to variation in gametocyte or gamete dysfunction and zygote mortality . This means that treatments that impact upon the viability of zygotes are not expected to have an evolutionary impact upon parasite sex ratios . In contrast , we predict that the best ( ES ) sex ratio strategy will vary depending on an interaction between gametocyte group size ( q ) , number of gametes formed per male gametocyte ( 0≤χ≤8 ) and gamete and/or gametocyte mortality . Although , our model makes no assumptions about whether parasites achieve an ES sex ratio through the evolution of facultative or fixed sex allocation strategies , facultative sex allocation is predicted for reasons already outlined in the introduction . In the context of clonal infections , the ES sex ratio maximises the expected number of viable zygotes , i . e . maximises the expected number of gametes of the minority sex present in the mating pool ( this excludes dead gametocytes/gametes , but includes dysfunctional gametocytes/gametes ) . For an infinite gametocyte group size ( i . e . q→∞ ) , that behaves deterministically , the ES sex ratio is one that leads to the same number of male and female gametes being present in the mating pool . This is the sex ratio z* that satisfies cz* = 1-z* , i . e . z* = 1/ ( c+1 ) , where c is the number of male gametes , able to mate , produced per male gametocyte [15] , [32] . Thus , the ES sex ratio is female biased if c>1 , and male biased if c<1 ( Figures 5 and S1 , S2 , S3 ) . However , for finite mating groups ( q<∞ ) – that behave stochastically – the expectation of mating success must be calculated over the whole distribution of possible outcomes . This will tend to reduce the extent to which the sex ratio is biased towards the sex favoured in the deterministic case [15] , [31] . For example , in the extreme of a gametocyte group size of two ( q = 2; the lowest mating group size for which mating success is possible ) , the ES sex ratio is always z* = 0 . 5 ( regardless of other parameter values ) , to maximise the probability of both sexes being present ( Figures 5 and S1 , S2 , S3 ) . Additionally , we reveal that , in a small portion of parameter space – corresponding to very small gametocyte group sizes , low female mortality , and high male gametocyte mortality and fecundity ( χ ) – fertility insurance can even lead to a sex ratio bias in the opposite direction ( i . e . producing a female biased sex ratio , despite the risk of the absence of males in the mating pool; Figures S2 and S3 ) . This non-intuitive result is due to the way stochastic variation in the number of gametocytes of each sex alters the variance as well as the expected number of gametes of each sex that reach the mating pool . Although the conditions under which this occurs are restrictive , they may be met in natural infections , as many individuals carry gametocytes at extremely low densities [64] . In the context of our experiments and assuming parasites can facultatively adjust sex ratios , our model predicts that if q is high enough to allow for sex ratio adjustment , then RNS should induce parasites to increase the production of male gametocytes . Our data suggest that RNS reduced female fertility by rendering gametocyte/gametes dysfunctional , so that their fertilisation results in the production of unviable zygotes . The reduction in ookinete production by TNF-α could also be due to male or female dysfunction or , more likely , through increasing zygote mortality . Therefore , we examined the influence of gametocyte and gamete dysfunction and zygote mortality on the evolution of parasite sex allocation strategies . We found that the ES gametocyte sex ratio is independent of these factors ( i . e . the occurrence of gametocyte/gamete dysfunction and zygote mortality does not change the relative fitness of different sex ratio strategies ) . Put simply , this suggests that zygote mortality or gametocyte/gamete dysfunction will not impose selection on parasite sex allocation strategies as parasites cannot compensate for the loss of reproductive success through sex ratio adjustment . More broadly , other immune factors , such as antibodies or complement , could also impair the sexual reproduction of malaria parasites and the effects of such factors should be easily interpreted in light of our theoretical models . To bring our mathematical modelling in line with our experiments we have focused on the importance of mortality and dysfunction throughout the sexual development of malaria parasites . However two additional factors have an important impact in sex allocation strategies of malaria parasites: ( 1 ) the inbreeding rate and ( 2 ) the rate at which asexually replicating parasites commit to gametocyte production ( conversion rate ) . The effect of inbreeding on the ES sex ratio is well understood , with theory ( Local Mate Competition ) enjoying strong empirical support [14] , [19] , [29] , [32]–[34] . For clonal mating groups , the ES sex ratio strategy is the one that maximises the overall mating success of the infection as the parasites behave as a single , unified decision maker [14] , [27] . In contrast , in mixed infections , conflicts between clones occur , such that the ES sex ratio is the one that maximises each individual clone's inclusive fitness and not the overall mating success of the infection [14] , [27] . But for the work we present here , extending our model to allow for a finite number of independent clones per host would not change the qualitative results we present . Fertility Insurance theory predicts that if a low conversion rate results in a small number of gametocytes being taken up by the vector ( i . e . small q ) , parasites should produce a less female biased sex ratio than expected by the inbreeding rate alone . This is due to the stochastic risk of too few males being present in the blood meal to fertilize the females when sex ratios are female biased [15] . One intuitive solution for this would be to produce more gametocytes . However , given that gametocyte production comes at a cost to asexual replication , parasites face a trade-off between investment in in-host survival and reproduction ( i . e . transmission ) . Increasing gametocyte conversion is a solution that will not always be available and might be impossible when parasites are ‘stressed’ ( e . g . by in-host competition and low doses of anti-malarial drugs ) [65] , [66] . Therefore , if transmission-blocking interventions also affect asexual stages and reduce in-host survival , parasites are likely to reduce conversion rates and produce fewer gametocytes . Our model reveals that an intervention with a sex-specific effect on mating ability will elicit an evolutionary response . However , sex ratio adjustment cannot fully rescue zygote production , given that an increase in the number of male gametocytes comes at the cost of decreasing the number of female gametocytes . Nevertheless , in a scenario of widespread transmission-blocking vaccination or treatment with gametocidal drugs with a sex-specific effect , natural selection will “compare” the fitness of parasites that do , and do not , adjust their sex allocation strategies , leading to an increase in the frequencies of the former . Therefore , quantifying the impact of sex ratio adjustment on rescuing fertility and thus , fitness is now required . In contrast , our model also reveals that a transmission-blocking factor resulting in zygote mortality or gametocyte/gamete dysfunction will be ‘evolution proof’ with respect to parasite sex allocation strategies . Therefore , we suggest that current efforts to prevent fertilization by targeting proteins with sex-specific phenotypes , such as P230 , P48/45 ( involved in gamete attachment ) or Pfg377 ( female emergence from the RBC ) , will be less effective than vaccines targeting zygote development ( e . g . P28 ) [5] , [60] , [67] . An alternative transmission-blocking approach could cause dysfunctional female gametes by targeting the expression of female-specific translationally repressed mRNAs [24] . Furthermore , a transmission-blocking intervention combining targets for gamete dysfunction and zygote death would minimize possible redundancy effects , which have been observed in several knock-outs of malaria parasites ( e . g . P48/45 ) [6] . Given the drive to develop transmission-blocking interventions that disrupt sexual reproduction in malaria parasites , there is an urgent need to evaluate how their short- and long-term success will be influenced by parasite mating strategies . Here , we combined experiments with mathematical modelling to predict how transmission-blocking factors influence parasite sex allocation strategies . Our model predicts that transmission-blocking interventions causing gametocyte/gamete dysfunction and/or zygote mortality will be ‘evolution-proof’ from the perspective of imposing selection on parasite sex ratio strategies , i . e . parasites may still evolve other strategies or traits to cope with a transmission-blocking intervention , but these will have to be independent of sex allocation . Put simply , understanding the behavioural strategies that parasites have evolved to cope with naturally occurring transmission-blocking immune factors , will inform predictions for how they will respond to a transmission-blocking factor . More broadly , understanding how , when and why parasites respond to changes in their in-host environment will facilitate the development of interventions that induce parasites to make decisions that are suboptimal for their transmission success , but that are clinically or epidemiologically beneficial . For efficient progress , synergy between research directed at evolutionary and mechanistic explanations for parasite traits and strategies is required .
We maintained MF1 mice , aged 8–10 weeks ( Harlan-Olac , UK; or in house supplier , University of Edinburgh ) , on ad libitum food ( RM3 ( P ) , DBM Scotland Ltd , UK ) and water ( supplemented with 0 . 05% PABA to enhance parasite growth ) , with a 12 hour light cycle , at 21°C . We initiated infections by intra-peritoneal inoculation of 107 parasitized RBCs in 100 µl carrier consisting of 50% Ringers ( 27 mM KCl , 27 mM CaCl2 , 0 . 15 M NaCl ) , 47 . 5% heat-inactivated foetal bovine serum and 2 . 5% heparin ( 5 units ml−1 ) . For experiments 1 and 2 , we inoculated female mice , previously ( day −3 or −4 ) treated with 60 mg/kg of phenylhydrazine ( PHZ ) , with P . berghei line 820 [68] . For experiment 3 we inoculated male mice ( PHZ treatment: 125 mg/Kg , day −2 ) with one of two P . berghei KO lines: Pbs48/45ko or Pbs47ko [4] , [6] , [22] . We treated mice with PHZ because the resulting release of young RBCs increases gametocyte production in P . berghei , which maximises the number of gametocytes that can be harvested for in vitro mating experiments [69] . For each experiment , parasites were collected from mice on day 3 or 4 post-infection , and each infection contributed parasites to all treatments to control for any potentially confounding influences of differences between infections . All the protocols involving mice passed an ethical review process and were approved by the U . K . Home Office ( Project License 60/3481 ) . Work was carried according to the Animals ( Scientific Procedures ) Act , 1986 . In order to manipulate the levels of RNS and TNF-α we used the following chemicals: recombinant mouse TNF-α ( Sigma , UK ) , L-ana ( Sigma , UK ) and SIN-1 ( Sigma , UK ) . We dissolved all chemicals in phospate buffered saline and exposed parasites to treatments in 1 ml cultures with 15 or 20 µl parasitized blood . L-ana is a specific inhibitor of the activity of the enzyme inducible nitric oxide synthase which becomes active in response to infection . SIN-1 donates NO and/or superoxide , in solution , at different rates depending on the specific conditions in which SIN-1 is incubated [54] , [70] , [71] . However , given that superoxide and NO react with each other at an extremely fast rate to produce peroxynitrite ( ONOO− ) , SIN-1 is likely to act as a donor of either NO or peroxynitrite , depending on the rates at which SIN-1 generates NO and superoxide [54] . The oxygen concentration of the solution is one of the major determinants of whether SIN-1 behaves as a NO or peroxynitrite donor , donating mostly NO in anaerobic conditions and peroxynitrite in aerobic conditions [54] . In our cultures , oxygen concentrations were in-between fully anaerobic and aerobic conditions , as parasites were incubated in closed 1 . 5 ml tubes . Biological agents , such as human plasma or heme proteins , which are similar to components of our cultures ( e . g . mouse plasma , haemoglobin ) increase the capacity of SIN-1 to donate NO [54] . Furthermore , as peroxynitrite can react to produce several RNS ( e . g . nitrite , nitrate , S nitrosothiols or nitrosyl-metal complexes ) and as we did not measure the specific contributions of each of these factors , we use the term RNS to refer to the factors present in cultures exposed to SIN-1 [50] , [61] , [72] . We did not measure RNS and TNF-α levels in our cultures for three reasons . First , our focus is on testing the effects of RNS and TNF-α on the sexual development of parasites . As our experiments were designed so that each host contributed blood and parasites to all treatment groups in a given experiment , this controls for any variation between infections and ensures that our results are due to the RNS and TNF-α manipulations each culture was subjected to . Second , TNF-α levels were directly manipulated with recombinant mouse TNF-α . Third , we are not aware of any method that would allow us to measure total levels of the different RNS in small volume cultures . We set up cultures with vector mimicking media for the following SIN-1 concentrations: 0 , 0 . 00001 , 0 . 0001 , 0 . 001 , 0 . 01 , 0 . 1 and 1 mg/ml [55] , with 6 mice contributing parasites to each treatment . We tested the following concentrations of recombinant mouse TNF-α: 0 , 0 . 005 , 0 . 01 , 0 . 5 and 1 µg/ml with 4 mice contributing parasites to each treatment . We recorded the densities of exflagellating males after 15–20 minutes and ookinetes after 18–20 hours using a haemocytometer . We used the following RNS and TNF-α levels: 1 mg/ml SIN-1 ( RNS+ ) , 1 mg/ml of L-ana ( RNS− ) , and presence ( TNF-α+ ) or absence ( TNF-α− ) of 1 µg/ml recombinant mouse TNF-α . Parasites from each of 20 mice were exposed to all four combinations of treatments . We used the following criteria to classify developmental stages of gametogenesis after 15 minutes incubation in vector mimicking media: ( 1 ) Mature gametocytes: still inside their RBC; females have blue-purple cytoplasm , small , well defined purple nucleus surrounded by a small nucleolus; males have pink-yellow cytoplasm and large disperse pale-pink nucleus . ( 2 ) Emerged females: female gamete condensed into a more circular shape , without a vacuole , cytoplasm staining a more intense blue and a less obvious nucleolus than in a female gametocyte . ( 3 ) Emerged male: male gamete with a large circular nucleus in the centre of the cell surrounded by a ring of cytoplasm . ( 4 ) Exflagellating male: emerged male gamete progressed to forming up to 8 flagella that protrude from the cell and stain red-purple [73]–[75] . We also recorded the densities of exflagellating males and ookinetes as described for experiment 1 . We infected 38 mice with Pbs47ko ( n = 19 ) or Pbs48/45ko ( n = 19 ) . We set up mating cultures following Reece et al . [19] , by pairing infections according to proximity of their sex ratios , calculated from the densities of Pbs48/45ko female gametocytes in giemsa stained smears ( using criteria described for Experiment 2 ) and Pbs47ko exflagellating males ( as for Experiment 1 ) . To avoid pseudo-replication , each infection was only used in 1 pair . For each pair of mice , we made 8 sets of 1 ml cultures , either with ( RNS+ ) or without ( RNS− ) 1×10−5 mg/ml SIN-1 , mimicking host ( 60 min . incubation ) or vector conditions ( 15 min . incubation ) , to which we added 15 µl of parasites from one of the infections in each pair . These single sex cultures provided ‘exposed’ parasites for fertility testing , and corresponded to the following factorial design: 2 conditions ( host/vector ) ×2 SIN-1 exposures ( RNS+/− ) ×2 sexes ( male/female ) . After incubation we replaced media in all cultures with 1 ml vector mimicking media ( without any SIN-1 manipulation ) . While ‘exposed’ parasites were incubating , we collected 60 µl of blood from each infection's pair and added these ‘unexposed’ parasites to 4 ml cultures in vector mimicking media ( without SIN-1 ) . Each 1 ml culture of the ‘exposed’ parasites was then added to a 4 ml culture containing its ‘unexposed’ pair and incubated to produce ookinetes ( as for Experiment 1 ) . This allowed us to ensure that the mating success of the ‘exposed’ sex would not be limited by the availability of ‘unexposed’ gametocytes from the opposite sex . All the cultures were timed so that ‘exposed’ parasites were added to the cultures containing their ‘unexposed’ mates at the same developmental stage . For example , a final 5 ml culture could contain 15 µl of blood from a RNS exposed Pbs48/45ko infection ( in which females are the ‘exposed’ sex ) and 60 µl of blood from a Pbs47ko infection ( in which ∼4 times more males are provided as ‘unexposed’ mates ) . We also set up cultures in vector mimicking media to verify that ‘unexposed’ parasites from each line are unable to produce ookinetes on their own . We recorded the densities of ookinetes as described for experiment 1 . We used linear mixed effects models ( R version 2 . 7 . 0; The R Foundation for Statistical Computing; www . R-project . org ) because , by treating each infection ( or pair of infections in Experiment 3 ) as a ‘random’ effect , we can account for problems associated with pseudoreplication arising from repeated measurements of each infection . In order to meet the assumptions made by parametric tests we arcsine square root transformed response variables where necessary . We minimised models following stepwise deletion of the least significant term and using log-likelihood ratio ( χ2 ) tests to evaluate the change in model deviance until only significant terms remained , and we present F-ratios for fixed effects remaining in minimal models . We then re-ran minimal models using restricted maximum likelihood to estimate the effect sizes reported in the text . Unless otherwise indicated , data and estimated effect sizes are presented as proportions of the focal parasite stage produced in a given treatment , relative to that produced across all treatments for each infection . We assume an infinite host population , divided into infected and uninfected individuals , with infected hosts containing a single infection producing haploid gametocytes that circulate in the blood . We assume that q gametocytes are transferred from host to vector during blood feeding , and that these gametocytes form a single mating group . The expected proportion of males in the mating group is z , i . e . the sex allocation strategy of the parasite strain that contributed the gametocytes . Hence , the actual number of males is a random variable α∼Bi ( q , z ) ( i . e . binomially distributed with q trials and probability of success z ) . Consequently , the number of female gametocytes is q-α . Male and female gametocytes are killed with probability dM and dF respectively , leaving Γ∼Bi ( α , 1-dM ) surviving males and φ∼Bi ( q-α , 1-dF ) surviving females . We assume every surviving male produces χ gametes , and every surviving female produces a single gamete . We consider that male gametes are killed with probability δM , and hence γ∼Bi ( χΓ , 1-δM ) male gametes enter the mating pool . We assume that all φ female gametes enter the mating pool ( death of female gametes is formally equivalent to that of female gametocytes , and hence is implicitly included in the parameter dF ) . Therefore , the number of zygotes is equal to the number of gametes of the limiting sex , i . e . ζ = min ( γ , φ ) . Finally , we assume that only a proportion p of zygotes are viable , due to either: ( a ) factors that kill each zygote with probability ΩZ; ( b ) factors acting on gametocytes resulting in the production of dysfunctional gametes at rate ΩM for males and ΩF for females; or ( c ) factors acting on gametes and causing them to become dysfunctional at rate ϖM for males and ϖF for females , i . e . p = ( 1-ΩZ ) ( 1-ΩM ) ( 1-ΩF ) ( 1-ϖM ) ( 1-ϖF ) . In this context , we use the term ‘dysfunctional’ to refer to a gamete that achieves fertilisation but carries sufficient damage to render the resulting zygote inviable ( i . e . unable to develop as an ookinete ) . Inviable zygotes will result when one or both of the parental gametes are dysfunctional . Hence , the number of viable zygotes produced by the mating group is W = ζ p , and this is our measure of fitness [15] , [31] , [32] . | Malaria and related parasites cause some of the most serious infectious diseases of humans , domestic animals and wildlife . To be transmitted , these parasites produce male and female sexual stages that differentiate into gametes and mate when taken up in a mosquito blood meal . Despite the need to develop a transmission-blocking intervention , remarkably little is understood about the evolution of parasite mating strategies . However , recent research demonstrates that producing the right ratio of male to female stages is central to mating success . Evolutionary theory predicts that sex ratios are adjusted in line with a variety of factors that affect mating success , including host immunity . We test this theory by investigating whether ubiquitous immune factors differentially affect the production and fertility of males and females . Our experiments demonstrate that immune factors have complex , sex-specific effects , from reducing gamete production to affecting offspring viability . We use these results to generate theory predicting how such effects shape the evolutionary trajectories of parasite sex ratio strategies . Given the drive to develop medical interventions that prevent transmission by blocking parasite mating , our results have important implications . Specifically , we suggest that medical interventions targeting offspring development are more likely to be ‘evolution-proof’ than interventions with sex-specific effects . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"evolutionary",
"biology/evolutionary",
"ecology",
"immunology/innate",
"immunity",
"immunology",
"infectious",
"diseases/protozoal",
"infections",
"evolutionary",
"biology",
"immunology/immunity",
"to",
"infections"
] | 2011 | Sex and Death: The Effects of Innate Immune Factors on the Sexual Reproduction of Malaria Parasites |
Campylobacter jejuni is one of the leading infectious causes of food-borne illness around the world . Its ability to persistently colonize the intestinal tract of a broad range of hosts , including food-producing animals , is central to its epidemiology since most infections are due to the consumption of contaminated food products . Using a highly saturated transposon insertion library combined with next-generation sequencing and a mouse model of infection , we have carried out a comprehensive genome-wide analysis of the fitness determinants for growth in vitro and in vivo of a highly pathogenic strain of C . jejuni . A comparison of the C . jejuni requirements to colonize the mouse intestine with those necessary to grow in different culture media in vitro , combined with isotopologue profiling and metabolic flow analysis , allowed us to identify its metabolic requirements to establish infection , including the ability to acquire certain nutrients , metabolize specific substrates , or maintain intracellular ion homeostasis . This comprehensive analysis has identified metabolic pathways that could provide the basis for the development of novel strategies to prevent C . jejuni colonization of food-producing animals or to treat human infections .
Campylobacter jejuni subsp . jejuni ( C . jejuni ) is one of the most common causes of infectious food-borne illness in industrialized countries [1 , 2] . The high incidence of this pathogen is due to its ability to persistently colonize the intestinal tract of food-producing animals . Contaminated food products—in particular , poultry meat—become a source of C . jejuni infection when improperly handled or undercooked [3] . While asymptomatic in most vertebrates , in humans , C . jejuni infection often leads to acute , although self-limiting , gastroenteritis [4] . Rarely , infections with C . jejuni lead to a sequelae known as Guillain-Barre syndrome , which is characterized as a serious neurodegenerative disorder [5] . A characteristic feature of C . jejuni that distinguishes it from other common enteropathogenic bacteria is the paucity of homologs of virulence factors that in other pathogens are engaged in specific interaction with the host [6 , 7] . In fact , C . jejuni has arguably more in common with commensal intestinal microbiota than with enteric pathogens . This is consistent with the observation that , other than in humans , the persistent presence of C . jejuni in the gut does not lead to pathology [8] . Why and how C . jejuni infection in humans leads to disease is very poorly understood , but it is expected that its ability to colonize and replicate within the intestinal tract to reach significant numbers must be central to its pathogenesis . Several studies have identified genes that are important for C . jejuni intestinal colonization using different animal models of infection . Other than genes required for motility or the modification of surface structures ( e . g . , protein glycosylation ) , the vast majority of genes identified to date as required for colonization are involved in the acquisition and metabolism of essential nutrients [7 , 9–12] . Therefore , deciphering the metabolic requirements of C . jejuni is central to the understanding of its ability to colonize a host and potentially cause disease . In fact , the understanding of the metabolism of bacterial pathogens during infection is quickly emerging as an extremely important area of research . While the basic metabolism of model bacteria during their growth in vitro has been extensively studied , knowledge of the metabolic requirements of bacterial pathogens during infection has lagged behind [13] . Indeed , it is becoming increasingly clear that the functional purpose of some virulence factors that specifically target host processes is to increase the availability of crucial nutrients or to facilitate their acquisition for bacterial growth . Therefore , the concept of “nutritional virulence factors” has been proposed to describe such pathogenic determinants [14–17] . Although most of the C . jejuni colonization determinants known to date have been discovered by candidate-mutant experimental approaches , there have been some attempts to comprehensively identify such determinants using genome-wide approaches [18–20] . However , because of technical limitations in the approaches or animal models used , those studies have not been comprehensive . The availability of high-throughput nucleotide sequencing technologies coupled to transposon mutagenesis has provided a powerful tool to interrogate highly saturated mutant libraries of insertion mutants for specific phenotypes [21–23] . This approach allows not only the comprehensive identification of virulence or colonization determinants but also the simultaneous measurement of the relative fitness cost resulting from the inactivation of potentially every nonessential gene under various environmental conditions . When applied to the understanding of metabolic requirements , such comparative analysis can provide a much more encompassing view of the relative importance of specific metabolic pathways under the conditions examined . An essential prerequisite for the application of high-throughput approaches to interrogate genomic libraries is the absence of bottlenecks that could limit the depth of coverage of the mutagenesis screen . Since in most animal models of infection C . jejuni cannot be recovered in large numbers [24] , this has been a major limitation in previous attempts to broadly interrogate mutant libraries for their ability to colonize the intestine [19 , 25] . Recently , a mouse model of infection has been described that allows the replication of C . jejuni to large numbers [26] , thereby overcoming previous limitations for the application of high-throughput genome-wide analyses . We report here the use of this animal model in combination with transposon mutagenesis and next-generation sequencing to comprehensively interrogate a highly saturated transposon insertion library of C . jejuni 81–176 for its ability to colonize the intestine . To provide a more robust framework for the interpretation of these results , we have also examined the mutant library after growth under different in vitro conditions . The combination of this extensive genetic approach with isotopologue profiling , metabolic flow examination , and the analysis of specific mutant strains amounted to the most complete analyses of the metabolic determinants necessary for a bacterial pathogen to colonize a mammalian host to date . Importantly , this analysis has identified key metabolic pathways that could be targeted for the development of novel strategies to prevent C . jejuni infection .
We recently reported the construction of a highly saturated and randomly dispersed transposon mutant library composed of approximately 50 , 000 individual insertions with an average density of 31 insertions/kb across the C . jejuni 81–176 genome [27] . Analysis of the library indicated that 90% of the 1 , 758 predicted open reading frames in the genome and its plasmids pTet and pVir harbored transposon insertions , whereas the remaining 10% of the genes were presumably essential under the conditions used to generate the library [27] . We used this transposon mutant library to interrogate the fitness of the different C . jejuni mutants after growth in solid rich ( blood agar ) or defined liquid medium supplemented with asparagine ( Asn ) , glutamine ( Gln ) , or serine ( Ser ) as the main carbon sources ( Fig 1 ) . Comparison of the relative number of each transposon mutant remaining after 48 hours of growth under the different conditions with the number in the original inoculum identified 58 genes whose inactivation resulted in a growth defect under all conditions tested ( Fig 1C , S1 and S2 Tables ) . Most of these genes are involved in basic metabolic processes such as nutrient transport and utilization , respiration , response to oxidative stress , cell envelope biogenesis , DNA replication and repair , translation , and protein turnover . This group of mutants also includes insertions in 13 genes whose functions are unknown . Insertions in 46 genes resulted in mutants with growth defects in defined liquid but not in solid rich media ( Fig 1C and S2 Table ) . Among those , we identified mutations in genes involved in proline ( proABC ) , biotin ( bioAFD ) , and purine and pyrimidine ( carB , pyrBF , and purEFHLMNQS ) biosynthesis . These observations validated our experimental approach since the defined medium used in the screen lacked these substrates . The analysis of the results described below combined with isotopologue profiling and metabolic flow analysis provides a comprehensive view of the metabolic requirements for C . jejuni to grow in rich and defined media . It was recently reported that mice , rendered dysbiotic by antibiotic treatment , could be stably colonized by C . jejuni [26] . We found that , predictably , antibiotic treatment drastically reduced though did not eliminate the bacterial load in the intestinal tract ( S4 Fig ) . More importantly , we found that , consistent with previous reports [26] , C . jejuni strain 81–176 can robustly colonize the intestinal tract of dysbiotic mice for up to 3 weeks after oral infection ( Fig 3A ) . Large numbers of C . jejuni colony-forming units ( CFUs ) were recovered from the large intestine and cecum of infected animals as early as 4 days after infection and up to 21 days post infection , although colonization of systemic tissues was inefficient . Consequently , this infection model is suitable to screen a library with a large number of transposon insertion mutants for their ability to colonize the intestine as a strategy to identify metabolic determinants necessary for C . jejuni growth in this compartment . After oral inoculation of dysbiotic C57BL/6 mice with the C . jejuni 81–176 transposon insertion mutant library , the infected animals were killed 4 , 7 , or 21 days after infection , and C . jejuni CFUs were recovered from the cecum for INSeq analysis . About 1010 C . jejuni CFUs were recovered from the cecum of infected mice 4 days after infection , and between 108 and 109 CFUs were recovered 21 days after infection ( Fig 3B ) . INSeq analysis of the inoculum revealed transposon insertions in 1 , 323 out of 1 , 758 predicted genes of C . jejuni 81–176 . Since only insertions within the first 80% of the coding sequence were considered in the analysis , it is likely that all these insertions resulted in inactivation of the targeted genes . Analysis of the mutants recovered 21 days post infection revealed a marked increase in insertions within the pVir and pTet plasmids , with a marked reduction in the representation of the mutant pool ( S5 Fig ) . The reasons for the “blooming” of these mutants late in infection are unclear , although our library already contained a preponderance of insertions within the resident plasmids , presumably because differences in DNA superhelicity may favor insertions into plasmid genes . Nevertheless , the blooming of plasmid insertion mutants observed at day 21 precluded the use of this time point to analyze C . jejuni colonization determinants . Examination of the mutant pool isolated 4 days after infection revealed the presence of an average of 935 mutants per animal ( approximately 70% of the inoculum pool ) , and no indication of blooming since the relative number of insertions within plasmid or chromosomal genes recovered from the intestine was similar to their ratio in the inoculum ( S5 Fig and S3 Table ) . Therefore , we chose this time point to identify genes involved in the initiation of C . jejuni colonization of the intestinal tract . The distribution of log2 values for the number of insertions in most genes showed an overall reduction in comparison with the inoculum ( Fig 3C ) . We observed a similar distribution even when the data were normalized by an alternative approach ( S6 Fig and S4 Table ) ( see “Materials and methods” ) . Although the reasons for this downshift in the population are not known and likely to be multifactorial , it is not unprecedented since it has been observed in previous similar studies [37] . Consequently , the fitness defect of a given mutant could not be confidently derived by a given cutoff in the log2 ratio of the difference between the number of insertions recovered from the intestine versus the number in the original inoculum . Instead , we identified genes with fitness defects by applying statistical analysis to determine the likelihood that the differences in log2 ratios for any specific gene could have happened by chance ( see “Materials and methods” ) . This analysis identified 143 genes that showed a significantly decreased number of insertions within the mutant pool recovered from the intestine relative to the inoculum ( Fig 3C and 3D and Table 1 ) . We found that insertions located immediately downstream of 126 out of the 143 genes identified did not result in attenuating phenotypes indicating that the transposon insertions do not result in polar effects on downstream genes ( S5 Table ) . Mutations in 101 of these genes did not result in a fitness reduction when grown in rich medium ( Table 1 ) , suggesting that they may be specifically required for initiation of intestinal colonization . Mutations in the remaining 42 genes resulted in different degrees of fitness loss when C . jejuni was grown in vitro ( Table 1 ) . Our INSeq screen identified attenuating transposon mutations in most of the genes that previous studies have shown to be required for intestinal colonization of C . jejuni 81–176 [19 , 33 , 38–42] , which demonstrates the comprehensive nature of our approach . Thus , C . jejuni 81–176 mutants harboring transposon insertions within genes involved in motility , lipooligosaccharide and capsule biosynthesis , or N- and O-protein glycosylation , as well as metabolic traits , were readily detected among the pool of mutants showing reduced colonization ( Table 1 and S6 Table ) . Notably , however , several mutations in genes encoding cytolethal distending toxin components ( CdtA , CdtB , and CdtC ) and other previously described putative virulence factors involved in cell adhesion and invasion such as CiaB/C/I , CadF , FlpA , JlpA , PldA , or HtrA resulted in either no fitness cost or a level of attenuation significantly lower than many mutations affecting metabolic processes ( see below ) ( S3 Table ) . These observations underscore the relative importance of metabolism in C . jejuni intestinal colonization . Our analysis also identified insertions in many genes that have not been previously linked to intestinal colonization . What follows is an analysis of these findings aided by the framework provided by the information obtained through studies of the requirements for C . jejuni growth under different in vitro culture conditions . It has become increasingly clear that the ability to secure nutrients is a key determinant of bacterial pathogenesis [14 , 16 , 80] . Our in vivo INSeq analysis has provided evidence that during the infection process C . jejuni relies more on specific metabolic adaptations than on specific virulence factors targeting host processes . In contrast to other intestinal pathogens , like Salmonella or pathogenic Escherichia coli [81–83] , C . jejuni utilizes a limited number of carbon and energy sources . However , since low C . jejuni infectious doses are sufficient to cause disease [84 , 85] , such metabolic restrictions do not seem to impede the ability of C . jejuni to overcome the microbiota-mediated colonization resistance . In fact , it is possible that C . jejuni may rely on the microbiota to provide essential nutrients to fuel its metabolism . We find that C . jejuni can utilize potential catabolic end products of the intestinal microbiota such as acetate or CO2-derived hydrogen carbonate and especially free amino acids and di-/ or oligopeptides , which are released from dietary or endogenous proteins . Although free amino acids and dipeptides are prominent in the mucus layer of the small intestine , our study suggests that the concentration of some amino acids may not be sufficient for C . jejuni since auxotrophic mutants unable to synthesize branched-chain and aromatic amino acids or Ser exhibit a colonization defect . We show that C . jejuni overcomes such substrate restriction through an active TCA cycle , gluconeogenesis , and nonoxidative PPP , which together facilitate its anabolic capacity and are crucial for its in vitro and in vivo growth . C . jejuni preferentially colonizes the mucus of the intestinal crypts [54–56] , where it is spatially segregated from the luminal microbiota [86 , 87] . This spatial separation may allow C . jejuni to thrive despite its limited metabolic capacity . Indeed , while other enteropathogenic bacteria like Salmonella exhibit redundant catabolic pathways that allow the exploitation of multiple nutrients [88] , our analyses clearly showed that the metabolic network of C . jejuni lacks comparable redundancy and harbors various bottlenecks , as it depends on the gluconeogenic activity of the EMP pathway and the nonoxidative PPP . These metabolic bottlenecks may allow the development of novel strategies to prevent colonization and novel anti-infective drugs .
The complete list of strains and plasmids used in this study is shown in S9 Table . The highly saturated Himar 1 Mariner transposon mutant library has been previously described [27] . The C . jejuni 81–176 strains were grown on brucella broth agar or on blood agar plates ( Trypticase soy agar supplemented with 5% sheep blood ) at 37°C in an incubator equilibrated to a 10% CO2 atmosphere . The C . jejuni transformants were selected on plates supplemented with 50 μg/ml kanamycin , 7 . 5 μg/ml chloramphenicol , and 10 μg/ml erythromycin as indicated . For liquid cultures , C . jejuni strains were grown in brain heart infusion ( BHI ) medium with no antibiotics added in most cases except for the growth of the mutant library or defined C . jejuni mutant strains ( see below ) . All C . jejuni strains were stored at −80°C in BHI broth containing 30% glycerol . For growth in rich-medium experiments , approximately 108 CFUs of the C . jejuni transposon mutant library were plated on blood agar in 15-cm petri dishes , and the bacterial cells were collected after 48 hours . Six replicates of this growth condition were processed as described below . For growth in liquid defined minimal medium , 108 CFUs of the C . jejuni transposon mutant library were added to 4 ml of DMEM ( GIBCO; catalogue number 11965 ) supplemented with 20 mM Asp , Gln , or Ser . The culture tubes were placed on a rotating wheel in 10% CO2 atmosphere for 48 hours , and the bacterial cells were collected by centrifugation . Three replicates for each of these growth conditions were processed as described below . DNA extraction and sequencing were carried out as previously described [27] . The sequencing data were analyzed using the INSeq_pipeline_v2 package [89] . For this analysis , the sequencing data of each sample were normalized to total counts per million ( i . e . , the number of transposon insertions at each site was multiplied by 106 and then divided by the total number of insertions in the sample ) [90 , 91] . This normalization step enabled comparison of the relative abundance change of each insertion mutant in the mutant library across different samples . Alternatively , the data were normalized using the median values of the number of reads . In this case , the median read number for all transposon insertions was calculated in each sample and normalized to a median read number of 50 . To identify significant fitness genes under different growth conditions , a Z-test was applied to all gene mutants whose log-transformed output:input ratios were significantly different from the overall distribution across all biological replicates [89] . Consequently , only genes with a q-value of <0 . 05 from the Z-test were considered significantly altered from the input . All deep-sequencing data filtering , normalization , mapping , and statistical analysis were conducted in Perl and R . All animal experiments were conducted according to protocols approved by the Yale University Institutional Animal Care and Use Committee . Six- to eight-week-old C57BL/6 mice were treated with antibiotics in drinking water for 4 weeks to eradicate their commensal gut flora and allow robust C . jejuni colonization as previously described [26] . Antibiotics were added to drinking water at the following concentration: ampicillin , 1 mg/ml; neomycin sulfate , 1 mg/ml; ciprofloxacin 200 μg/ml; metronidazole , 1 mg/ml; and vancomycin , 500 μg/ml . Antibiotics were removed from drinking water 2 days prior to the oral administration through stomach gavage of 100 μl of sodium bicarbonate ( to neutralize the stomach pH ) followed by the oral administration of 109 CFUs of C . jejuni in 100 μl of PBS . Mutants were reisolated 4 and 21 days after the mice were inoculated with the transposon insertion pools . Analysis of the mutants recovered 21 days post infection revealed a marked increase in insertions within the pVir and pTet plasmids with a drastic reduction in the representation of the mutant pool . Although the reasons for the “blooming” of these mutants late in infection are unclear , their emergence precluded the use of this time point to analyze C . jejuni colonization determinants . Consequently , the mutant library screen was performed by killing the mice and homogenizing their ceca with a tissue homogenized 4 days after infection . Bacteria were recovered from the cecum homogenates by plating on 15-cm blood agar petri dish plates ( 3 plates per cecum sample ) containing Campylobacter selective supplements ( Karmali , Oxoid SR0167 ) . Colonies were scraped off the plates , and the genomic DNA was isolated as previously described [27] . The INseq DNA sample preparation , sequencing , and data analysis were carried out as described above . C . jejuni 81–176 knockout mutant strains were constructed as previously described [41] . Briefly , flanking regions of the selected open reading frames ( ORFs ) were amplified with specific primers , linked to either side of a kanamycin ( aphA3 ) , an erythromycin ( erm ) , or a chloramphenicol ( cat ) resistance cassette , and cloned into the pBluescript II SK using the Gibson assembly protocol [92] . The resulting plasmids were used to move the mutated alleles into the chromosome of C . jejuni 81–176 by natural transformation and allelic recombination . Complementation of the mutant strains of C . jejuni was achieved by introducing a wild-type copy of the gene at the hsdM locus as previously described [93] . The plasmids used to generate the mutants are listed in S9 Table . Six- to eight-week-old C57BL/6 mice were infected by oral gavage with 109 CFUs of the different C . jejuni mutants mixed with an equal number of the wild-type parent strain . To enumerate bacterial loads in the cecum , mice were killed 4 days after infection , ceca were homogenized in 3 ml PBS containing 0 . 05% sodium deoxycholate , and the dilutions were plated on blood agar plates with Campylobacter selective supplements with and without the selection antibiotics to determine the CFUs of the wild-type and mutant strains . Cultivation of C . jejuni for isotopologue analysis was performed as described in S1 Text . Sample preparation for isotopologue analysis by GC/MS was performed as previously described [94] . During sample preparation , the acidic hydrolysis of cellular proteins did not allow the determination of Asn , Cys , Gln , Met , and Trp: Cys and Met are destroyed during the hydrolysis procedure , whereas Asn and Gln are converted to Asp and Glu , respectively . Consequently , the values for Asp and Glu correlate with the averages of Asn/Asp and Gln/Glu [95 , 96] . C . jejuni 81–176 was grown in 50 ml Hank’s Balanced Salt Solution supplemented with iron ( II ) ascorbate ( Sigma ) , MEM Vitamin Solution ( Invitrogen ) , and 1% Casamino acids ( Roth ) as a carbon source . The liquid cultures were incubated under reduced-oxygen atmosphere using Anaerocult C-Packs ( Merck ) at 37°C under shaking ( 150 rpm ) conditions . The uptake of amino acids through C . jejuni 81–176 was investigated by analyzing the changes in the amount of the amino acids in the culture supernatants . Samples of 2 ml supernatant were taken at the indicated time points , the bacterial cells were removed by centrifugation ( 17 , 000 g , 5 min , 4°C ) , and an aliquot of the filtrated supernatant was used for GC/MS analysis . The 100 μl aliquot of the supernatant was diluted in 500 μl water , containing 4 μg of ribitol as an internal standard . The samples were mixed , dried under vacuum at room temperature , and stored at −20°C . Derivatization of the samples was done with 40 μl pyridine , containing methoxyamine hydrochloride ( 20 mg/ml ) and 60 μl N-Methyl-N-trimethylsilyltrifluoro-acetamide ( MSTFA ) . The GC/MS analysis was done on a Thermo GC Ultra coupled to a DSQII mass spectrometer equipped with an AS3000 autosampler ( ThermoScientific , Dreieich , Germany ) as described before [97] , with the following exceptions: helium flow was set to 1 . 1 ml/min , and the temperature was increased to a final 325°C . The solvent delay time was 5 . 80 min . Data analysis was performed with Metabolite Detector ( version 2 . 07; [98] ) , and quantification was done using 1 unique fragment ion for each metabolite . For statistical analysis , data were first normalized by dividing the peak area of every detected compound in each sample by the peak area of the respective internal standard ribitol . Afterwards , the mean and the standard deviation were calculated from 3 biological samples with 3 technical replicates each . To investigate the growth properties of amino-acid auxotrophic C . jejuni 81–176 mutants , the different strains were cultivated in a defined DAAM based on Hank’s Balanced Salt Solution with each amino acid present in 2 mM concentration supplemented with vitamin mix , 10 μM ferrous ascorbate , and 20 mM lactate as described before [61] . To compare the growth of the C . jejuni ktr and kdp mutants with the wild-type strain in the presence or absence of K+ , strains were inoculated ( starting OD600 of 0 . 02 ) into 4 ml of liquid defined rich medium ( S10 Table ) with or without the addition of 0 . 5 or 5 mM KCl . Cultures were incubated on a rotating wheel under 10% CO2 atmosphere , and the OD600 were measure at the indicated times after inoculation . All experiments were performed at least 3 times . For the motility assay , the optical density of the bacterial cultures was adjusted to an OD600 of 0 . 3 , and 10 μl was spotted onto soft agar ( 0 . 5% , wt/vol ) . The plates were incubated for 24 hours at 37°C , and the swarming diameters of the different strains were compared to the wild type and the nonmotile C . jejuni ΔmotA mutant strain . | There is accumulating evidence that in addition to canonical virulence factors such as toxins , adhesins , or invasins , bacterial pathogens utilize specific metabolic traits to colonize and proliferate within their hosts , a concept that is increasingly referred to as “nutritional virulence” . We have used transposon insertion mutagenesis combined with next-generation sequencing , a mouse model of infection , isotopologue profiling , and metabolic flow analysis to obtain a comprehensive view of the metabolic requirements for the intestinal colonization of C . jejuni , a leading cause of food-borne gastroenteritis in industrialized countries . This information could provide the basis to control C . jejuni colonization of food-producing animals or the human host . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"dna-binding",
"proteins",
"microbiology",
"campylobacter",
"jejuni",
"genetic",
"elements",
"molecular",
"biology",
"techniques",
"bacteria",
"campylobacter",
"bacterial",
"pathogens",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"amino",
"acid",
"analysis",
"proteins",
"medical",
"microbiology",
"microbial",
"pathogens",
"recombinant",
"proteins",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gastrointestinal",
"tract",
"biochemistry",
"anatomy",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"genomics",
"mobile",
"genetic",
"elements",
"serum",
"proteins",
"organisms"
] | 2017 | Metabolic and fitness determinants for in vitro growth and intestinal colonization of the bacterial pathogen Campylobacter jejuni |
Schistosomiasis is listed as one of most important tropical diseases and more than 200 million people are estimated to be infected . Development of a vaccine is thought to be the most effective way to control this disease . Recombinant 26-kDa glutathione S-transferase ( rSjGST ) has previously been reported to achieve a worm reduction rate of 42–44% . To improve the efficiency of the vaccine against Schistosoma japonicum , we immunized mice with a combination of pcDNA vector-encoded 26-kDa SjGST ( pcDNA/SjGST ) , IL-12 expressing-plasmid ( pIL-12 ) , and rSjGST . Co-vaccination with pcDNA/SjGST , pIL-12 , and rSjGST led to a reduction in worm burden , hepatic egg burden , and the size of liver tissue granulomas than that in the untreated infection controls . In addition , we detected high levels of specific IgG , IgG1 , and IgG2a against the rSjGST antigen in infected mice vaccinated with this combination of pcDNA/SjGST , pIL-12 , and rSjGST . Moreover , high expression levels of Th2 cytokines , including IL-4 and IL-10 , were also detected in this group , without diminished levels of IL-12 , INF-γ , and TNF-α cytokines that are related to parasite killing . In conclusion , we have developed a new vaccination regimen against S . japonicum infection and shown that co-immunization with pcDNA/SjGST vaccine , pIL-12 , and rSjGST has significant anti-parasite , anti-hepatic egg and anti-pathology effects in mice . The efficacy of this vaccination method should be further validated in large animals such as water buffalo . This method may help to reduce the transmission of zoonotic schistosomiasis japonica .
Schistosomiasis is an important helminth parasitic disease , and it remains a major health problem worldwide , especially in tropical and subtropical countries [1] . Schistosoma japonicum causes the most severe pathological symptoms , and it is estimated that several million people in China are infected every year , with considerable economic loss due to infection of both humans and domestic animals [2 , 3] . Although effective chemotherapeutic drugs , such as praziquantel and artemether ( artemisinin derivatives ) , are available for the treatment and prevention of schistosomiasis [4] , reinfection and decreased susceptibility to the drugs restrict their effectiveness [5] . Therefore , development of a safe and efficient vaccine would be a better strategy for control and prevention of schistosome infection [6] . Progress continues in the development of an anti-schistosomiasis vaccine . Sjc26GST ( S . japonicum 26-kDa glutathione S-transferase; GenBank accession no . BU711548 . 1 ) is a potential vaccine against S . japonicum [7 , 8] . Both native and recombinant purified Sjc26GST have been shown to provide a certain level of protection against infection , in terms of reduced worm burden , female fecundity , and egg viability [9–12] . We have also reported that reSjc26GST can be used for diagnosis of schistosomiasis in buffaloes , and that it provides high sensitivity and specificity [13] . In recent years , Sjc26GST has been developed into a DNA vaccine with the capacity to potentiate mainly Th1 immune responses against S . japonicum [14–16] . However , the effectiveness of the Sjc26GST DNA vaccine in reducing the worm burden was not significantly elevated , although we previously demonstrated that T helper type 1 ( Th1 ) responses are important in providing protective immunity against schistosome infection [17] . The effectiveness of DNA vaccination alone is limited , because it often generates only a weak cellular immune response; therefore , the complementary use of adjuvants may be required to improve vaccine potency and enhance its immunoprotective effects against S . japonicum [15 , 18 , 19] . IL-12 , which is involved in the differentiation of naïve T cells toward Th1 [20] , is an effective adjuvant in increasing the protective immunity from vaccination with rSm14 against S . mansoni [21] , as well as with Sj23 plasmid DNA against S . japonicum [22] . IL-12 co-administration with DNA vaccine priming can induce strong cell-mediated type 1 immune responses [20 , 23] . Although Th1 immune responses are important in providing protective immunity against schistosome infection [21 , 24 , 25] , a rapidly induced and excessive Th1 response may also cause damage to tissues of the infected host during parasite killing [26] . In addition , it has been shown that different adjuvants may be appropriate for various purposes , including prolonged antigen release , activation of nonspecific immune stimuli , and even reduction of side effects [27] . Research with a novel finding has shown that an immunization strategy employing combined DNA and recombinant protein vaccines can induce strong cellular and humoral responses [28] . Recently , this immunization strategy has also been used to provide a basis for optimizing vaccination against schistosomiasis japonicum [29–31] . In this study , we used pIL-12 as an adjuvant and co-immunized with recombinant SjGST ( rSjGST ) in an attempt to improve the protective efficacy of the SjGST DNA vaccine against Schistosoma japonicum . We found that combining pcDNA/SjGST with pIL-12 and rSjGST for immunization of mice increased the protective efficacy and simultaneously reduced pathologic injury of the liver caused by granulomas .
The DNA vaccine construct was modified from pQE31/Sjc26GST , which contains the intact open reading frame ( ORF ) of 26 kDa SjGST from Schistosoma japonicum ( GenBank accession no . BU711548 . 1 ) [13] . Briefly , the intact ORF of Sjc26GST was amplified by PCR using the pQE31/Sjc26GST vector as a template and primers containing EcoRV and BamHI restriction sites ( Forward-5′-CGGGATCCCGTCATGTCCCCTATACTAGGTTAT-3′ and Reverse-5′-GGGATATCCCTTTATTTTGGAGGATGGTCGCCA-3′ ) . The gene products was digested with BamHI/EcoRV at 37°C for 4 h , and then subcloned into the pcDNA4 vector ( Invitrogen , Waltham , MA , USA ) at 37°C overnight to generate the pcDNA/SjGST construct . Constructs were confirmed by restriction analysis and sequencing ( Fig 1A ) . The DH5α strain of Escherichia coli ( Invitrogen , USA ) was transformed with plasmid DNA . The DNA vaccine was amplified and purified according to the manufacturer’s instruction for endotoxin-free Giga Prep Kits ( Qiagen , Valencia , CA , USA ) . The plasmid DNA vaccine was resuspended in 5% glucose and stored at −80°C until use . In addition to pcDNA/SjGST , we used rSjGST , prepared previously ( Fig 1B ) [13] , and an IL-12-encoding plasmid ( pIL-12 ) , kindly provided by the laboratory of Dr . Tauo [32] , as an adjuvant to enhance the immune response . The mouse immunization protocol is shown in Fig 2 . One week before vaccination , all C57BL/6 mice ( 6-week-old males ) were intramuscularly primed with 100 μl ( 10 μM ) of cardiotoxin in the quadriceps femoris muscle to enhance the permeability of the cellular membrane prior to the first DNA vaccination [33] . Mice were divided into six groups , each of which included 6–8 mice . The mice in Group I ( untreated infection control ) were pretreated with cardiotoxin only; mice in Groups II and III were immunized with 100 μg of control plasmid ( pcDNA4 ) or 100 μg of pcDNA/SjGST alone , respectively; mice in Group IV were co-immunized with 100 μg of pcDNA/SjGST + 100 μg of pIL-12; mice in Group V were co-immunized with 100 μg of pcDNA/SjGST + 100 μg of rSjGST; mice in Group VI were immunized with a combination of 100 μg of pcDNA/SjGST , 100 μg of pIL-12 , and 100 μg of rSjGST . Mice were boosted two or three times at 2-week intervals ( weeks 2 , 4 , 6 ) following the initial vaccination in a manner specific for each group . After the final immunization , each group of mice was challenged with 30 cercariae by direct skin penetration through the abdomen at week 8 , as previously described [34] . All experiments were performed in triplicate to ensure the reliability of the data . Blood was collected from the tail vein of the mice at week 8 ( after vaccination ) and week 16 ( after infection ) . ELISA was used to determine the anti-SjGST antibody titer . In brief , 96-well plates were pre-coated with purified rSjGST ( 5μg/well ) at 4°C overnight . FBS ( 5% ) was added into each well to block non-specific binding and incubated at 37°C for 2 h . Diluted serum ( 1:20 dilution , 200μl ) was added and incubated at 37°C for 2 h . HRP-conjugated goat anti-mouse IgG , IgG1 , IgG2a , or IgG2b ( 1:2000 dilution ) was added and incubated at 37°C for 1 h . ABTS ( Zymed ) was used as a substrate , and the optical density was measured at 405 nm using an ELISA reader ( Dynatech MR5000 ) . After 8 weeks of infection , mice were bled , sacrificed , and perfused . Worms were collected from the hepatic portal system and counted by anatomic microscopy . Part of the mouse liver was cut , weighed , and digested with 5 ml of 5% KOH at 37°C overnight , then the egg number per gram was determined under microscope and calculated . The reduction rate was calculated using the following formula: Worm burden reduction rate ( % ) = ( 1—mean number of worms in immunized mice/mean number of worms in mice immunized with the control plasmid ) × 100 . Egg reduction rate ( % ) = ( 1—mean number of eggs per gram in immunized mice/mean number of eggs per gram in mice immunized with the control plasmid ) × 100 . The method was modified as our previously reports described [35] . The spleens of sacrificed mice were removed and placed in a Petri dish and then washed to flush out the spleen immune cells . The spleen immune cells were dispersed using a PBS-filled syringe equipped with a 23-G needle . Residual red blood cells were lysed in phosphate buffered solution ( PBS ) containing 150 mM ammonium chloride , 1 mM potassium bicarbonate , and 0 . 1 mM ethylenediaminetetraacetic acid ( EDTA ) . The intact spleen immune cells were recovered by centrifugation ( 500 ×g , for 3 min , 4°C ) , and resuspended for later use . Routinely , 95% of the isolated cells were viable , as determined by Trypan Blue exclusion assay . mRNA quantification was assessed by real-time polymerase chain reaction ( PCR ) . Total RNA was extracted using TRIzol™ reagent ( Life Technologies , Carlsbad , CA , USA ) and reverse transcribed with Moloney Murine Leukemia Virus ( MMLV ) reverse transcriptase ( Promega , Madison , WI , USA ) to generate cDNA . Each cDNA pool was stored at -20°C until further real-time PCR analysis . Primer pair specificity was validated by performing a RT-PCR reaction using common reference RNA ( Stratagene , La Jolla , CA , USA ) as a DNA template . The primers used are: Interleukin ( IL ) -2 , forward: 5'-ATGGACCTACAGGAGCTCCTG-3' , reverse: 5'-TCAAATCCAGAACATGCCGCAG-3'; IL-12 , forward: 5'-ACATGGTGAAGACGGCCAG-3' , reverse: 5'-GAAGTCTCTCTAGTAGCCAG-3'; Interferon-gamma ( INF-γ ) , forward: 5'-CTTCCTCATGGCTGTTTCTG-3' , reverse: 5'-TGTCACCATCCTTTTGCCAG-3'; IL-4 , forward: 5'-CAGAGAGTGAGCTCGTCTG-3' , reverse: 5'-GGTGCAGCTTATCGATGAATC-3'; IL-10 , forward: 5'-ATGCAGGACTTTAAGGGTTAC-3' , reverse: 5'-CCTGAGGGTCTTCAGCTTC-3'; Tumor Necrosis Factor-alpha ( TNF-α ) , forward: 5'-CCTCACACTCAGATCATCTTC-3' , reverse: 5'-CGGCTGGCACCACTAGTTG-3'; Glyceraldehyde-3-phosphate dehydrogenase ( GAPDH ) and 18s were used as endogenous reference genes . 18s , forward: 5'-AAGACGGACCAGAGCGAAAGCA-3' , reverse: 5'-ATCGCCAGTCGGCATCGTTTATG-3'; GAPDH , forward: 5'-TCCTGGTATGACAATGAATACGG-3' , reverse: 5'-GATGGAAATTGTGAGGGAGATG-3' . Real-time PCR reactions were performed on the lightcycler nano real-time PCR system ( Roche Diagnostics , Mannheim , Germany ) using LightCycler 480 SYBR Green I Master ( Roche Diagnostics ) . Briefly , 10 μl reactions containing 2 μl of Master Mix , 2 μl of 0 . 75 μM forward primer and reversed primer , and 6 μl of the cDNA sample . Each sample was run in triplicate . The real time PCR program were for 3 minutes at 95°C; 45 cycles for 10 seconds at 95°C , 30 seconds at 60°C . At the end of the program , a melt curve analysis was done . Data analysis was performed using the LightCycler Nano software version 1 . 0 ( Roche ) . The entire livers of sacrificed mice were collected , fixed , and then stained with hematoxylin and eosin , as described previously [36] . The extent of pathological changes in the liver was assessed macroscopically on the day of sacrifice . Briefly , liver specimens were fixed in 10% formalin , embedded in paraffin , and sectioned at 3-μm thickness . The sizes of non-confluent granulomas formed around a single egg were measured in the stained sections using the ImageJ software ( NIH ) . For all histological quantification , 7 mice livers from each group were analyzed and at least 3 slides were used to determine the width of the granuloma borders of each specimen by referring the previous description [37] . Two liver sections , selected to be sufficiently distant from each other to ensure that a granuloma was not measured twice , were used for all granuloma measurements . Data were collected from three independent experiments . Antibody and cytokine levels were compared by one-way ANOVA using SPSS 18 . 0 software ( SPSS Inc . Chicago , IL , USA ) . Other results were analyzed using two-tailed Student’s t-test . In all analyses , differences of p < 0 . 05 were considered significant .
Eight weeks after vaccination with pcDNA/SjGST alone ( Fig 3A left bars ) , the titer of anti-rSjGST IgG increased more than pcDNA4 control group . Levels of anti-rSjGST IgG antibodies in mice vaccinated with pcDNA/SjGST + pIL-12 alone ( Group IV ) , + rSjGST alone ( Group V ) , or + pIL-12 and rSjGST ( Group VI ) were significantly higher than those in Group II or III ( p <0 . 001; p <0 . 05 ) . Furthermore , the dominant antibody isotypes in mice vaccinated with pcDNA/SjGST plus rSjGST were IgG1 and IgG2b ( p <0 . 01; p <0 . 001 ) ( Fig 3B and 3D , left bars ) . Immunization with a combination of pcDNA/SjGST and pIL-12 significantly induced IgG2a production ( p <0 . 05 ) ( Fig 3C , left bars ) . On the other hand , eight weeks after infection with S . japonicum , sera were collected and the mice were sacrificed . We found that levels of anti-SjGST specific total IgG were higher in the infected mice vaccinated with pcDNA/SjGST + pIL-12 , pcDNA/SjGST + rSjGST , or plus both pIL-12 and rSjGST ( Group IV–VI ) than in the control group mice vaccinated with pcDNA4 ( Group II ) ( p <0 . 05 ) ( Fig 3A , right bars ) . However , Fig 3B ( right bars ) showed that IgG1 was strongly elicited only in mice vaccinated with pcDNA/SjGST + rSjGST , whether pIL-12 was included or not ( Group V–VI ) ( p < 0 . 001 ) ; vaccination with pcDNA/SjGST sufficiently induced the production of IgG2a except in Group V that plus rSjGST alone ( p < 0 . 001 ) ( Fig 3C , right bars ) ; and only vaccination with pcDNA/SjGST alone ( Group III ) induced specific IgG2b ( p < 0 . 001 ) ( Fig 3D , right bars ) . Vaccination with the pcDNA/SjGST plasmid efficiently induced production of specific anti-SjGST antibodies recognizing the rSjGST antigen . Next , we tested the protective efficacy against S . japonicum infection . The vaccinated mice groups were challenged with 30 cercariae . Eight weeks following infection , mice were bled and sacrificed . Adult worms were collected from the hepatic portal system after perfusion and the egg burden was measured . Fig 4A and Table 1 show the worm count for the control group ( Group II , immunized with the pcDNA4 plasmid alone ) did not differ from that in untreated infected mice ( Group I ) . Compared to Group II , the worm burden of Group III was significantly reduced ( the reduction rate was 62 . 02% ) ( p < 0 . 001 ) . Co-immunization with pcDNA/SjGST + pIL-12 or rSjGST enhanced the worm reduction rate to 68 . 99% and 72 . 33% , respectively ( p < 0 . 001 ) . Combined vaccination with pcDNA/SjGST + pIL-12 + rSjGST sufficiently reduced the number of recovered worms , from 12 . 9 to 3 . 33 ( reduction rate = 74 . 19% ) ( p < 0 . 001 ) . The number of eggs in the liver tissue of each group is presented in Fig 4B and Table 1 . Compared to Group II mice , the number of eggs per gram of liver in Group III was only reduced by 17 . 08% . However , the addition of pIL-12 to pcDNA/SjGST led to a significant reduction of the hepatic egg burden ( 37 . 54% , p < 0 . 05 ) compared to Group II mice . A significant difference ( a decrease of 46 . 50% ) was observed between the group co-immunized with pcDNA/SjGST + pIL-12 + rSjGST and the group immunized with pcDNA4 alone ( p < 0 . 05 ) , but not the group immunized with pcDNA/SjGST + rSjGST . Thus , vaccination of mice with pcDNA/SjGST + pIL-12 + rSjGST can significantly reduce the worm and hepatic egg burdens . Splenic immune cells were isolated immediately after mice were sacrificed and then subjected to gene expression assay . As shown in Fig 5A , mice vaccinated with pcDNA/SjGST showed significant induction of IL-2 , except for Group V that plus rSjGST alone ( p < 0 . 01 ) when compared with the pcDNA4 control . Mice vaccinated with pcDNA/SjGST , particularly when plus pIL-12 alone ( Group IV ) , effectively upregulated IL-12 ( p < 0 . 001 ) as well as INF-γ ( p < 0 . 01 ) mRNA , with the exception of INF-γ expression in Group V ( Fig 5B and 5C ) . Conversely , as presented in Fig 5D and 5E , the Th2 IL-4 gene and immunosuppressive IL-10 gene were significantly upregulated when rSjGST was used as an adjuvant whether pIL-12 was included or not ( p < 0 . 01; p < 0 . 05 ) . IL-4 mRNA levels in Group V were much higher than in the other groups without rSjGST ( p < 0 . 05 ) , whereas IL-10 expression was the highest in the group vaccinated with the pcDNA/SjGST + pIL-12 + rSjGST combination ( p < 0 . 05 ) . It is noteworthy that TNF-α gene expression was significantly induced in all mice vaccinated with pcDNA/SjGST ( p < 0 . 001 ) . TNF-α mRNA expression was the highest in Group VI , compared to that in all of the other experimental groups ( p < 0 . 05 ) ( Fig 5F ) . Liver damage caused by the egg-induced granulomas is the major pathogenicity of S . japonicum infection . This egg deposition results in the formation of granulomas . However , we found that liver pathology by granuloma formation was not entirely consistent with the reduction rates of egg deposition which significant reduced in Group IV and Group VI . As shown in Fig 6 , deposition of many schistosome eggs and severe granuloma formation were observed in liver tissue collected from Groups I and II . After vaccination with pcDNA/SjGST or co-immunization with pIL-12 , we found that granuloma pathogenesis was not significantly reduced ( Fig 6 , III–VI ) , while hepatic egg burdens were decreased ( Table 1 and Fig 4B ) . However , few sites of egg deposition or granuloma were observed Groups V and VI , which were vaccinated with a combination of pcDNA/SjGST + rSjGST and pcDNA/SjGST + pIL-12 + rSjGST . Morphological observations of liver tissue from mice in these groups are concordant with these results ( Fig 7A ) . To determine whether granuloma formation would be affected in immunized mice , we measured the size of granulomas observed in the hepatic tissue samples . As shown in Fig 7B , mice immunized with pcDNA/SjGST exhibited smaller granulomas than mice in the untreated infection control group ( p < 0 . 05 ) . Moreover , the size of granulomas was not reduced in mice simultaneously co-immunized with pIL-12 as much as in other groups . In contrast , the area of egg-induced granulomas was significantly reduced in mice vaccinated with pcDNA/SjGST + rSjGST or pcDNA/SjGST + pIL-12 + rSjGST compared with mice vaccinated with pcDNA alone or untreated infected mice ( p < 0 . 05; 0 . 01 ) . In addition , Group V and Group VI were also different from each other ( p < 0 . 05 ) . This suggests that recombinant SjGST suppressed the excessive liver damage caused by hepatic granuloma . The above results show that vaccination with SjGST DNA and recombinant SjGST can protect mice from S . japonicum infection .
In the public health area , development of a vaccine is thought to be an efficient strategy to control schistosoma infection . The WHO recommends that the minimal protection level provided by candidate vaccines is 40% [38] . Several studies have shown that effective vaccination against schistosomes depends on the simultaneous induction of both cell-mediated and humoral immunity [39–42] . Using DNA priming/protein vaccine boosting protocols may facilitate achievement of this immunization goal . In this study , we focused on protection enhancement and pathogenesis reduction of the pcDNA/SjGST DNA vaccine , by co-immunizing with IL-12 plasmid DNA in addition to recombinant SjGST protein . Immunization with SjGST DNA-based vaccines against the zoonotic pathogen S . japonicum has been much studied , even though they produce only a limited reduction in worm burden [14 , 29 , 43] . Insufficient uptake of intramuscularly delivered DNA plasmid leading to poor immune responses to antigens is one of the primary reasons underlying the overall low efficacy of DNA vaccines . Combining SjGST plasmid DNA vaccines with adjuvants , such as Cimetidine , IL-18 , TLR ligands , levamisole , and others , could enhance the protection level [15 , 18 , 19 , 44] . Here , we showed that our pcDNA/SjGST plasmid DNA provides efficient worm reduction ( 60% ) , up to >70% if used with the adjuvant pIL-12 , rSjGST , or both . However , the anti-egg effects of vaccination with pcDNA/SjGST were not as good as the worm reduction effects , unless the mice were co-immunized with pIL-12 , whereas the addition of both pIL-12 and rSjGST produced the highest egg reduction rate ( 46% ) . Use of IL-12 as a gene adjuvant induces production of IFN-γ and TNF-α cytokines , and enhances cytotoxic type-1 responses as well as protective immunity [21 , 45] . Additionally , similar results were observed in our cytokine data ( Fig 5 ) . Moreover , vaccination experiments in pigs and buffalos have shown that combining IL-12 with a DNA vaccine can significantly improve the hepatic or fecal egg reduction rate and the female worm reduction rate [22 , 46] . Cheng et al . confirmed the anti-fertility effect of IL-12 , which can lead to impaired worm development and a significantly decreased number of eggs per pair of worms in the liver as well as in the uteri of ovigerous females [47] . Siddiqui et al . found that pIL-12 co-injected with Sm-p80 DNA yielded an augmentation of total IgG and IgG2a , predominantly of Th-1–type antibodies [23 , 48] . Our data showed that vaccination with pcDNA/SjGST alone or a combination of pcDNA/SjGST and pIL-12 induce sufficient levels of total IgG and Th1-associated IgG2a to protect mice from schistosome infection . Furthermore , IgG2a and IgG1 , Th1- and Th2-associated antibodies , were the dominant subtype detected in mice co-immunized with pIL-12 , rSjGST , and pcDNA/SjGST . In addition , the gene expression data demonstrated that immunization with pcDNA/SjGST , particularly with pIL-12 as an adjuvant , induced significant Th1-associated IL-12 and INF-γ cytokines . When rSjGST was included in the immunization program , the expression of Th2 and Treg cytokines , IL-4 and IL-10 , was increased markedly without leading to a decline in IL-12 expression levels . Our data show that co-vaccination with pcDNA/SjGST vaccine plus pIL-12 and rSjGST can induce parasite-specific humoral immunity but does not interfere with the protective efficacy of cell-mediated immune responses . Excessive Th1 cytokines that rapidly expand in the acute stage of infection may cause lethal immunopathology in a parasitized host [26 , 49] , whereas suitable Th2 cytokines can attenuate excessive inflammatory injury and promote tissue repair [50 , 51] . Although previous reports have shown that co-immunization with a recombinant vaccine and rIL-12 can reduce the size of hepatic granulomas in infected mice [21] , and even IL-12 itself can reduce the size of granulomas [47 , 52] , other studies have indicated that vaccination with a plasmid DNA vaccine plus an IL-12-expressing plasmid might instead increase hepatic damage [22] . Recombinant IL-12 and plasmid IL-12 DNA as adjuvants co-immunized with a DNA vaccine may have a different effect , because the plasmid DNA adjuvants might amplify and sustain the immune effect of IL-12 by cell expression [53] . DNA vaccines themselves also induce Th1-type cellular responses with high levels of IFN-γ production [43] . Therefore , co-immunization of pcDNA/SjGST with pIL-12 may induce an excessive Th1 response in a short time . Nonetheless , IL-12 was still indispensable in our vaccination regimen , improving protective immunity and anti-egg effects of the pcDNA/SjGST vaccine . Using a recombinant protein vaccine to boost the effect of a DNA vaccine has been advanced as a new regimen for development of vaccines against schistosomiasis [31] , even in the application of SjGST vaccine [29 , 30] . Our results show that pcDNA/SjGST vaccination by boosting with recombinant SjGST proteins not only enhanced the anti-parasite efficacy against schistosomiasis ( Table 1 and Fig 4A ) but also significantly increased the anti-pathological effects , as evident in the reduction in the quantity and size of liver granulomas observed in the rSjGST-boosted groups ( Fig 4B and Fig 7B ) . The positive impact of the SjGST boost on induced Th2 immunity was also demonstrated by inducing SjGST-specific IgG1 antibody and Th2 cytokine increases; in particular , IgG1 and IgG2a levels as well as Th1 and Th2 cytokines were elevated with pIL-12 and rSjGST co-immunization . The Th2 cytokines , IL-4 and IL-13 , can suppress excessive neutrophil recruitment and proinflammatory cytokine production , which are mainly responsible for hepatic damage during schistosomiasis japonica infection [54] . Thus , rSjGST boosting may play a role in alleviating hepatic damage caused by strong Th1-promoting DNA vaccines , plus adjuvants such as pIL-12 . Results of cytokine expression explained the possible mechanism of additive effects as co-immunization with pcDNA/SjGST + pIL-12 and rSjGST . Immunization of pcDNA/SjGST alone or pcDNA/SjGST + pIL-12 both rapidly enhanced Th1 responses , e . g . IL-2 , IL-12 and INF-γ that are important in providing protective immunity against S . japonicum [48 , 55] . This effect is particularly significant when pcDNA/SjGST combining with pIL-12 and no matter plus rSjGST or not . During schistosoma infection , TNF-α participated not only the cell-mediated protective immunity , but also involved in the immune responses that affect liver fibrosis and hepatosplenomegaly [56 , 57] . Our data showed that the pcDNA/SjGST vaccine induced significant levels of TNF-α and was the highest in Group VI with both pIL-12 and rSjGST as adjuvants . Meanwhile , rSjGST protein added to the immunization protocol induced the expression of Th2 cytokines IL-4 , and especially IL-10 . IL-10 plays a crucial role in regulating not only the severity of acute liver pathology , but also granulomatous organization and cohesiveness [58 , 59] . Egg-induced inflammation , large granulomas , hepatic fibrosis , and tissue eosinophilia were observed in IL-10/IL-12-deficient mice with schistosome infection [49] . In the acute stage of schistosome infection , tolerance to worm egg antigens led to the enhancement of the Th1 response and a reduction in the Th2 response , as well as an increase in host mortality rate [60] . Therefore , inducing the appropriate Th2 response will decrease unnecessary damage to host tissue within immune attacks . We have developed a new vaccination regimen against schistosomiasis japonica . Co-immunization of mice with pIL-12 , rSjGST , and a pcDNA/SjGST vaccine produced significant anti-parasite , anti-hepatic egg , and anti-pathology effects . This regimen can induce both specific cellular and humoral responses to attain a balance between parasite elimination and prevention of pathological tissue injury . The efficacy of our method of vaccination should be further validated in large animals , such as water buffalo . These results may help to reduce the transmission of zoonotic schistosomiasis japonica . | Schistosomiasis continues to be a serious global public health problem that considered by World Health Organization ( WHO ) . More than 200 million people are infected and cause 280 thousand deaths every year . Among , Schistosoma japonicum causes the most severe pathological damages and the slowest immune resistance manifestation . It is estimated there is considerable economic loss in China due to the infection of human and domestic animals . Therefore , development of a useful vaccine is thought to be an efficient strategy to control and prevent schistosome infection . In this study , we co-immunized mice with pcDNA/SjGST vaccine , pIL-12 and rSjGST to develop a new vaccination regimen against schistosomiasis japonica . And we found this regimen can induce both specific cellular and humoral responses to attain a balance between parasite elimination and prevention of pathological tissue injury . The new regimen produced significant anti-parasite , anti-hepatic egg , and anti-pathology effects . Our method of vaccination can be applied in large livestock , such as water buffalo or cow that may help to reduce the transmission of zoonotic schistosomiasis japonica . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"schistosoma",
"invertebrates",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"helminths",
"granulomas",
"immunology",
"animals",
"vaccines",
"preventive",
"medicine",
"developmental",
"biology",
"immunologic",
"adjuvants",
"molecular",
"development",
"vaccination",
"and",
"immunization",
"schistosoma",
"japonicum",
"public",
"and",
"occupational",
"health",
"dna",
"vaccination",
"animal",
"cells",
"immune",
"system",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms",
"recombinant",
"vaccines"
] | 2016 | Combined IL-12 Plasmid and Recombinant SjGST Enhance the Protective and Anti-pathology Effect of SjGST DNA Vaccine Against Schistosoma japonicum |
During echinoderm development , expression of nodal on the right side plays a crucial role in positioning of the rudiment on the left side , but the mechanisms that restrict nodal expression to the right side are not known . Here we show that establishment of left-right asymmetry in the sea urchin embryo relies on reciprocal signaling between the ectoderm and a left-right organizer located in the endomesoderm . FGF/ERK and BMP2/4 signaling are required to initiate nodal expression in this organizer , while Delta/Notch signaling is required to suppress formation of this organizer on the left side of the archenteron . Furthermore , we report that the H+/K+-ATPase is critically required in the Notch signaling pathway upstream of the S3 cleavage of Notch . Our results identify several novel players and key early steps responsible for initiation , restriction , and propagation of left-right asymmetry during embryogenesis of a non-chordate deuterostome and uncover a functional link between the H+/K+-ATPase and the Notch signaling pathway .
Left-right ( L/R ) asymmetry is an essential feature of development in most bilaterian animals . In vertebrates , the morphology and positioning of many internal organs as well as development of the nervous system is left-right asymmetric and failure to establish these asymmetries can result in pathological disorders [1]–[7] . Left-right asymmetric processes have also been analyzed during development of a number of invertebrates including cephalochordates [8] , [9] , ascidians [8] , sea urchins [10] , snails [11] and insects [12] , [13] . How left-right asymmetries arise from embryos that are initially bilaterally symmetrical and how the left-right axis aligns consistently with the antero-posterior and dorsal-ventral axes are important questions that have recently become the subject of intensive research in a number of laboratories . Studies in vertebrates suggest that specification of the left-right axis can be conceptually divided into four distinct steps [1] , [5] , [14] , [15] . The first step involves a directional symmetry-breaking event that allows the L/R axis to be aligned with respect to the A/P and D/V axes . A failure to establish this directional asymmetry results in randomized left-right asymmetries ( heterotaxia ) characterized , for example by the stochastic positioning of the visceral organs on the left or the right side . In mouse , zebrafish or Xenopus , a leftward flow generated by a ciliated left-right organizer , ( the node in the mouse , Küpffer vesicle in zebrafish , and archenteron roof in Xenopus ) plays a key role in setting up this initial asymmetry [16] . In contrast , an asymmetrical cell migration at Hensen's node is responsible for establishment of left-right asymmetry in the chick [17] . Furthermore , in both Xenopus and chick , there is evidence for left-right asymmetries being established well before the appearance of cilia in the derivative of the organizer [18]–[20] . It is therefore generally believed that the mechanisms used during the initial symmetry-breaking phase are divergent in different species [2] , [21] . The second step in left-right axis determination involves establishment of asymmetric gene expression on the left and/or right side of the embryo in response to the flow of laterality information from the organizer . In contrast to the apparent variety of mechanisms used to break the bilateral symmetry in vertebrates , there is a striking conservation in the role played by the TGF beta Nodal in this process . In all vertebrate and chordate species studied so far , including zebrafish , Xenopus , mouse , rabbit , amphioxus and in the tunicate Ciona , nodal is the earliest known gene expressed in the periphery of the node and in the left lateral plate mesoderm in response to signals from the left-right organizer [2] , [8] . During the third step , left-right information is transferred from the organizer to the lateral plate . Elegant genetic experiments in the mouse revealed that during this process , Nodal produced in the node region activates its own expression in the distant lateral plate [22] , [23] and that this induction requires the expression of the TGF beta GDF1 in the node [24] . In the lateral plate , Nodal activates the expression of its downstream target pitx2 , which by regulating cell proliferation , cell migration and cell adhesion , participates in the fourth and crucial step of left-right axis i . e . the translation of asymmetric gene expression into asymmetric placement and morphogenesis of the organ primordia [4] , [25]–[27] . An important and heavily debated question in the field of L/R axis establishment is whether there is a conserved early cascade of laterality upstream of nodal expression [21] , [28]–[30] . In several species , the earliest event involved in the establishment of the L/R axis upstream of nodal expression involves the activity of the H+/K+-ATPase . Pharmacological inhibition of the H+/K+-ATPase induces heterotaxia in several vertebrate animal models including zebrafish [31] , Xenopus and chick [19] , causes random left-right determination in embryos from basal chordates such as tunicates ( Ciona intestinalis ) and disrupts left-right determination in embryos of basal deuterostomes organisms such as the sea urchin [10] , [32] . This strongly suggests that a mechanism involving the activity of the H+/K+-ATPase plays a central and perhaps ancestral role in determination of left-right asymmetry . The exact role played by the H+/K+-ATPase is largely enigmatic . Levin and colleagues suggested that an asymmetric activity of a H+/K+-ATPase may generate gradients of membrane potential that in turn may regulate the directionality of gap junction communication or , alternatively , that the activity of the H+/K+-ATPase may regulate the synthesis or secretion of a right sided determinant [19] . In contrast , Gros et al . reported that chick embryos incubated in the presence of omeprazole , an inhibitor of the H+/K+-ATPase , do not display the asymmetrical cell movements that initiate left-right asymmetry in birds , suggesting that the H+/K+-ATPase may regulate cell movements [17] . Raya et al reported that omeprazole treatment abolishes the Notch-dependent asymmetrical expression of Delta around the Hensen's node and suppressed the expression of nodal in the perinodal region indicating that omeprazole treatments interfere with the transcriptional activation of nodal in the node [33] . More recently , Walentek et al . proposed that the activity of the H+/K+-ATPase is required for canonical and non canonical Wnt signaling and foxJ expression [34] . Therefore , a unifying mechanism for the role of the H+/K+-ATPase is still lacking . In vertebrates , an early requirement for Notch signaling upstream of nodal expression is another conserved feature of left-right determination . In mouse , chick , and zebrafish , Notch signaling is required to initiate nodal expression around the node and mouse mutant lacking the activity of Delta1 , CSL ( CBF1/RBPJ/Su ( H ) /Lag-1 ) /Suppressor of Hairless or of Notch1 and Notch2 , fail to express nodal in the node region and show severe defects of left-right patterning [35] , [36] , [37] . Work from Izpisua Belmonte and coll . suggested a possible link between the role of ionic flux generated by the H+/K+-ATPase and Notch signaling . These authors proposed that , in addition to promoting the asymmetric expression of Delta1 around the node , an asymmetry in the activity of the H+/K+-ATPase may regulate an accumulation of extracellular calcium on the left side that may in turn promote the activation of the Notch signaling pathway [33] . Clearly , our understanding of the role of proton pumps in determination of L/R asymmetry remains scarce and further studies are required to clarify the links between the activity of the H+/K+-ATPase , extracellular calcium and Notch signaling . Recently , we started to dissect the process of left-right axis specification in the sea urchin [10] . Sea urchins are invertebrates but , like vertebrates , they belong to deuterostome superclade . This basal evolutionary position , as a sister group of the chordates , makes them an interesting phylum to study the conservation of mechanisms used to build the body plan of deuterostomes . Sea urchin development offers a striking example of left-right asymmetry ( Figure 1 ) . Like most echinoderms , sea urchins develop indirectly and their larvae undergo a metamorphosis during which most larval tissues are replaced by adult tissues generated from an imaginal disk called the adult rudiment , that forms exclusively on the left side of an otherwise bilaterally symmetric larva [38] , [39] . The rudiment derives from the left coelomic pouch and from a portion of the ectoderm located on the left side of the vestibule , where the mouth is located . Precursors of the coelomic pouches have a double origin: part of these precursors derive from the non-skeletogenic mesoderm that is induced by Delta-Notch signaling at the vegetal pole while another contribution comes from the small micromeres [40]–[43] . Although formation of the rudiment is a textbook example of left-right asymmetry , very little was known until recently on the mechanism that control the asymmetric positioning of this organ [44]–[46] . In particular , rudiment positioning has been shown to depend on a signal released by the micromeres but the identity of this signal is unknown [46] . We showed previously that a Nodal-Lefty-Pitx2 signaling pathway regulates left-right asymmetry during development of the sea urchin embryo [10] . However , intriguingly , nodal in the sea urchin is expressed on the right side of the ectoderm and in the right coelomic pouch at the end of gastrulation and not on the left side as in all vertebrates where its expression has been analyzed . Functional analysis revealed that one function of Nodal signals on the right side is to repress formation of the adult rudiment . Inhibition of Nodal signaling after gastrulation caused formation of an ectopic rudiment while ectopic activation of the pathway after gastrulation prevented formation of the rudiment [10] . Furthermore , we showed that inhibition of the H+/K+-ATPase disrupted the directional left-right asymmetry and randomized both nodal expression and positioning of the rudiment [10] . We now report that establishment of left-right asymmetry in the sea urchin embryo involves reciprocal signaling between the ventral ectoderm that expresses nodal and a left-right organizer of endodermal origin and that this long-range signaling requires Univin/Vg1 . We show that in the absence of this organizer or when an organizer forms both on the left and the right sides , nodal expression in the ectoderm is randomized along the left-right axis suggesting that this endomesodermal left-right organizer is only responsible for orienting the symmetry breaking and for making it directional . We provide evidence that establishment of this organizer requires the activity of several signaling pathways including the Notch , FGF-ERK , BMP2/4 and Univin/Vg1 . Finally , we report the unexpected finding that the activity of the H+/K+-ATPase is critically required for Notch signaling and that inhibiting the activity of this ATP driven proton pump phenocopies inhibition of Notch signaling in the early embryo leading to complete suppression of the expression of Notch target genes and to the absence of mesodermal derivatives . Our results therefore open the way to the analysis of the molecular pathway that regulates left-right asymmetry in the sea urchin embryo and uncover a functional link between two essential players of left-right asymmetry i . e . the H+/K+-ATPase and Notch signaling .
Asymmetric expression of nodal along the left-right axis could be detected as early as the mid-gastrula stage ( Figure 2A ) . At this stage ( about 22 hpf ) , while the archenteron had not yet reached the animal pole region , nodal expression was detected in a group of about 2–5 cells embedded into the wall of the archenteron on the right side . Double fluorescent in situ hybridization with the endodermal marker foxA and the mesodermal marker foxF confirmed that these nodal expressing archenteron tip cells are located at or near the boundary between the mesoderm and endoderm , immediately adjacent to the coelomic pouch precursors that express foxF ( Figure 2B ) . During the next 2 . 5 h period , the territory expressing nodal was progressively displaced towards the animal pole and at 24 hpf , a cluster of about 10–15 cells arranged in a rosette expressed nodal asymmetrically at the tip of the archenteron on the right side ( Figure 2A and Figure S1 ) . Based on their position immediately adjacent to the delaminating secondary mesenchymal cells , these nodal expressing cells at 24 h likely correspond to precursors of the right coelomic pouch . Importantly , during this period , nodal expression remained symmetric in the ventral ectoderm . Weak asymmetric expression of nodal was first detected in the ectoderm , on the right side of the presumptive ciliary band territory around 24 hpf . This asymmetry in the distribution of nodal transcripts in the ectoderm further accentuated during the following 3 h period and at 26 hpf , strong asymmetric expression of nodal on the right side was detected both at the tip of the archenteron and on the right side of the ectoderm in most embryos ( Figure 2A ) . Therefore , this analysis revealed that the first asymmetric expression of nodal occurs in the endomesoderm and not in the ectoderm , as previously thought , and that nodal expression subsequently expands from the endomesoderm to the mesoderm . Similarly , L/R asymmetric expression of univin started to be detected in the right coelomic pouch around 24 hpf , well after asymmetrical nodal expression had been initiated in the endomesoderm , while asymmetric expression of univin in the ectoderm occurred only after 26 hpf , well after nodal expression had switched to the right side of the ventral ectoderm ( Figure S1 ) . The finding that the first manifestation of left-right asymmetry determination during sea urchin embryogenesis is asymmetric expression of nodal in the archenteron strongly suggested that during normal development , the first symmetry-breaking event occurs in the endomesoderm . Furthermore , the later shift of nodal and univin expression from a bilaterally symmetric expression in the ectoderm to an asymmetric expression on the right side suggested that the asymmetry initiated in the endomesoderm is subsequently transferred to the ectoderm . Previous work [10 and unpublished data] as well as unpublished results from our lab indicated that in the sea urchin , like in vertebrates , the H+/K+ ATPase and Delta/Notch are key players required upstream of nodal expression during left-right axis establishment . We therefore first investigated if the activities of Notch and of the H+/K+ ATPase are required for the asymmetric expression of nodal in the endoderm at gastrula stage . Surprisingly , inhibition of Notch signaling by treatment with the γ-secretase inhibitor DAPT ( [N- ( 3 , 5-Difluorophenacetyl ) -L-alanyl]-S-phenylglycine t-butyl ester ) or by injection of a morpholino against Delta did not abolish nodal expression in the endoderm but caused instead ectopic expression of this gene on the left side of the archenteron ( Figure 3A ) . Starting at 22 hpf , while in control gastrulae nodal was expressed exclusively on the right side of the archenteron , in DAPT-treated embryos and in Delta morphants , nodal transcripts were expressed bilaterally in two groups of cells in the archenteron . Similarly , blocking the activity of the H+/K+-ATPase by treatment with omeprazole caused nodal to be expressed bilaterally in the endomesoderm at gastrula stage ( Figure 3A ) . These results suggest that Delta/Notch signaling and the activity of the H+/K+-ATPase are required ( either directly or indirectly ) to repress nodal expression in cells located on the left side of the archenteron . At pluteus stage , DAPT-treated larvae and Delta morphants expressed nodal and univin asymmetrically in the ectoderm but the expression was detected either on the right side or on the left side ( Figure 3B ) . Consistent with the random expression of nodal at pluteus stage , DAPT treated larvae developed with a rudiment that was randomly positioned on either the right or the left side ( Figure 3B ) . As controls for the effect of DAPT treatment and of the Delta morpholino , we analyzed the expression of marker genes transcribed either asymmetrically ( pitx2 , sox9 ) or symmetrically ( foxF ) in the coelomic pouches precursors or in the muscle cell precursors ( tropomyosin ) in response to Delta-Notch signaling . Indeed , expression of all four mesodermal marker genes was abolished in most of the DAPT-treated embryos as well as in the Delta morphants consistent with the expected severe reduction of mesodermal derivatives caused by inhibition of Notch signaling ( Figure 3C ) [41] , [42] . Therefore , inhibition of Notch signaling , in addition to preventing specification of the coelomic pouch precursors , caused the early endodermal expression of nodal to become bilateral and randomized nodal and univin expression in the ectoderm at pluteus stage . To determine when Notch signaling is required for establishment of left-right asymmetry , we treated embryos with DAPT for various time windows and analyzed the expression of nodal ( Figure 3D ) . This analysis revealed that the period during which DAPT is effective at perturbing left-right asymmetry corresponds to early development , with treatments performed during the cleavage/early blastula period being the most effective , the efficiency of the treatment rapidly dropping after early blastula stage , and treatments performed after hatching no longer perturbing left-right asymmetry . The period during which Notch signaling is required to establish left-right asymmetry largely overlaps with the period during which secondary mesodermal precursors are induced by Delta signals expressed in the primary mesenchymal cell precursors [42] , [47] . This suggests that Notch signaling regulates nodal expression indirectly , likely through signaling between the mesoderm that is induced by Delta/Notch signaling and the endoderm that expresses nodal . This also suggests that Delta is the signal released by the micromeres that regulates positioning of the rudiment [46] . In the sea urchin embryo like in vertebrates , treatments with the H+/K+ pump inhibitor omeprazole randomize L/R nodal expression [10] . Interestingly , we found a striking similarity between the phenotypes resulting from treatments with H+/K+ ATPase inhibitors , and treatments that interfere with Delta-Notch signaling ( Figure 4A ) . Treatments with omeprazole , like treatments with DAPT or injection of the morpholino against Delta , strongly delayed gastrulation and resulted in development of embryos that largely lacked delaminating secondary mesenchymal cells at the tip of the archenteron and that later were largely albino ( Figure 4A ) . Furthermore , the window during which omeprazole is mostly effective extends from fertilization to the very early blastula stage , i . e . a period very similar to the window of action of the Notch inhibitor DAPT ( Figure S2 ) . These observations raised the possibility that omeprazole treatments inhibit Notch-Delta signaling . To test this possibility , embryos were treated with omeprazole during cleavage and blastula stages and the expression of mesodermal marker genes activated in response to Notch activation , such as the immunocyte markers gcm , papss and GATA1/2/3 , was analyzed ( Figure 4B ) [48] , [49] , [50] , [51] . As a control , we analyzed the expression of the Delta ligand and of msp130 , two genes that are expressed in the skeletogenic mesoderm territory independently of Delta/Notch signaling [42] , [52] as well as the expression of the endodermal marker gene foxA [53] . Strikingly , in most embryos treated with the proton potassium pump blocker , expression of the immunocyte marker genes , which are regulated by Delta signaling , was strongly downregulated or absent . In contrast , foxA was expressed at apparently normal levels in the endoderm precursors . Furthermore , consistent with the previously described expansion of endodermal precursors at the expense of non skeletogenic mesodermal precursors caused by inhibition of Delta-Notch signaling [41] , [47] , [54]–[56] , the vegetal boundary of the territory expressing foxA in the DAPT , Delta-Mo injected embryos or omeprazole treated embryos was shifted towards the vegetal pole ( Figure 4B ) . In contrast , expression of Delta and msp130 in the skeletogenic mesoderm precursors was largely normal in the omeprazole treated embryos . This shows that inhibition of the H+/K+-ATPase does not perturb specification of the skeletogenic mesoderm and endoderm but that it specifically interferes with specification of the non-skeletogenic secondary mesoderm . Since the non-skeletogenic mesoderm is induced by Delta signals emanating from the adjacent skeletogenic mesoderm precursors , this further suggests that omeprazole treatment , may block reception of the Delta signal in the surrounding cells . We next sought to determine in which step of the Notch pathway , the activity of H+/K+-ATPase may be required by combining Notch gain of function and omeprazole treatments . During secretion in the trans Golgi network , the Notch protein is first processed by proteases of the Furin family that generate a non-covalent heterodimer between the Notch extracellular domain NECD and Notch tethered intracellular domain that interact in a Ca2+ dependent manner [57] . Upon binding of Delta , Notch is cleaved at the S2 site by proteases of the ADAM/TACE family , generating a membrane bound activated form of Notch called NEXT ( Notch Extracellular Truncation ) . NEXT is then the substrate for the gamma secretase , which catalyzes the intramembranous S3 cleavage that releases the Notch intracellular domain NICD [58] . To further define the step in which the activity of H+/K+-ATPase is required for Notch signaling , we used luciferase assays . We overexpressed mRNAs encoding the P . lividus Delta , NEXT or NICD proteins and measured the activity of the Notch reporter gene RBP-JK [59] in the presence or absence of omeprazole ( Figure 4C ) . Omeprazole treatment strongly inhibited the stimulation of Notch signaling induced by overexpression of Delta , consistent with a disruption of Notch signaling caused by the inhibitor . In contrast , omeprazole treatment had no effect on the activation of Notch signaling caused by overexpression of NEXT or NICD . This strongly suggests that the H+/K+-ATPase is required before or at the level of the S2 cleavage that generates NEXT . In vertebrates , the FGF/MAP kinase pathway is involved in establishment of left-right asymmetry . FGF signaling has been implicated in the symmetry breaking process and in the release of nodal vesicular parcels ( NVPs ) that carry Sonic Hedgehog and retinoic acid [60] . Furthermore , inhibition of FGF signaling disrupts left-right asymmetry in Xenopus and zebrafish , an effect that has been correlated to a reduction of ciliary length [61] . To investigate if FGF/MAP kinase signaling is required for the early asymmetry of nodal expression in the endomesoderm and for establishment of left-right asymmetry during sea urchin development , we analyzed the expression of nodal following treatments with the FGFR inhibitor SU5402 and with the MEK inhibitor U0126 ( Figure 5 ) . As controls for the effects of the inhibitors , we verified that the expression of pax2/5/8 and sprouty , two downstream targets of FGFA in the ectoderm [62] , is downregulated in the treated embryos ( Figure S3 and data not shown ) . While treatments with DAPT caused bilateral expression of nodal in the archenteron , in contrast , treatments with U0126 or with SU5402 abolished nodal expression in the endoderm at gastrula stage ( Figure 5A ) . Therefore , a positive input from the FGF/MAP kinase signaling pathway is required for nodal expression in the endomesoderm . To test if FGF/MAP kinase signaling is required for nodal induction through inhibition of Notch signaling , embryos were treated simultaneously with DAPT and U0126 . Double inhibition of Notch and MAP kinase signaling prevented nodal expression in the endomesoderm indicating that FGF signaling is likely required downstream or in parallel to Notch signaling for induction of nodal expression ( Figure 5A ) . Interestingly , despite the absence of nodal expression in the endomesoderm at gastrula stage , nodal was expressed asymmetrically in the ectoderm of SU5402 or U0126 treated embryos at pluteus stage , but as in the case of DAPT treated embryos , its expression was randomized along the left-right axis ( Figure 5B ) . Similarly , expression of sox9 , pitx2 and univin was randomized following inhibition of ERK signaling ( Figure 5B ) . About one third of the U0126-treated larvae later developed with two rudiments while in the remaining larvae the rudiment was either on the left ( 31% ) or the right side ( 38% ) ( Figure 5C ) . Time-course analysis revealed that the window during which SU5402 and U0126 are effective at perturbing left-right asymmetry extends from early mesenchyme blastula up to the early gastrula stage , i . e . , immediately before the initiation of asymmetric nodal expression in the archenteron tip cells ( Figure 5D and Figure S3 ) . Collectively , these results demonstrate that FGF/MAP kinase signaling is critically required to initiate asymmetric expression of nodal in the endoderm and that perturbations of nodal expression in this endodermal territory ultimately result in randomized left-right asymmetry in the ectoderm and random positioning of the rudiment ( Figure 5E ) . Furthermore , both the absence of nodal expression in this endomesodermal territory and the bilateral expression of nodal in this region resulted in randomized ectodermal nodal expression along the left-right axis suggesting that this endomesodermal region has the properties of a left-right organizer . Although this organizer does not appear to be necessary for the process of symmetry breaking itself , it is responsible for orienting the symmetry breaking and for making it directional . Finally , since Nodal and BMP2/4 play antagonistic roles during patterning of the ectoderm in the sea urchin embryo [63] , [64] and since BMP signaling is active in the upper part of the archenteron that expresses nodal during gastrulation [65] , we investigated if BMP signaling is required for specification of this left-right mesendodermal organizer and for the subsequent establishment of left-right asymmetry ( Figure 6 ) . We first tested the effects of perturbations of BMP signaling on nodal expression on the right side at gastrula stage . Treatments with recombinant BMP4 protein very efficiently suppressed nodal expression in the archenteron tip cells and in the ectoderm ( Figure 6A , Figure S4 ) suggesting that elevated BMP signaling can antagonize Nodal signaling in the context of left-right asymmetry . Injection into the egg of morpholino oligonucleotides directed against the bmp2/4 transcript or against the transcript encoding Alk3/6 , a type I BMP receptor that is required to transduce BMP2/4 signals [65] , also eliminated nodal expression in the left-right organizer indicating that BMP signaling is essential for the early nodal expression in the endomesoderm ( Figure 6A ) . Consistent with the observed loss of nodal expression in the endomesoderm at gastrula stages , nodal expression in the ectoderm was randomized in the bmp2/4 or alk3/6 morphants at pluteus stage . In the absence of BMP signaling , nodal expression in the ectoderm also expanded dorsally suggesting that BMP signaling is required as a dorsal barrier in the ectoderm ( Figure 6B ) . Taken together , these observations suggest that BMP signaling is first required in the endomesoderm to establish nodal expression in the mesendodermal organizer , then , that it is required as a dorsal barrier in the ectoderm to prevent expansion of nodal expression to the dorsal side . To test if BMP signaling is required in the endomesoderm or in the ectoderm for nodal expression in the left-right organizer , we specifically blocked BMP signaling in the endomesoderm by injecting the alk3/6 morpholino in the four vegetal blastomeres of embryos at the 8-cell stage and analyzed nodal expression at gastrula and pluteus stages . Inhibition of BMP signaling in the endomesoderm prevented nodal expression in the endomesoderm at gastrula stage in 93% of the injected embryos ( two experiments n: 30 ) ( Figure 6C , 6D ) . All the embryos injected with the alk3/6 morpholino in vegetal blastomeres nevertheless developed into pluteus larvae . However , consistent with the absence of nodal expression in the endomesoderm at gastrula stage , ectodermal nodal expression in these larvae was randomized . This result extends the previous observations made after inhibition of BMP signaling at the 1-cell stage and indicates that in the sea urchin embryo , BMP signaling in the endomesoderm plays a positive and essential role in the initiation or maintenance of nodal expression in the mesendodermal organizer . To better define the role of BMP signaling in the establishment of left-right asymmetry , we injected the bmp2/4 morpholino into one blastomere at the two cell-stage and , at gastrula stage , selected the embryos that inherited the morpholino on either the left or the right side and analyzed nodal expression at gastrula and pluteus stage ( Figure 6E–6G ) . Intriguingly , while targeting the bmp2/4 morpholino to the left side resulted in either the complete suppression ( 85% n = 20 ) or strong reduction ( 15% ) of nodal expression in the organizer at gastrula stage , normal nodal expression could be detected in 45% of the embryos that had received the morpholino on the right side ( n = 11 ) . The different sensitivities of the left and right sides to the bmp2/4 morpholino raised the possibility that BMP signaling on the left side may be required on the right side for nodal expression in the left-right organizer . Consistent with this idea , injection of the bmp2/4 morpholino into the presumptive right side territory did not perturb the sidedness of nodal at pluteus stage but strikingly , injection of the bmp2/4 morpholino on the presumptive left side randomized nodal expression in the ectoderm . To test if BMP signaling is asymmetric in the archenteron at gastrula stage , we tried to detect endogenous BMP signaling using an antiphosphoSmad1/5/8 antibody . Anti-phospho Smad1/5/8 immunostaining revealed the presence of a domain with strong BMP signaling in the archenteron at gastrula stages ( Figure 6H ) . In most embryos ( 13/19 ) , nuclear staining in the archenteron was asymmetric , with more intense staining being visible in the dorsal-left quadrant opposite to the region where nodal is expressed ( see also Figure S5 ) . These results suggest that in the sea urchin embryo , BMP signaling in the endomesoderm is required to establish nodal expression in the left-right organizer located on the right side . Furthermore , they suggest that at gastrula stage , BMP signaling itself is asymmetric , with stronger signaling occuring on the left side of the archenteron . As described above , treatments that perturb the early expression of nodal , resulting in either bilateral expression of nodal ( inhibition of Delta/Notch signaling ) or in the absence of expression of nodal in the endoderm ( inhibition of FGF/MAP kinase or of BMP signaling ) , ultimately randomize the expression of nodal in the ectoderm at later stages . This suggested that during sea urchin development , the first left-right asymmetry appears in the endomesoderm and that this asymmetry is subsequently transmitted to the ectoderm in the form of an asymmetric expression of nodal and univin on the right side of the ciliary band region . Consistent with this idea , previous experiments had shown that inhibition of nodal mRNA translation at the egg stage followed by local injection of nodal mRNA into one animal blastomere ( belonging to the presumptive ectoderm ) , efficiently rescued dorsal-ventral polarity , but failed to rescue left-right polarity in the endomesoderm and did not restore ectodermal expression of nodal and pitx2 on either side of the larva [10] . However , paradoxically , previous results from our laboratory also showed that inhibition of Nodal function in the ectoderm abolished the asymmetric expression of pitx2 in the endomesoderm suggesting that ectodermal Nodal signals were required upstream of endomesodermal Nodal expression [10] . One scenario that may reconcile these observations is that Nodal signals coming from the ectoderm may first be required for the asymmetric expression of nodal and pitx2 in the endomesoderm , then this asymmetry may be subsequently transmitted through Nodal signaling from the endomesoderm to the right ectoderm . To test this idea , we blocked Nodal signaling in either the ectoderm or the endomesoderm and analyzed nodal expression in the endomesoderm at gastrula stages as well as nodal and pitx2 expression in the ectoderm and coelomic pouches at pluteus stages ( Figure 7 ) . Injection of Nodal morpholino into the four animal blastomeres at the 8-cell stage abolished nodal expression in the endomesoderm at gastrula stage and produced radialized embryos consistent with previous results ( Figure 7A , 7B ) and Table 1 [10] . Therefore ectodermal Nodal signals are required upstream of endomesodermal nodal expression . In embryos radialized by treatments with recombinant Nodal or nickel chloride , however , nodal was expressed radially in the ectoderm but expression in the endomesoderm was abolished ( Figure S6 ) . Therefore , normal dorsal-ventral patterning of the ectoderm is required for nodal expression in the endomesoderm . Consistent with the idea that endomesodermal nodal expression requires ectodermal Nodal signals , blocking translation of nodal mRNA or blocking reception of Nodal signals in the endomesoderm by injection of alk4/5/7 morpholinos into the four vegetal blastomeres prevented nodal expression in the endomesoderm at gastrula stage ( Figure 7C , 7D and Table 2 ) . Injection of alk4/5/7 morpholinos into the four vegetal blastomeres did not affect establishment of dorsal-ventral polarity but it randomized nodal expression in the ectoderm at pluteus stage and eliminated pitx2 expression in the right coelomic pouch . Therefore , endomesodermal Nodal signals are indeed required to establish the directional asymmetry of nodal expression in the ectoderm . We also investigated if interfering with Nodal function in the endoderm perturbs establishment of left-right asymmetry in the ectoderm by using chimeras ( Figure 7E , 7F ) . Eggs were injected with the Nodal morpholino together with a lineage tracer and allowed to develop up to the 16/32-cell stage , then , the animal and vegetal regions were separated and recombined with their complementary halves derived from wild type embryos . When the function of Nodal was inhibited in the animal hemisphere , the resulting chimeras displayed a phenotype very similar to that observed following injection of the morpholino into the egg: the embryos lacked both dorsal-ventral and left-right polarity , consistent with the essential role of nodal in establishment of these embryonic axes ( not shown ) [10] , [66] . In contrast , chimeras in which the Nodal morpholino was present in the vegetal hemisphere developed into morphologically normal pluteus larvae ( Figure 7E , 7F ) ( 100% n = 12 ) . However , in these embryos , nodal expression in the ectoderm was randomized ( Figure 7F ) . This shows that , while Nodal function in the ectoderm is clearly important for establishment of left-right asymmetry in the endomesoderm , Nodal signaling in the endomesoderm is in turn essential for transmission of left-right asymmetry to the ectoderm . Therefore determination of left-right asymmetry in the sea urchin embryo most likely requires reciprocal signaling between the ectoderm and endomesoderm . If Nodal signals emitted from the ventral ectoderm drive nodal expression in the endomesoderm , why , in the dorsal-ventral axis rescue experiments mentioned above , local expression of nodal into one animal blastomere at the 8-cell stage is not able to rescue the expression of L/R markers in the endomesoderm of nodal morphants ? We reasoned that in the rescue experiment , the size of clone expressing nodal is much smaller , than the presumptive ventral ectoderm that normally expresses nodal . Furthermore , in these rescue experiments , the progeny of the nodal expressing blastomere typically occupies the center of the ventral ectoderm that gives rise to the region surrounding the stomodeum and , importantly , it does not overlap with the more lateral ectoderm that normally expresses univin at gastrula stage . Univin is a Vg1/GDF1 related factor that is very important during dorsal-ventral axis formation and Nodal/GDF1 heterodimers have been shown to be much more potent and to signal over a longer range compared to Nodal homodimers in other systems [24] . This raised the possibility that the failure of ectopic nodal to rescue left-right patterning in the endomesoderm might be due to the absence of overlap between the nodal expressing clone and the univin expressing territory and to the failure to form Nodal-Univin heterodimers at gastrula stages . To test this possibility , we analyzed pitx2 and sox9 expression following injection of nodal mRNA alone or of a mixture of nodal and univin mRNAs into one blastomere at the 8-cell stage of nodal morphants ( Figure 8 ) . While injection of nodal mRNA alone into an ectodermal precursor was unable to induce expression of pitx2 in either the endomesoderm or in the ciliary band , strikingly , co-injection of nodal and univin rescued expression of pitx2 in the right coelomic pouch and induced a massive expression of pitx2 throughout the right and left portions of the distant ciliary band ( Figure 8B ) . This shows that local and symmetric expression of nodal and univin in the ectoderm of nodal morphants is sufficient to rescue asymmetric expression of pitx2 in the endomesoderm , consistent with previous data showing that Nodal signaling in the ectoderm is essential for driving asymmetric nodal/pitx2 expression in the endomesoderm . Conversely , targeting the Univin morpholino to the right side of the embryo completely blocked the asymmetric expression of nodal in the ectoderm on the right side ( Figure S7 ) . Taken together , these results strongly suggest that Nodal and Univin synergize to signal both locally and over a long range during left-right patterning in the sea urchin embryo . In conclusion , these results ( summarized in Figure 9B ) strongly suggest that determination of left-right asymmetry in the sea urchin embryo involves two successive reciprocal long-range signaling events between the ectoderm and the endomesoderm mediated by Nodal-Univin heterodimers ( Figure 9A ) . First , during gastrulation , a Nodal/Univin signal emitted by the ventral ectoderm cooperates with an FGF signal of unknown origin and with a BMP signal coming from the left side of the archenteron to initiate nodal expression in cells on the right side of the tip of the archenteron ( Figure 10 ) . On the left side , an unidentified signal , likely emitted by the mesoderm induced by Delta/Notch signaling is required to repress nodal expression . Together , these positive and negative signals are responsible for establishment of a left-right mesendodermal organizer on the right side of the tip of the archenteron , which starts to express nodal then univin . At late gastrula/prism stage , Nodal/Univin signals emitted from this organizer are responsible for transferring left-right asymmetry from this mesendodermal organizer to the lateral ectoderm by inducing nodal and univin expression in cells located on the right side of the ventral ectoderm and ciliary band .
The Notch signaling pathway plays a key role in establishment of left right asymmetry in vertebrates . However , the mechanisms by which Notch acts in this pathway differ significantly between the mouse and the zebrafish . Genetic analysis in the mouse showed that expression of nodal in the node is crucial for subsequent propagation of nodal expression to the lateral plate [22] , [23] . Several studies have demonstrated that perturbations of the Notch pathway strongly affect this early expression of nodal in the node and disrupt establishment of left-right asymmetry . Embryos mutant for Delta1 , or double mutant for Notch1 and Notch2 or lacking the function of CSL , ( the main transcriptional effector of the Notch pathway ) , fail to express nodal in the node and subsequently are unable to establish the left-sided expression of nodal in the lateral plate [35]–[37] . Indeed , expression of nodal in the node is directed by a cis-regulatory module that contains binding sites for CSL and mutations of these sites abolish the activity of this enhancer . In zebrafish , however , nodal expression in the node is not eliminated by disruption of Delta/Notch signaling . In this case , Notch signaling appears to control cilia length in the Kupffer's vesicle by regulating the expression of the master cilia regulator foxJ1 [75] . Another primary target of Notch signaling in the zebrafish appears to be the gene encoding the Nodal antagonist of molecule Charon [75] , [76] . Charon is first expressed symmetrically in the node region , then asymmetrically with a stronger expression on the right side of the node where Charon antagonizes Nodal signaling . The finding that in Delta mutants or following DAPT treatments , expression of charon , but not nodal expression , is strongly reduced and the presence of several CSL binding sites in the charon promoter strongly suggest that Notch signaling regulates charon expression in the zebrafish . While in the mouse inhibition of Notch signaling prevents nodal expression , in zebrafish , inhibition of Notch signaling causes instead nodal to be expressed bilaterally in the lateral plate [76] . Our results clearly showed that the Notch pathway also plays a crucial role during establishment of left-right asymmetry in the sea urchin embryo . Inhibition of Notch signaling by injection of morpholino directed against Delta or treatment of embryos with a γ-secretase inhibitor caused bilateral expression of nodal in the endoderm at gastrula stage and randomized nodal expression in the ciliary band at later stage . The function of Notch signaling in the sea urchin embryo therefore does not appear to be in the activation of nodal expression like in the mouse , but instead in the repression of nodal expression on the left side , like in the zebrafish , since inhibition of Notch signaling caused bilateral expression of nodal in the endomesoderm . How Notch signaling promotes unilateral expression of nodal on the right side in the sea urchin is presently unclear . Since the mesodermal precursors lie immediately on the top of the invaginated archenteron , and since Notch signaling is primarily required for specification of these mesodermal precursors , one possibility is that Notch signaling is required early for specification of mesodermal cells , which in turn send an inhibitory signal during gastrulation that prevents nodal expression on the left side of the underlying endoderm ( Figure 10 ) . Alternatively , Notch may be required for the correct positioning of a signal that induces Nodal expression on the right side . A third possibility is that , by analogy to the role of Notch signaling in the chick , Notch signaling may regulate cell rearrangements that would be required for establishment of left-right asymmetry . In line with this idea , previous studies reported that the progeny of the small micromeres partition asymmetrically into the two coelomic pouches with the left coelomic pouch inheriting a larger fraction than the right coelomic pouch [43] . It is important , however , to keep in mind that the period during which Notch signaling is required for correct right sided expression of nodal is separated from the onset of nodal expression in the archenteron by 15 h and therefore that the effect of Notch signaling on nodal is most likely very indirect . Future studies are required to understand how Notch signaling regulates left-right asymmetry in the sea urchin embryo . In particular , the identity of the inhibitory signal X remains to be established . It is interesting to draw a parallel between the repressive effect of the non-skeletogenic mesoderm on endodermal precursors of the left-right organizer and the repressive effects that the PMCs exert on the non skeletogenic mesodermal precursors . When the skeletogenic precursors ( micromeres or PMCs ) are removed , non skeletogenic precursors transfate to replace the missing skeletal precursors [77] . It will interesting to determine if the repressive effects of the PMCs on SMC conversion to a skeletogenic fate and the repressive effects of the non skeletogenic precursors on endodermal precursors conversion into a nodal expressing left-right organizer rely on similar molecular mechanisms . Finally , ablation of micromeres at the 16-cell stage has been reported to perturb left-right asymmetry and to randomize positioning of the rudiment suggesting that micromeres release a signal that regulates left-right asymmetry [46] . Our results strongly suggest that this signal is Delta , which is expressed in the progeny of the large micromeres where it induces non skeletogenic mesoderm precursors from surrounding endomesodermal precursors [42] , [78] , [79] . One of the most striking results of our study is that treatments with the H+/K+-ATPase inhibitor omeprazole mimicked inhibition of Notch signaling in the early embryo . Treatments with omeprazole , like injection of the Delta morpholino or treatments with DAPT , abolished formation of non-skeletogenic mesodermal precursors causing a strong delay in gastrulation [80] , [81] and resulting in gastrulae with a smooth archenteron , devoid of secondary mesenchymal cells , and later , in larvae lacking pigment cells [41] , [78] . At the molecular level , expression of several marker genes expressed in the secondary mesodermal precursors ( gcm , papss and GATA1/2/3 ) was abolished following inhibition of the H+/K+-ATPase . Therefore , both in the context of establishment of left-right asymmetry and in the context of induction of the germ layers , omeprazole treatments mimicked inhibition of Notch signaling . One study had implicated the activity of the H+/K+-ATPase in the modulation of Notch signaling at the extracellular level . In the chick , the activity of H+/K+-ATPase has been associated with a transient left-right accumulation of extracellular calcium and this transient rise in extracellular calcium has been proposed to promote Notch signaling partly by promoting asymmetrical expression of Delta [33] . It is very unlikely that Notch activity is regulated by an increase in extracellular calcium in the sea urchin since this organism develops in an environment that already contains an extremely high ( 10 mM ) concentration of extracellular calcium . More recent studies have implicated Wnt signaling in the regulation of foxJ1 [82] , and the activity of the H+/K+-ATPase in canonical Wnt signaling [34] . In the sea urchin embryo , the phenotypes caused by omeprazole treatment are very different from those resulting from inhibition of Wnt signaling [83] , [84] . Furthermore , we showed that omeprazole treatment did not interfere with the Wnt dependent expression of endodermal marker genes such as foxA , ruling out a role for the H+/K+-ATPase in the Wnt pathway . Instead , omeprazole specifically interfered with expression of mesodermal markers , indicating a more direct role in the Notch pathway . To our knowledge , this is the first report that the activity of H+/K+-ATPase is fundamental for Notch signaling . So how may the activity of the H+/K+-ATPase regulate Notch signaling ? Two recent studies reported that Delta-Notch signaling is highly pH dependent and that the activity of the V-ATPase , a proton pump that controls the acidity of lysosomes , plays a central role in Notch signaling . In one study Vaccari and coll . showed that cells mutants for the V-ATPase accumulate an uncleaved form Notch in the endosomes and lysosomes and fail to activate Notch signaling [85] . Similarly , in a screen for mutations that disrupt the Notch pathway , Yan et al found that mutations that inactivate genes encoding either Rabconnecting 3 ( Rbcn3 ) , a known regulator of V-ATPase in yeast , or VhaC39 , a gene encoding a subunit of the V-ATPase , recapitulate a number of phenotypes caused by inactivation of the Notch pathway including defective oogenesis and abnormal patterning of imaginal discs [86] . Cells lacking Rbcn3 or VhaC39 function fail to acidify intracellular compartments and accumulate Notch in late endosomes . How the function of V-ATPases regulates Notch signaling is presently unknown but a number of studies have implicated V-ATPases in the regulation of a number of essential cellular processes such as endocytosis , lysosomal degradation or secretion . It is therefore possible that V-ATPase is required for trafficking of Notch or Delta . Another possibility is that the V-ATPase mediated acidification is required for generation of NICD , the intracellular and active form of Notch . The active form of Notch requires two successive proteolysis events mediated by ADAM metalloprotease and γ-secretase [58] . Interestingly , in Drosophila , expression of NICD , the form of Notch generated by γ-secretase cleavage , but neither expression of full length Notch nor expression of NEXT , can rescue the defects caused by inactivation of Rbcn3 or V-ATPase function , strongly suggesting that V-ATPase is required at or downstream of γ-secretase-mediated S3 cleavage of NEXT [86] . In the sea urchin embryo , omeprazole treatment inhibited the stimulation of Notch signaling induced by overexpression of Delta but had no effect on overexpression of NEXT or NICD . Therefore , omeprazole treatments appear to affect a step located at or upstream of the S2 mediated cleavage of Notch . Since S2 cleavage is mediated by secreted metalloproteases of the ADAM/TACE/Kuzbanian family , one possibility is that the activity of the H+/K+-ATPase is required for the activity of these enzymes . Alternatively , the activity of the H+/K+-ATPase may be required in the signal sending cells through processes such as trafficking or endocytosis of Delta . The localization of the H+/K+-ATPase on the apical surface of epithelial cells is consistent with these proposed roles [87] . The activity of the H+/K+-ATPase was previously shown to be essential for establishment of left-right asymmetry in zebrafish and Xenopus . However , to our knowledge , its role in the regulation of Notch signaling had never been investigated . We showed that the function of the H+/K+-ATPase is mandatory for Notch signaling in the sea urchin embryo and that embryos treated with omeprazole fail to express Notch target genes and later display randomized expression of nodal along the left-right axis . Therefore our results tie together and extend different observations on the roles of proton pumps on establishment of left-right asymmetry and Notch signaling . Studies in vertebrates suggested the existence of three distinct steps in the establishment of left-right asymmetry: symmetry breaking , initiation of asymmetric expression of nodal in a left-right organizing center and propagation of left-right asymmetry to more distant tissues . These three steps can be identified during establishment of L/R asymmetry in the sea urchin embryo and although the picture is still incomplete , what emerges from this study is that there are both conserved as well as to notably divergent features in the strategies and mechanisms used in echinoderms and vertebrates to establish left-right asymmetry . The role of a discrete mesendodermal region playing the role of a left-right organizer emerges as a conserved feature . Similarly , the implication of BMP signaling in the regulation of nodal expression is another feature that appears to be conserved between sea urchin and vertebrates . Finally , the role of Vg1/GDF1 in promoting propagation of left-right asymmetry to more distant regions is a third feature that appears to be common to sea urchin and vertebrate embryos . In contrast , the role of Notch may not be conserved since in the sea urchin , unlike in vertebrates , the role of Notch signaling appears to be very indirect and temporally separated from nodal expression . In vertebrates , the node plays the function of a left-right organizer . Left-right asymmetry first becomes apparent in and around the node and subsequently propagates to the rest of the embryo . In the sea urchin embryo , the first manifestation of left-right asymmetry is expression of nodal on the right side of the tip of the archenteron . Several lines of evidence strongly suggest that this asymmetry of mesendodermal precursors is crucial for establishment of left-right asymmetry in the ectoderm and that this asymmetry is transmitted to the ectoderm at later stages resulting in right-sided expression of nodal in the ciliary band ( Figure 9 ) . First , both the absence of nodal expression and bilateral expression of nodal in the mesoderm result in random expression of nodal in the ectoderm . Second , inhibition of Univin function on the right side forced nodal to be expressed on the left side of the ciliary band . Finally and importantly , using chimeras , we showed that inhibition of nodal function in the endomesoderm randomizes nodal expression in the ectoderm . Our results are largely consistent with results of Amemiya et al . who showed that ablation at gastrula stage of the tip of the archenteron together with part of the ectoderm on the right side reversed positioning of the rudiment in 70% of the embryos while excision that removed the right ectoderm but left the archenteron intact had a more much more modest effect on left-right asymmetry , reversing positioning of the rudiment in only 30% of the embryos [45] . We therefore propose that the nodal expressing mesodermal cells located at the tip of the archenteron may therefore play the role of a left-right organizer similar to the node of vertebrates ( Figure 11 ) . However , this organizer is only responsible for orienting the symmetry breaking and for making it directional . Left right asymmetry can be established in the absence of this organizer but it is not directional . There is accumulating evidence that the BMP pathway plays a dual and crucial role in vertebrates both in promoting expression of nodal on the left side and in preventing nodal activation on the right side . Nearly as many studies have implicated BMP signaling in the repression of nodal expression on the right side [88]–[93] as in the positive regulation of nodal expression on the left side [94]–[97] . For example in the mouse embryo , a reduction of BMP signaling causes nodal to be expressed bilaterally in the lateral plate . In the sea urchin , inhibition of BMP signaling by injection of a bmp2/4 or alk3/6 morpholino into the egg or blocking BMP signaling specifically in the endomesoderm prevented nodal expression in the organizer at gastrula stage and randomized nodal expression at pluteus stage . Intriguingly , targeting of the BMP2/4 morpholino to either the left or the right side revealed that BMP signaling on the left side is required for nodal expression on the right side . Consistent with this idea , we found that BMP signaling is stronger on the left side of the archenteron at gastrula stage and that the sector in which pSmad1/5/8 is detected and the region where the nodal expressing left-right organizer is formed are complementary . Furthermore , the asymmetry of nodal expression in the left-right organizer was detected slightly before the asymmetry of BMP signaling . It is therefore likely that an initially symmetric BMP signaling participates in the induction of nodal expression on the right side and that asymmetric Nodal signaling is in turn responsible for the asymmetry of BMP signaling possibly by antagonizing BMP signaling in the dorsal-right sector of the endomesoderm . The fact that all the genes encoding BMP ligands and BMP antagonists are expressed symmetrically along the left-right axis ( our unpublished data ) is consistent with this idea . Taken together , these observations suggest that formation of the left-right organizer is regulated by a combination of both positive and negative regulatory interactions ( Figure 10 ) . On the left side of the archenteron , a repressive signal produced by the secondary mesoderm prevents nodal expression . On the right side of the archenteron , three signals cooperate to induce nodal in the left-right organizer . The first signal is Nodal/Univin produced from the ventral ectoderm , the second signal is a member of the FGF family of growth factors ( the tissue that produces it is presently not identified ) , and the third signal is likely produced in the dorsal part of the archenteron downstream of BMP signaling . In vertebrates , left-right asymmetry propagates from the node to the lateral plate . Elegant rescue experiments using transgenic lines driving expression of GDF1 in the node or in the lateral plate demonstrated that the activity of GDF1 in the node is required for expression of nodal in the lateral plate [23] , [24] . In addition communication between the node and the lateral plate has been recently shown to require functional gap junctions in the adjacent endodermal cells [98] . It is unlikely that gap junctions are involved in long range communication between the ectoderm and the endomesoderm since genes encoding gap junction proteins ( connexins , innexins ) are absent from the sea urchin genome [99] . In contrast , we showed that Univin , a TGF beta related to Vg1 and GDF1 , is critically required for long range signaling between the ectoderm and the endomesoderm and for propagation of the left-right asymmetry signal . This suggests that the role of Univin as a TGF beta critically required for long range signaling by Nodal during left-right patterning is an evolutionary conserved and probably ancient feature in the left-right determination pathway ( Figure 11 ) . In vertebrates the expression of lefty1 in the midline is thought to play a crucial role in the initiation and maintenance of unilateral expression of nodal [100] , [101] . In the sea urchin embryo , there is presently no argument to suggest that there is a midline similar to the lefty expressing midline of vertebrate embryos that would act as a barrier to prevent propagation of nodal expression to the right side . Consistent with this idea , lefty in the sea urchin embryo is not expressed in the midline . Despite the absence of expression of lefty in the midline , a robust expression of nodal on the right side of sea urchin embryos is established at the end of gastrulation . How is this asymmetric expression established ? There is strong evidence that in the sea urchin like in vertebrates [102] , the epigenetic system constituted by short range Nodal autoregulation and long range inhibition by Lefty plays a crucial role in restricting nodal expression [103] . Lefty is both a very potent and highly diffusible inhibitor of Nodal signaling in the sea urchin embryo and lefty expression shifts to the right side at the end of gastrulation . Any small bias of nodal expression towards the right side will therefore be amplified and maintained by the self enhancement and lateral inhibition mechanism resulting in a robust expression of nodal and lefty on the right side in the absence of any midline barrier . We propose that the function of the left-right mesendodermal organizer on the right side of the archenteron is to provide this initial bias of nodal expression and that the reaction-diffusion mechanism between Nodal and Lefty further amplifies this bias , establishing a stable nodal expression on the right side . Of the three steps involved in establishment of left-right asymmetry , the first i . e . symmetry breaking , remains the most enigmatic . Our data in the sea urchin embryo , point to the endomesoderm as the site where the symmetry is first broken and identify the Notch , FGF and BMP signaling pathways as critical early actors in the molecular cascade leading to determination of laterality . However , many questions remain on the mechanism by which Notch signaling represses nodal expression on the left side . Does Notch signaling regulate nodal expression by promoting asymmetrical cell movements , as proposed in the chick or does Notch signaling regulate nodal expression by promoting the local production by mesodermal cells of a factor that inhibits nodal expression ? To answer these questions , future experiments should attempt to identify the inhibitory signal X that prevents nodal expression on the left side and should define the identity of the cells that send it . Similarly , the identity of the FGF ligand that promotes nodal expression on the right side is presently unknown and whether there is any connection between these inhibitory ( Notch/factor X ) or activating ( FGF , BMP ) signals remains to be explored . Future experiments should also examine the mechanisms responsible for asymmetrical BMP signaling in the archenteron and clarify the mechanisms by which BMP signals promote nodal expression . Finally , two important questions that future experiments should address are: to what extent is the left-right organizer of the sea urchin embryo homologous to the left-right organizer of vertebrates and do the archenteron tip cells require cilia to fulfill their role of left-right organizing cells ? In conclusion , our results provide a framework for the future dissection of the molecular pathway that regulates establishment of left-right asymmetry in the sea urchin . Furthermore , they demonstrate a strong connection between two players of the left-right determination pathway that were previously thought to be largely independent: the H+/K+-ATPase and Notch signaling . Finally , in addition to regulating left-right asymmetry , Notch signaling plays multiple and crucial roles in the etiology of various cancers [104] and particularly in acute T cell leukemia ( T-ALL ) . Our finding that omeprazole , an extremely well tolerated and world-wide standard drug used to treat gastritis and ulcers , inhibits Notch signaling in the sea urchin embryo may be of clinical interest . In line with this idea , previous studies reported that omeprazole has an antiproliferative effect on pancreatic or colon cancer cells leading to the suggestion that omeprazole treatments could be used to develop new therapeutic strategies [105] , [106] . Our finding that omeprazole inhibits Notch signaling in echinoderm embryos raises the possibility that the effect of omeprazole on tumor reversion may be linked to inhibition of Notch signaling , an hypothesis that should be investigated in future studies .
Adult sea urchins ( Paracentrotus lividus ) were collected in the bay of Villefranche-sur-Mer . Embryos were cultured at 18°C in Millipore-filtered sea water and at a density of 5000 per ml . Fertilization envelopes were removed by adding 1 mM 3-amino-1 , 2 , 4 triazole ( ATA ) 1 min before insemination to prevent hardening of this envelope followed by filtration through a 75 µm nylon net [107] . Treatments with the γ-secretase inhibitor DAPT ( 10–30 µM in sea water , Calbiochem ) , omeprazole ( 150–200 µM in sea water , Sigma ) , U0126 ( 5–10 µM in sea water , Calbiochem ) , SU5402 ( 30–50 µM in sea water , Calbiochem ) were performed by adding the chemical diluted from stocks in Dimethylsulfoxyde ( DMSO ) in 24-well plates protected from light at the desired time . As controls , DMSO was added alone at 0 . 1% final concentration . Treatments by these inhibitors were performed continuously starting after fertilization . Treatments with recombinant BMP4 protein ( 0 . 5 µg/ml ) were started at the 16-cell stage . Experiments involving treatments with pharmacological inhibitors ( DAPT , omeprazole , U0126 , SU5402 ) were repeated multiple times with the same results . Larvae were reared in 2-liter beakers with constant stirring at a density of one larva per 5 ml . They were fed every day with a freshly grown culture of the unicellular alga Isocrysis thaliana at a density of about 1000–5000 cells per ml . The presence and position of the rudiment was scored with a dissecting microscope after 3–4 weeks of culturing , and the larvae were photographed with a Zeiss Axiophot with dark-field and DIC illumination . To observe metamorphosis , single larvae competent to metamorphose were transferred to a Petri dish and observed at regular intervals . Metamorphosis was usually completed in 1–3 h . Embryos devoid of fertilization envelopes were operated in Ca2+-free artificial sea water . Embryos microinjected with the nodal-Morpholino and a fluorescein-lysine dextran ( FLDX ) at the 16–32-cell stage were placed in a Kiehardt chamber on a dissecting microscope and vegetal halves were recombined to animal halves of unlabeled control embryos at the same stage in Ca2+-free seawater . Thirty-six hours post-fertilization , the embryos were imaged using a fluorescent microscope to record morphology and the presence of the dyes . The embryos were then fixed individually and analyzed by in situ hybridization with a nodal probe . Immunostaining with the phosphoSmad1/5/8 antibody was performed as described by Lapraz et al . 2009 [65] . Morpholino antisense oligonucleotides were obtained from Gene Tools LLC ( Eugene , OR ) . Characterization of the nodal , BMP2/4 , univin , alk4/5/7 and alk3/6 morpholinos has been described in [65] , [108] , [109] . The specificity of the alk4/5/7 , alk3/6 and nodal morpholinos has been demonstrated by rescue experiments . In the case of Delta , we designed and tested two morpholinos . The phenotypes observed with the Delta morpholino were considered specific since this morpholino caused a phenotype identical to the phenotype caused by DAPT treatment or by injection of a dominant negative form of Delta ( truncation of the cytoplasmic domain ) . This phenotype is characterized by development of embryos lacking secondary mesenchymal cells at the tip of the archenteron during gastrulation [42] and lacking pigment cells and blastocoelar cells at later stages [48] . The phenotypes observed were therefore very consistent with the zygotic expression pattern and with previous well-established functional data . These phenotypes are very similar to those caused by inhibition of Notch signaling in other species [42] , [79] . The sequences of all the morpholino oligomers used in this study are listed below . The most efficient morpholino of each pair is labeled with a star . Delta morpholinos are both directed against the 5′ UTR of the Delta transcript . Delta Mo1*: 5′-GTGCAGCCGATAGCCTGATCCGTTA-3′ . Delta Mo2: 5′-CTTTTCTTATCAGTCCAAACCAGTC-3′ . univin Mo1*: 5′-ACGTCCATATTTAGCTCGTGTTTGT-3′ . univin Mo2: 5′-GTTAAACTCACCTTTCTAAACTCAC-3′ . nodal Mo1*: 5′-ACTTTGCGACTTTAGCTAATGATGC-3′ . nodal Mo2: 5′-ATGAGAAGAGTTGCTCCGATGGTTG-3′ . alk4/5/7 Mo 1: 5′-TAAGTATAGCACGTTCCAATGCCAT-3′ . alk3/6: Mo1: 5′-TAGTGTTACATCTGTCGCCATATTC-3′ . bmp2/4 Mo1*: 5′-GACCCCAGTTTGAGGTGGTAACCAT-3′ . bmp2/4 Mo2: 5′-CATGATGGGTGGGATAACACAATGT-3′ . Morpholino oligonucleotides were dissolved in sterile water and injected at the one-cell stage together with Tetramethyl Rhodamine Lysine Dextran ( RLDX ) ( 10000 MW ) at 5 mg/ml or Fluoresceinated Dextran ( FLDX ) ( 70000 MW ) at 5 mg/ml . Fluoresceinated Dextran is used as a lineage tracer of the injected cell . For each morpholino a dose-response curve was obtained and a concentration at which the oligomer did not elicit non-specific defect was chosen . Approximately 2–4 pl of oligonucleotide solution at 0 . 5 mM were used in most of the experiments described here . For morphological observations , about 150–200 eggs were injected in each experiment . To analyze gene expression in the morphants a minimum of 50–75 injected embryos were hybridized with a given probe . All the experiments were repeated at least twice and only representative phenotypes observed in more than 80% of embryos are presented . Synthesis of capped mRNA coding for Nodal and Univin are respectively described in [66] and [109] . The pCS2 Delta construct is described in [48] . The Notch NICD and NEXT constructs were derived from a full length Paracentrotus lividus cDNA clone . For the NEXT construct the coding sequence of Notch corresponding to the aminoacids 1570–2528 of Notch ( from the lin12 repeats up to the end of the protein ) was amplified and cloned in pCS2 . For the NICD construct , a region corresponding to aminoacids 1728–2528 of Notch ( starting immediately after the transmembrane domain and extending to the end of the protein ) was similarly cloned into pCS2 . Delta induced overproduction of pigment cells when injected at 500 µg/ml while mRNA encoding NEXT caused the same effect when injected at 1 mg/ml and mRNA encoding NICD when injected at 200 µg/ml . The Genebank accession numbers for the sequences discussed in this paper are: Notch ( JQ861276 ) , Nodal ( AAS00534 ) , BMP2/4 ( DQ536194 ) , Alk3/6 ( FJ976181 ) , FoxA ( ABX71819 ) , Univin ( ABG00200 ) , Pitx2 ( AAW51825 ) , Sox9 ( AAW51826 ) , Delta ( ABG00198 ) , Gcm ( ABG66953 ) , PAPSS ( DQ531774 ) , GATA1/2/3 ( ABX71821 ) . Gene regulatory network diagrams were constructed using the biotapestry program available at http://www . biotapestry . org/ [110] . Dual luciferase assays were performed with the Promega Dual Luciferase Reporter system ( Promega ) . Microinjection of purified and linearized plasmids was carried out by established protocols [111] . In the case of RBPJ-K luciferase reporter , the linearized plasmid was injected at 3 . 5 µg/ml , together with Endo 16-Renilla DNA at 1 ng/µl and carrier DNA ( Hind III digested sea urchin DNA ) at 17 µg/ml . For induction of Delta/Notch signaling , Delta mRNA was used at 500 µg/ml , NICD ( Notch Intracellular Domain ) mRNA at 200 µg/ml and NEXT ( Notch extracellular truncation ) RNA at 1000 µg/ml . For each measurement , 200 embryos were injected , collected at hatching blastula stage then lyzed following the manufacturer's instructions . The level of RBP1 derived Firefly Luciferase was detected according to the manufacturer's instructions using a GloMax luminometer with an integration of 10 s . The level of luciferase activity was normalized to the level of Renilla activity for each experiment . All the experiments were repeated two to three times using separate batches of embryos . In situ hybridization was performed following a protocol adapted from Harland et al . 1991 [112] with antisense RNA probes and staged embryos . Probes derived from pBluescript vectors were synthesized with T7 RNA polymerase after linearization of the plasmids by NotI , while probes derived from pSport were synthesized with SP6 polymerase after linearization with SfiI . Control and experimental embryos were developed for the same time in the same experiments . The nodal , univin , pitx2 , sox9 probes have been described already respectively in [10] , [66] , [109] . For double fluorescent in situ hybridizations , embryos were incubated overnight in hybridization buffer with the two probes . The nodal probe was labeled with Digoxigenin ( DIG mix from Roche- Ref: 11277073910 ) ; The foxA and foxF probes were labeled with fluorescein ( Fluo Mix frome Roche- Ref: 11427857910 ) . After washing of the probes , embryos were incubated with an Anti-Digoxygenin Antibody coupled to HRP ( Roche-ref: 11 207 733 910 ) , diluted at 1/2000 overnight at 4°C , and staining was developed with the Cy3-Tyramide Signal Amplification System ( TSA-Plus Kit-Perkin Elmer-Ref: NEL753 ) . Embryos were rinsed with TBST until disappearance of background . The anti-digoxygenin-HRP antibodies were removed by treatment with Glycine , 0 , 1 M pH: 2 . 2 , H2O2 1% , Tween 0 . 1% in TBST , and embryos were incubated with the Anti-Fluorescein Antibody coupled with HRP ( Roche- Ref: 11 426 346 910 ) , diluted 1/2000 during two hours at room temperature , and revealed with Cy2-tyramide signal amplification . Embryos were rinsed with TBST then mounted with City fluor and observed with a DIC and fluorescence microscope Axioimager . To visualize the clones of injected cells after in situ hybridization , we used an antibody against fluorescein coupled to alkaline phosphatase . At the end of the in situ hybridization protocol , embryos were rinsed with PBST+EDTA 5 mM then incubated in a buffer containing glycine 0 . 2 M pH: 2 . 2 , Tween 0 . 1% to inactivate the anti-digoxigenin antibody . Embryos were then washed six times in PBST , incubated in blocking solution ( 1% BSA , 2% Sheep serum inactivated in TBST ) then with the anti-Fluorescein antibody coupled to Alkaline phosphatase ( 1/4000 ) at 4°C overnight . For Alkaline phosphatase staining , embryos were washed six times with TBST and briefly rinsed in Tris 100 mM pH: 8 . 2 and stained using FastRed as substrate in Tris 100 mM pH: 8 . 2 . Staining was stopped by four rinses with PBST+EDTA 5 mM , then two rinses with PBST 25% Glycerol and 50% Glycerol . Embryos were then mounted and observed with a DIC microscope . | Asymmetries between the left and the right sides of the body are an essential feature of most bilaterian animals , and failure to establish these asymmetries can result in pathological disorders in humans . Left-right asymmetries are established during early development by the asymmetric activity of a signaling pathway in a discrete region of the embryo that plays the role of a left-right axis organizer . Although the role of this signaling pathway appears to be conserved among vertebrates , whether the mechanisms involved in the initial breaking of the symmetry and in the establishment of the left-right organizer are also conserved remains an open question . We report that left-right axis determination in the sea urchin embryo also relies on the activity of a left-right organizer located within the gut of the sea urchin embryo . We also report the unexpected finding that the activity of the H+/K+-ATPase , a previously known but enigmatic player in this pathway , is critically required for activation of the Notch receptor . Our results therefore open the way to analysis of the molecular pathway that regulates establishment of laterality in the sea urchin embryo and uncover a functional link between two essential players of left-right asymmetry . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"evolutionary",
"biology",
"biology",
"pattern",
"formation",
"evolutionary",
"developmental",
"biology"
] | 2012 | Reciprocal Signaling between the Ectoderm and a Mesendodermal Left-Right Organizer Directs Left-Right Determination in the Sea Urchin Embryo |
The misexpressed imprinted genes causing developmental failure of mouse parthenogenones are poorly defined . To obtain further insight , we investigated misexpressions that could cause the pronounced growth deficiency and death of fetuses with maternal duplication of distal chromosome ( Chr ) 7 ( MatDup . dist7 ) . Their small size could involve inactivity of Igf2 , encoding a growth factor , with some contribution by over-expression of Cdkn1c , encoding a negative growth regulator . Mice lacking Igf2 expression are usually viable , and MatDup . dist7 death has been attributed to the misexpression of Cdkn1c or other imprinted genes . To examine the role of misexpressions determined by two maternal copies of the Igf2/H19 imprinting control region ( ICR ) —a chromatin insulator , we introduced a mutant ICR ( ICRΔ ) into MatDup . dist7 fetuses . This activated Igf2 , with correction of H19 expression and other imprinted transcripts expected . Substantial growth enhancement and full postnatal viability was obtained , demonstrating that the aberrant MatDup . dist7 phenotype is highly dependent on the presence of two unmethylated maternal Igf2/H19 ICRs . Activation of Igf2 is likely the predominant correction that rescued growth and viability . Further experiments involved the introduction of a null allele of Cdkn1c to alleviate its over-expression . Results were not consistent with the possibility that this misexpression alone , or in combination with Igf2 inactivity , mediates MatDup . dist7 death . Rather , a network of misexpressions derived from dist7 is probably involved . Our results are consistent with the idea that reduced expression of IGF2 plays a role in the aetiology of the human imprinting-related growth-deficit disorder , Silver-Russell syndrome .
Parthenogenetic mouse embryos usually die before 6½ days post coitum ( dpc ) . Occasionally they develop to the 25 somite forelimb bud stage or approximately 9½ dpc [1]–[5] . Parthenogenones possess two maternally-derived genomes and would be expected to possess abnormal levels of transcript of all known imprinted genes , that is , lack of expression of paternally expressed genes ( two inactive copies ) , and over-expression of maternally expressed genes ( two active copies ) . Their death is likely a composite effect of at least some of these misexpressions , although those involved are not well defined . Defining the causes is important for improving understanding of the aetiology of genomic imprinting [6]–[9] and the prevalence of sexual reproduction , which ‘has long been an evolutionary enigma’ [10] . Knowledge of the causes of parthenogenetic death has come from two sources . First , the union of unbalanced complementary gametes in intercrosses of mice carrying reciprocal or Robertsonian translocations yield , at low frequency , embryos with maternal duplication and paternal deficiency for particular Chr regions as defined by the translocation breakpoint [11]–[13] . Maternal duplication of twelve Chr regions results in developmental anomalies . Only three of these are associated with peri- or prenatal death , these being maternal duplication of proximal Chr 6 ( MatDup . prox6 ) —prior to 11½ dpc [14] , maternal duplication of distal Chr 7 ( MatDup . dist7 ) —late fetal death [15] , and maternal disomy of Chr 12—perinatal death , probably attributable to the distal region [16] . Second , knockouts of imprinted genes and imprinting control regions ( ICRs ) have provided information on the effects of disregulation of imprinted genes , for example , [17]–[21] . To better define the causes of failed parthenogenetic development , and learn more of how imprinted genes at dist7 work together to regulate normal development , we have examined some of the misexpressions of imprinted genes thought to contribute to the abnormal development of MatDup . dist7 conceptuses . These display a pronounced growth deficit of the fetus and placenta and die at the late fetal stage , or possibly at birth . Live MatDup . dist7 young have never been observed [13] , [15] ( J . Mann , unpublished data ) . Dist7 is an important region in terms of genomic imprinting , containing over 20 imprinted genes [13] , [22] . At least three of these are regulated by the Igf2/H19 imprinting control region ( ICR ) , these being ‘insulin like growth factor 2’ ( Igf2 ) —paternally expressed and encoding a mitogen important for embryonic growth [23] , [24] , ‘insulin II’ ( Ins2 ) —paternally expressed in yolk sac [25] , and the non-coding ‘H19 fetal liver mRNA’ ( H19 ) gene—maternally expressed [26] . Other non-coding transcripts have been described , these being Mir483 , contained within an intron of Igf2 [27] and for which imprinting status is unknown , Mir675 , contained with an H19 exon and therefore likely to follow the imprinting pattern of the host gene [28] , [29] , and antisense transcripts within Igf2 [30] . The targets of the Mir483 and Mir675 miRNAs are unknown . The maternally-derived Igf2 allele is inactive due to the hypo-methylated maternal Igf2/H19 ICR functioning as a‘CCCTC-binding factor’ ( CTCF ) -based chromatin insulator . This lies between the Igf2 promoter and the shared Igf2-H19 enhancers , preventing their interaction . The maternal H19 promoter lies on the same side of the insulator as the enhancers , therefore interaction occurs . On the paternal Chr the ICR is hyper-methylated , preventing CTCF binding and insulator formation and allowing for paternal Igf2 promoter and enhancer interaction . The paternal H19 promoter , just distal to the methylated ICR , also becomes methylated , and is inactive . The Ins2 gene is located just distal to Igf2 . The Ins2 parental alleles are affected in the same way as their Igf2 counterparts , but only in yolk sac . Ins2 is expressed biallelically in pancreas [25] , [31]–[33] . Telomeric or distal to the Igf2/H19 ICR domain is a large cluster of imprinted genes under regulatory control of the Kv differentially methylated region ( DMR ) -1 ( KvDMR1 ) ICR . The active state of maternally-derived genes within this cluster is coincident with maternal-specific ICR methylation and the inactive state of the promoter of the ‘KCNQ1 overlapping transcript 1’ ( Kcnq1ot1 ) gene contained within the ICR . The paternal ICR is hypo-methylated , and paternal-specific elongation of the Kcnq1ot1 transcript is coincident with silencing in cis of genes within the cluster [17] , [34] , [35] . One of the genes regulated by this ICR is the ‘cyclin-dependent kinase inhibitor 1C ( P57 ) ’ ( Cdkn1c ) gene encoding a protein facilitating reduced cell proliferation , increased apoptosis and delayed cell differentiation [36] , [37] . MatDup . dist7 fetuses are maternally duplicated for the hypo-methylated Igf2/H19 ICR and hyper-methylated KvDMR1 ICR regions , as well as for other imprinted transcripts at dist7 . This epigenetic configuration is highly similar to that associated with the human imprinting-related growth deficit disorder , Silver-Russell syndrome ( SRS ) ( OMIM 180860 ) . More than half of cases are associated with hypo-methylation of the IGF2/H19 ICR , also known as ‘ICR1’ . The disease is also associated with maternal duplication of the KvDMR1 ICR region , also known as ‘ICR2’ , and maternal duplication of the 11p15 . 5 Chr region encompassing both ICRs . It is strongly suspected that SRS is caused by downregulation of IGF2 , and , in a minority of cases , excess CDKN1C or other imprinted genes regulated by ICR2 . However , empirical evidence is lacking [38]–[40] . The death of MatDup . dist7 fetuses has been difficult to decipher . Available evidence suggests that maternal duplication of the Igf2/H19 ICR regulatory domain alone is insufficient to explain the total phenotype observed . Mice with paternal inheritance of a tandem duplication of a chicken β-globin CTCF-based chromatin insulator , substituted for the endogenous Igf2/H19 ICR , are similar to MatDup . dist7 mice in having a fully functional hypo-methylated insulator on both parental Chrs . They lack Igf2 activity , have at least twofold over-expression of H19 , with both parental alleles probably active , and would be expected to lack Ins2 activity in yolk sac . Nevertheless , their phenotype—dwarfism combined with postnatal viability—is essentially identical to Igf2 mutants [41] . Mice homozygous for this genetic modification , in a mix of strains 129S1/SvImJ and outbred Swiss CF-1 , showed normal fecundity and were maintained as a random-bred line for several years ( J . Mann , unpublished data ) . Further , lack of Igf2 activity is unlikely to be the sole cause of reduced growth in MatDup . dist7 fetuses . At 17½ dpc , their weight is approximately 40% of wild-type [42] ( J . Mann and Walter Tsark , unpublished observations ) compared to 50–60% of wild-type for Igf2 mutants and mice maternally inheriting the chicken insulator [41] . Overall , these observations indicate that the MatDup . dist7 phenotype of fetal growth deficit and death involves the misexpression of imprinted genes outside the influence of the Igf2/H19 ICR , and this has previously been suggested [42] . Available evidence also indicates that maternal duplication of the KvDMR1 ICR regulatory domain alone is insufficient to explain the total phenotype observed . Mice with paternal inheritance of a deletion of this element exhibit biallelic expression of adjacent imprinted genes . These mice , in a mix of mouse strains 129S4/SvJae and C57BL/6J , are postnatally viable . They show some reduction in size , and it has been indicated that this is caused by over-expression of Cdkn1c [35] . Reduced growth has also been observed in Cdkn1c-BAC transgenic mice . While these displayed high frequency perinatal mortality in strain 129/Sv , high postnatal viability was obtained in a mix of strains 129/Sv and outbred Swiss MF1 [43] . These observations indicate that MatDup . dist7 late fetal death , occurring in the context of mixed strains including outbred Swiss , involves the misexpression of imprinted genes outside the influence of the KvDMR1 ICR . Overall , these observations have led to suggestions that MatDup . dist7 death could be a composite effect of misexpressions derived from both imprinted domains , for example , Igf2 inactivity combined with Cdkn1c over-expression [43] . To define the role of imprinted genes regulated by the Igf2/H19 ICR in the MatDup . dist7 phenotype , we evaluated the effects of introducing a mutated Igf2/H19 ICR ( ICRΔ ) which cannot bind CTCF and form an insulator [44] . MatDup . dist7 fetuses carrying ICRΔ would be expected to be corrected in terms of the number of active alleles of Igf2—activation of one of two inactive alleles , H19—repression of one of two active alleles , and Ins2—activation of one of two inactive alleles in yolk sac . MatDup . dist7 fetuses carrying ICRΔ were significantly rescued in terms of growth and were able to survive to adulthood . These results demonstrate that the aberrant phenotype of MatDup . dist7 fetuses is highly dependent on the presence of two maternally-derived Igf2/H19 ICR chromatin insulators .
Maternal inheritance of ICRΔ results in activation of Igf2 in cis such that total Igf2 RNA is 1 . 7 and 2 . 1 times the normal level in the liver and kidney of 17½ dpc fetuses , respectively , and also repression of H19 in cis , such that total H19 RNA is 0 . 2 and 0 times the normal level in these same tissues , respectively [44] . This configuration of expression—two active Igf2 and two inactive H19 alleles—is coincident with increased growth , an effect thought to be due to the former misexpression [18] , [45] , [46] . Lack of H19 RNA alone has no effect on Igf2 expression or imprinting and results in no discernible phenotype [47] . Maternal inheritance of ICRΔ would also be expected to result in activation of Ins2 in yolk sac . To confirm that maternal inheritance of ICRΔ can mediate normal growth , we tested its function in mice paternally inheriting a null mutation of Igf2 ( Igf2− ) . Mice of genotype ( ICR+/+ , Igf2+/− ) are small due to lack of Igf2 activity , with the maternal allele inactive , and the paternal allele null [24] . Results are shown in Figure 1 . Experimental young of genotype ( ICRΔ/+ , Igf2+/− ) , in which the maternally-derived Igf2 allele is activated in cis by ICRΔ , were not significantly different in weight to control ( ICR+/+ , Igf2+/+ ) mice at 6 weeks of age ( females , P = 0 . 271; males , P = 0 . 035 ) . Thus , a single maternal copy of ICRΔ induces sufficient Igf2 activity for achieving normal postnatal growth . We note that , in respect to growth with one versus two active Igf2 alleles , experimental ( ICRΔ/+ , Igf2+/− ) animals with one active allele ( maternal ) , were not significantly different in weight to ( ICRΔ/+ , Igf2+/+ ) animals with two active alleles ( females , P = 0 . 378; males , P = 0 . 089 ) . Further , ( ICR+/+ , Igf2+/+ ) females with one active allele ( paternal ) , were not significantly different in weight to ( ICRΔ/+ , Igf2+/+ ) females with two active alleles ( P = 0 . 04 ) . However , in males , mice with one active allele ( paternal ) were lighter than mice with two active alleles , as expected ( P = 0 . 002 ) . Given the borderline probability values obtained , greater numbers of animals need to be analysed to accurately determine the relative growth rates of mice of the various genotypes . MatDup . dist7 zygotes were produced in intercrosses of mice carrying the reciprocal translocation T ( 7;15 ) 9H ( T9H ) . Such intercrosses give rise to a high proportion of unbalanced zygotes , and litter size is small . Of balanced zygotes , only one in seven are expected to be MatDup . dist7 , these identified by the dist7 marker , albino ( c ) , a mutation of the ‘tyrosinase’ ( Tyr ) gene [15] . The ICRΔ mutation was introduced into female T9H/+ parents and was inherited by MatDup . dist7 zygotes ( Figure 2 ) . Expected allelic activity of Igf2 and Cdkn1c in the three possible MatDup . dist7 genotypes is shown ( Figure 2B ) . ICRΔ-induced activation of Igf2 was confirmed in 13½ dpc MatDup . dist7 fetuses obtained in ( T9H/+ , Tyrc/c , ICRΔ/+ ♀×T9H/+ , Tyr+/+ , ICR+/+ ♂ ) intercrosses . The level of Igf2 transcript in MatDup . dist7 ICRΔ . + fetuses was the same as in control ICR+/+ fetuses with one active allele , while it was almost double the normal amount in MatDup . dist7 ICRΔ . Δ fetuses with probably two active alleles ( Figure 3A and 3B ) . Increased total Igf2 RNA was also seen in mice which maternally inherit ICRΔ and have an active maternal and paternal allele of Igf2 ( Figure 3A and 3B ) . Also , MatDup . dist7 fetuses of all genotypes contained at least double the amount of Cdkn1c RNA relative to controls , probably because of two active alleles ( Figure 3A and 3B ) . These intercross matings were allowed to proceed to term and we immediately began to observe viable albino or MatDup . dist7 young which were of overtly similar size to agouti littermates . A MatDup . dist7 animal and its two littermates at 10 days post-partum is shown ( Figure 4 ) . All MatDup . dist7 young obtained were of genotype ICRΔ . + or recombinant ICRΔ . Δ . Seven of 52 mice born were MatDup . dist7 which is similar to the expected frequency , indicating that ICRΔ was always able to increase growth and rescue viability . In age- and litter-matched animals , a significant weight deficit of approximately 17% in MatDup . dist7 animals became apparent at 6 weeks of age when compared with controls carrying an equivalent number of active Igf2 alleles , that is , MatDup . dist7 ICRΔ . + with control ICR+/+ ( one active allele each ) and MatDup . dist7 ICRΔ . Δ recombinant with control ICRΔ/+ ( probably two active alleles each ) ( Figure 5A ) . CDKN1C may antagonize the growth promoting effects of IGF2 [17] , [48] , and it has been suggested that excess CDKN1C may combine with lack of IGF2 to cause MatDup . dist7 death [43] . To test this possibility , we introduced a null allele of Cdkn1c ( Cdkn1c− ) into MatDup . dist7 fetuses to enforce its monoallelic expression . In ( T9H/+ , Tyrc/c , ICR+/+ Cdkn1c+/− ♀×T9H/+ , Tyr+/+ , ICR+/+ Cdkn1c+/+ ♂ ) matings , all of 55 young obtained were agouti controls , that is , at least six albino MatDup . dist7 ( ICR+ . + , Cdkn1c+ . − ) pups were expected , but none were observed . This result is not consistent with the idea that MatDup . dist7 death results only from the combined action of the Cdkn1c and Igf2 misexpressions . To test for a role of Cdkn1c over-expresssion in the growth deficit at 6 weeks of age of rescued postnatal MatDup . dist7 ICRΔ . + animals , we introduced Cdkn1c− into MatDup . dist7 fetuses such that they were of genotype ( ICRΔ . + , Cdkn1c+ . − ) . This genotype should be normalized for the number of active alleles of imprinted genes regulated by the Igf2/H19 ICR , and also be normalized for Cdkn1c expression , that is , all of these imprinted genes should be monoallelically expressed . In ( T9H/+ , Tyrc/c , ICRΔ/+ , Cdkn1c+/− ♀×T9H/+ , Tyr+/+ , ICR+/+ , Cdkn1c+/+ ♂ ) matings , viable MatDup . dist7 ICRΔ . + , Cdkn1c+ . − young were obtained and these did not display a significant weight deficit at 6 weeks of age—with the caveat that the weight measurements are relative to control young obtained in the previous matings ( Figure 5B ) . Their weights could not be compared to littermates as , given the mating scheme , agouti littermates were always positive for ICRΔ—inheritance of Cdkn1c− being lethal—and therefore possessed two active copies of Igf2 . In any event , these results are consistent with the possibility that biallelic expression of Cdkn1c does contribute to a reduction in postnatal growth in MatDup . dist7 ICRΔ . + or ICRΔ . Δ , Cdkn1c+ . + animals .
We have shown that maternal introduction of a mutant Igf2/H19 ICR , which lacks chromatin insulator activity , into MatDup . dist7 fetuses substantially alters their abnormal phenotype—small size and death at the late fetal stage—to one of near normal growth rate and survival to adulthood . This result clearly demonstrates the dependence of this phenotype on a misexpression of imprinted genes caused by the presence of two active maternally-derived Igf2/H19 ICR chromatin insulators . As this ICR is known to regulate the expression of at least three dist7 imprinted genes—H19 , Ins2 , Igf2 , and a number of non-coding transcripts—correction in the misexpression of one or more of these was probably responsible for the result obtained . Activation of Igf2 was likely an important correction , this being the only alteration in expression induced by ICRΔ expected to affect growth . The survival of MatDup . dist7 mice with ICRΔ is more difficult to decipher . As discussed in the Introduction section , it is unlikely that the Igf2/H19 ICR-derived misexpressions are solely responsible for their death , as mice with two functional chromatin insulators—a maternally-derived Igf2/H19 ICR , and a paternally-derived chicken insulator substituted for the Igf2/H19 ICR , possess the same combination of misexpressions as MatDup . dist7 mice in respect to this region , yet these animals have normal postnatal viability [41] . Further evidence is provided by observations of the effects of misexpression of each imprinted gene alone . First , for H19 , no overt effect on phenotype is observed in transgenic mice with ectopic over-expression [49]–[52] . Biallelic or over-expression of H19 has been suggested to cause perinatal death of mice produced by combining a non-growing oocyte genome ( ng ) , carrying a deletion of the distal Chr 12 IG-DMR ICR ( Δ12 ) , with a fully grown oocyte genome ( fg ) —ngΔ12/fg mice [53] . However , these mice would be predicted to have the equivalent expression profile of imprinted genes as mice with maternal inheritance of the chicken insulator substitution . The latter mice are viable , despite twofold over-expression of H19 [41] . Therefore , the perinatal death of ngΔ12/fg mice may result from the combined action of H19 RNA excess—or possibly Igf2 RNA absence—and small imperfections in expression derived from the non-growing oocyte genome , for example , as related to the IG-DMR ICR deletion . Second , for Ins2 , mice lacking in expression of this gene are viable [54] . Third , for Igf2 , mice lacking expression are dwarfed and have impaired lung development [55] , but are usually viable . High postnatal survival frequency of Igf2 mutants is seen in inbred strain 129/SvEv [23] , [24] although in this strain we have observed a low level of perinatal death ( J . Mann , unpublished observations ) . In the present study , in a mix of strains 129/SvEv and outbred Swiss CF-1 , we observed high frequency survival . Also , in this same strain mix , we maintained a Igf2−/− random-bred line for a number of years which had normal fecundity ( J . Mann , unpublished data ) . On the other hand , use of a second Igf2 null mutation [56] revealed that lack of IGF2 in strain C57BL/6J results in death at birth . This effect was not peculiar to this second knockout allele as homozygous mutants can be obtained in strain 129 ( M . Constancia , personal communication ) . In the present study , MatDup . dist7 young were a mix of strains 129S1/SvImJ , CF-1 , C57BL/6J and CBA/Ca . In this mix , lack of Igf2 activity is highly likely to be compatible with survival . Given these various lines of evidence , the present experiments strongly suggest that misexpressed imprinted genes , as regulated by the Igf2/H19 ICR , work in combination with misexpressions derived outside of this region of influence in causing the total MatDup . dist7 phenotype . The significant rescue in growth probably mediated by Igf2 activation may also be directly related to MatDup . dist7 survival in that it could compensate for negative effects derived from outside the Igf2/H19 ICR region . Nevertheless , we cannot rule out the possibility that Ins2 inactivity in yolk sac , excess H19 RNA , or the misexpression of non-coding RNAs regulated by the Igf2/H19 ICR make a contribution to the lethal effect . These possibilities could be investigated through correction of their misexpression in MatDup . dist7 fetuses , then determining growth and survival . For example , correction of H19 over-expression could be achieved by introducing a deletion of the transcript region only . The imprinted genes operating outside the influence of the Igf2/H19 ICR that contribute to MatDup . dist7 death would be expected to require maternal- , rather than paternal-specific imprinting or methylation for attaining differential expression in the normal context . This is because for full-term development , there is apparently no other requirement , aside from Igf2/H19 ICR methylation , for paternal imprinting at dist7 [57] . The cluster of genes requiring maternal-specific methylation of the KvDMR1 ICR for activity fulfills this criterion . While the introduction of a null mutation of Cdkn1c , and hence enforced monoallelic expression of this gene , did not rescue MatDup . dist7 fetuses , this does not rule out the possibility that CDKN1C excess has a role in causing MatDup . dist7 death . In MatDup . dist7 ( ICR+ . + , Cdkn1c+ . + ) fetuses , Cdkn1c RNA levels were found to be more than three times that of controls , suggesting that each maternally-derived Cdkn1c allele was upregulated 1 . 5-fold . Therefore , CDKN1C could still be in excess in MatDup . dist7 ( ICR+ . + , Cdkn1c+ . − ) animals . Also , there remains the possibility that excess Cdkn1c RNA may contribute as part of a network of misexpressions derived from the cluster regulated by the KvDMR1 ICR . For example , biallelic expression of the ‘pleckstrin homology-like domain , familiy A , member 2’ ( Phlda2 ) gene results in placental growth retardation and marginal fetal growth restriction [58] , and upregulation of PHLDA2 is correlated with growth retardation in humans [59] , [60] . Also , it has been suggested that excess expression of the ‘achaete-scute complex homolog 2 ( Drosophila ) ( Ascl2 ) gene could cause the MatDup . dist7 lethal effect [42] . The phenotype of MatDup . dist7 fetuses could also involve misexpressions of dist7 imprinted genes lying outside of the influence of the two known ICRs . For example , ‘adenosine monophosphate deaminase 3’ ( Ampd3 ) —maternally expressed in placenta , and identified in a transcriptome analysis of MatDup . dist7 conceptuses [22] , ‘inositol polyphosphate-5-phosphatase F’ ( Innp5f ) —an isoform paternally expressed in brain [61] , and ‘cathepsin D’ ( Ctsd ) —possible paternal-specific expression [62] . The postnatal weight deficit of approximately 17% in MatDup . dist7 young at 6 weeks of age was similar to that in mice paternally inheriting a deleted KvDMR1 ICR . This deletion results in biallelic expression of imprinted genes regulated by this ICR , including Cdkn1c [17] , [34] . Indeed , excess CDKN1C has been indicated as the cause of the weight deficit [35] . Consistent with this possibility is that the weight of MatDup . dist7 ICRΔ . + , Cdkn1c+ . − young was normal at 6 weeks of age . However , we note that MatDup . dist7 neonates displayed no significant weight deficit until reaching adulthood , while in mice paternally inheriting the deleted KvDMR1 ICR , the weight deficit is present in fetuses and persists throughout postnatal development [17] . More data regarding weight gain in relation to the inheritance of ICRΔ , in MatDup . dist7 young and otherwise , is required to confirm these observations . In terms of MatDup . dist7 death , additional experiments are required to determine exactly which combination of misexpressions are involved . The total MatDup . dist7 phenotype has been ascribed to the very distal portion of Chr 7 as defined by the reciprocal translocation T ( 7;11 ) 65H ( T65H ) [42] . This translocation has a breakpoint far more distal on Chr 7 relative to the T9H translocation used in this study , although is still proximal to the two clusters of imprinted genes regulated by the Igf2/H19 and KvDMR1 ICRs . However , some caution should be exercised in ascribing the total effect to this region . While it was shown that T65H- and T9H-MatDup . dist7 fetuses are of similar morphology [42] , the postnatal viability of the former was not investigated . If T65H-MatDup . dist7 fetuses are also inviable , then the composite lethal effect is likely to be contained within the two aforementioned clusters of imprinted genes . Evidence that the KvDMR1 cluster contributes to the effect could be obtained by determining the viability of MatDup . dist7 fetuses carrying a deletion of this whole cluster . This would result in enforced monoallelic expression of all genes under regulation of the KvDMR1 ICR , including Cdkn1c , and these mice and would be expected to be postnatally viable , although small because of Igf2 inactivity . Such a deletion , made through truncation of Chr 7 at a point distal to the Ins2 gene , has been described [63] . A complication with this possible experiment is the existence of imprinted genes at dist7 which are not regulated by either ICR . Another experiment could be to breed mice with paternal inheritance of the chicken β-globin insulator substituted for the Igf2/H19 ICR [41] combined with paternal inheritance of the KvDMR1 ICR deletion [17] . These would misexpress all imprinted genes under regulatory control of both ICRs . If these were the only misexpressions involved in the MatDup . dist7 phenotype , then the phenotype should be reproduced . MatDup . dist7 fetuses provide an epigenetic model of a subtype of human Silver-Russell syndrome ( SRS ) involving maternal duplication of the orthologous Chr region , 11p15 . 5 , which encompasses ICR1 and ICR2 . In these fetuses , we have shown that abrogation of ICR1 insulator function was able to restore Igf2 expression , concomitant with restoration of growth and survival . The most common subtype of SRS , that involving hypo-methylation of ICR1 , is perhaps better modelled in mice maternally inheriting the chicken insulator in place of ICR1 . These animals provide information on the effects of the presence of two functional insulators at the Igf2/H19 region as the only epigenetic lesion . In these fetuses , we previously showed that DNA methylation was abrogated while insulator function remained intact . This resulted in reduced Igf2 activity and growth retardation [41] . Both of these findings support the idea that reduced expression of IGF2 during fetal development is causal in the development of SRS . They also support the suggestion that the failure to detect low concentrations of serum IGF2 in SRS patients is related to downregulation of IGF2 by this stage [38] . Further genetic manipulation in these mouse models should provide additional implications for the human disease . Our experiments suggest that misexpression of imprinted genes caused by two maternal copies of the Igf2/H19 ICR constitute one component of a composite barrier to parthenogenetic development that was not previously predicted . The lethal effect in MatDup . dist7 fetuses may be specific to later stages of development , and may not normally occur in parthenogenones given their peri-implantation death . Nevertheless , high-level paternal- and maternal-specific expression of Igf2 and H19 , respectively , is present shortly after implantation , at least by 6½ dpc [64] . Therefore , it cannot be ruled out that these misexpressions , and others regulated by the Igf2/H19 ICR , play a role in what probably is a complex composite lethal effect involving a network of misexpressed imprinted genes . Indeed , the fact that parthenogenones fail earlier in development than embryos with maternal duplication of any single Chr region , indicates that misexpressions of imprinted genes from different regions are cumulative or synergistic in their deleterious effects . Further , at the molecular level , it has been shown that disregulation of the imprinted genes ‘pleiomorphic adenoma gene-like 1’ ( Plagl1 ) and H19 can affect the expression of other imprinted genes in an imprinted gene expression network [65] , [66] . Previous observations have shown that the normal activity of imprinted genes regulated by the Igf2/H19 ICR are one of a small number of developmentally critical expression profiles provided exclusively by imprinting through the male germ line , provided that most if not all other imprinted genes are not misexpressed [57] . The present results raise the possibility that full-term parthenogenetic development could be achieved by correcting the misexpressions of only a few imprinted genes in order to repair the total expression network . One necessary correction would be to activate the ‘paternally expressed 10’ ( Peg10 ) gene . Lack of expression of this gene results in death by 10½ dpc , and this misexpression alone would be expected to present a barrier to parthenogenesis . It would be expected to contribute to , or could be solely responsible for , the embryonic death of MatDup . prox6 mice , which occurs prior to 11½ dpc [20] .
Line no . ; genotype; strain; source , how produced , or reference: Line-1; 129S1/SvImJ ( 129S1 ) ; Tyr+/+; The Jackson Laboratory , stock no . 002448 . Line-2; outbred Swiss CF-1; Tyrc/c; Charles River Laboratories . Line-3; T9H/T9H , Tyr+/+; mix of C57BL/6J ( B6 ) and CBA/Ca ( CB ) ; The Jackson Laboratory , stock no . 001752 . Line-4; T9H/T9H , Tyrc/c; mix of B6 , CB and CF-1; made by mating line-2 with -3 , then intercrossing . Line-5; Tyrc/c , ICRΔ/Δ; mix of CF-1 and 129S1; made by mating ICRΔ/+ mice [44] with line-2 , then intercrossing . Line-6; Tyrc/c , Cdkn1c+/−; mix of 129S7/SvEvBrd ( 129S7 ) , B6 and CF-1; made by mating Cdkn1c+/− mice [37] with line-2 , then intercrossing . Line-7; T9H/T9H , Tyrc/c , Cdkn1c+/−; mix of strains B6 , CB , CF-1 , and 129S7; made by mating line-4 with -6 , then intercrossing . Line-8; Igf2+/−; 129/SvEv [23] . Production of experimental ( ICRΔ/+ , Igf2+/− ) mice ( Figure 1 ) : Female parents ( ICRΔ/+ , Igf2+/+ ) were bred in ( line-5 ♀×line-1 ♂ ) matings . Male parents ( ICR+/+ , Igf2+/− ) were of line-8 . Young were a mix of strains 129 and CF-1 . Production of MatDup . dist7 ICRΔ . + and ICRΔ . Δ mice ( Figure 2 ) : Female parents ( T9H/+ , Tyrc/c , ICRΔ/+ ) were bred in ( line-5 ♀×line-4 ♂ ) matings . Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings . Young were a mix of strains 129S1 , CF-1 , B6 , and CB . Production of MatDup . dist7 Cdkn1c− . + young , attempted: Female parents ( T9H/+ , Tyrc/c , Cdkn1c+/− ) were bred in ( line-4 ♀×line-6 ♂ ) matings . Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings . Young were a mix of strains 129 , B6 , CB , and CF-1 . Production of MatDup . dist7 ( ICRΔ . + , Cdkn1c+ . − ) young ( Figure 5B ) : Female parents ( T9H/+ , Tyrc/c , ICRΔ/+ , Cdkn1c+/− ) were bred in ( line-5 ♀×line-7 ♂ ) matings . Male parents ( T9H/+ , Tyr+/+ , ICR+/+ ) were bred in ( line-3 ♀×line-1 ♂ ) matings . Young were a mix of strains 129S1 , 129S7 , B6 , CB , and CF-1 . For the ICR , two pairs of primers were used . The first pair was specific for the mutant ICR , identical to a pair previously described [41]: 5′- GCCC ACCA GCTG CTAG CCATC -3′ and 5′- CCTA GAGA ATTC GAGG GACC TAAT AAC -3′ , 240 bp amplicon identified ICRΔ . + and ICRΔ . Δ animals . The second pair was specific for ICR+ , with primers binding to sequence positions that were modified in ICRΔ [44]: 5′- AACA AGGG AACG GATG CTAC CG -3′ and 5′- GCAA TATG TAGT ATTG TACT GCCA CCAC -3′ , lack of a 506 bp amplicon identified ICRΔ . Δ animals . For Cdkn1c , the null allele was identified using primers specific for the selection cassette using in gene targeting: 5′- CTCA GAGG CTGG GAAG GGGT GGGT C -3′ , within the mouse ‘phosphoglycerate kinase 1’ ( Pgk1 ) promoter , and 5′- ATAC TTTC TCGG CAGG AGCA AGGT G -3′ , within the neo coding sequence , 520 bp amplicon . Fetuses were genotyped , and total RNA recovered using RNAzol ( Tel-Test ) after homogenization of the total fetus minus the head . Probes for Igf2 and ‘glyceraldehyde-3-phosphate dehydrogenase’ ( Gapdh ) RNA were as previously described [67] . The Cdkn1c probe was made by RT-PCR using primers; 5′- GCCG GGTG ATGA GCTG GGAA -3′ and 5′- AGAG AGGC TGGT CCTT CAGC -3′ , 221 bp amplicon . Northern blots were performed with 32P radiolabelled probes as described previously [68] . The three probes were hybridized independently to the same blots after stripping . Radioactivity of bands was quantitated using a Typhoon PhosphorImager ( Molecular Dynamics ) . For each lane , values for Igf2 and Cdkn1c were normalized to the Gapdh value . | Parthenogenetic mouse embryos with two maternal genomes die early in development due to the misexpression of imprinted genes . To gain further insight into which misexpressions might be involved , we examined some of the misexpressions that could determine the small size and fetal death of a “partial parthenogenone”—embryos with maternal duplication of distal Chr 7 ( MatDup . dist7 ) . We investigated the involvement of two maternal copies of the Igf2/H19 imprinting control region ( ICR ) , which is associated with lack of activity of the Igf2 gene , encoding a growth factor , and over-activity of H19 . By introducing a mutant ICR , we activated Igf2 and expected to correct other misexpressions , such as that of H19 . The result was substantial increase in growth and full postnatal viability of MatDup . dist7 fetuses , demonstrating the dependency of their abnormal phenotype on two maternal copies of the ICR . Activation of Igf2 was probably the main effector of this rescue . These results are consistent with the idea that reduced expression of IGF2 is causal in the human growth deficit disorder , Silver-Russell syndrome . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"developmental",
"biology/developmental",
"evolution",
"genetics",
"and",
"genomics/epigenetics",
"developmental",
"biology/developmental",
"molecular",
"mechanisms",
"evolutionary",
"biology/developmental",
"evolution"
] | 2010 | Postnatal Survival of Mice with Maternal Duplication of Distal Chromosome 7 Induced by a Igf2/H19 Imprinting Control Region Lacking Insulator Function |
Lethal recessive alleles cause pre- or postnatal death in homozygous affected individuals , reducing fertility . Especially in small size domestic and wild populations , those alleles might be exposed by inbreeding , caused by matings between related parents that inherited the same recessive lethal allele from a common ancestor . In this study we report five relatively common ( up to 13 . 4% carrier frequency ) recessive lethal haplotypes in two commercial pig populations . The lethal haplotypes have a large effect on carrier-by-carrier matings , decreasing litter sizes by 15 . 1 to 21 . 6% . The causal mutations are of different type including two splice-site variants ( affecting POLR1B and TADA2A genes ) , one frameshift ( URB1 ) , and one missense ( PNKP ) variant , resulting in a complete loss-of-function of these essential genes . The recessive lethal alleles affect up to 2 . 9% of the litters within a single population and are responsible for the death of 0 . 52% of the total population of embryos . Moreover , we provide compelling evidence that the identified embryonic lethal alleles contribute to the observed heterosis effect for fertility ( i . e . larger litters in crossbred offspring ) . Together , this work marks specific recessive lethal variation describing its functional consequences at the molecular , phenotypic , and population level , providing a unique model to better understand fertility and heterosis in livestock .
Lethal recessive alleles cause pre- or postnatal death in homozygous affected individuals , reducing fertility in various populations [1] . Although recessive lethals are generally widespread throughout populations , their effect is generally masked by the extremely low frequency of individual mutations . However , within small sized domestic and wild populations , those alleles might be exposed by inbreeding [2 , 3] , caused by matings between related parents that inherited the same recessive lethal allele from a common ancestor . The precise impact of recessive lethals depends on the population structure ( i . e . effective population size ) and recessive lethal mutation rates . In livestock , populations have been subject to intensive ( genomic ) selection resulting in relative small effective population sizes [4] . With small effective population size , genetic drift can rapidly increase the frequency of recessive lethals in the population . Although genomic selection has enabled substantial improvement on various traits including production , fertility , and disease resistance [5] , it does not provide much advantage over traditional selection when it comes to controlling the frequency of recessive lethal mutations [6] . Several studies have reported recessive lethal variation ( i . e . death of embryo or foetus prior to birth ) , likely derived from a single sire origin , to be maintained in livestock populations [1 , 7] . In fact , the frequency of some recessive lethals were driven by heterozygote advantage for important production traits , e . g . milk yield in cattle [8] , or growth in pigs [9] , although the majority was likely the result of genetic drift . Together these studies show that lethal recessive alleles can have a considerable impact on population fitness , emphasizing the need for early detection . Although various recessive embryonic lethal loci have been reported in livestock , pinpointing the causal mutation can be extremely difficult . Charlier et al ( 2016 ) showed , using a reverse genetic screen , that loss-of-function mutations and deleterious missense mutations cause embryonic lethality in cattle populations . Nevertheless , the discovery of recessive embryonic lethals is often hampered by the lack of affected individuals and the relative low frequency . Genotyping and sequencing large cohorts of animals within single populations can therefore facilitate the discovery of such detrimental variation , and point directly to the causal mutations . Pig fertility has increased steadily over the past years [10] . Breeding for improved fertility concerns a large number of traits with a combined effect on overall fertility , and lethal recessives are increasingly considered to substantially affect fertility in purebred livestock populations [11] . However , in pigs , the final production animals are crossbreds between purebred populations , usually derived from three-way crosses [12 , 13] . First , crossbred sows are created from two elite purebred populations selected for high production of piglets ( i . e . ‘maternal lines’ ) , which then are crossed with a third elite purebred population especially selected for meat production traits ( i . e . ‘paternal line’ ) . These crossbreds are known to perform better on multiple traits compared to their parental purebred lines , in particular for traits related to fertility and robustness [14] , as a result of the heterosis effect . Heterosis is caused by different non-additive effects , such as dominance , and it has been subject to a scientific controversy; the dominance hypothesis emphasizes the suppression of undesirable recessive alleles ( by dominant alleles ) , while the overdominance hypothesis emphasizes on heterozygote advantage [15] . However , the magnitude of recessive lethals contributing to heterosis is largely unknown . In this study we aim to explore the impact of lethal recessive variation in two pig populations using the following stepwise approach ( 1 ) perform simulations to assess the impact of genetic drift on lethal recessives , ( 2 ) identify haplotypes harboring lethal alleles using large-scale genotype data as developed by VanRaden et al . [16] , ( 3 ) confirm lethality by reduced fertility in carrier animals , ( 4 ) identify causal mutations segregating on these haplotypes using whole genome sequence data ( WGS ) and RNA-sequencing data , ( 5 ) study the impact of recessive lethals on heterosis for fertility related traits .
To identify lethal alleles segregating in the pig populations we examined genotype data from 28 , 085 ( Landrace ) , and 11 , 255 ( Duroc ) animals . All animals were genotyped or imputed to a medium-density 50K SNPchip ( S1 Table ) . The genotypes were phased to build haplotypes , and then we applied an overlapping sliding window approach to identify haplotypes that show a deficit in homozygosity , likely harbouring a lethal recessive allele [16] . The analysis yielded one strong candidate haplotype ( DU1 ) harbouring a lethal recessive allele in the Duroc population , and four candidates in the Landrace population ( LA1-4 ) , respectively ( Table 1 ) . Haplotype lengths range from 0 . 5 to 5 Mb and carrier frequencies range from 4 . 6 to 13 . 4% . We observe no homozygotes for DU1 , LA1 , and LA3 haplotypes , while we expected 26 , 126 , and 16 , respectively . We do observe two , and three homozygotes for LA2 ( 50 expected ) and LA4 ( 14 expected ) , suggesting incomplete linkage disequilibrium ( LD ) of the haplotypes with the causal lethal recessive mutation . Four out of five haplotypes show deviation from Hardy-Weinberg equilibrium with over 50% carrier offspring for carrier-by-carrier matings . This is in concordance with the absence of homozygous offspring , resulting in a 1:2 offspring ratio instead of the expected 1:2:1 genotype offspring ratio ( Fig 1A , Table 1 ) . We analysed the effect of the haplotypes on fertility phenotypes including total number born ( TNB ) , number born alive ( NBA ) , number of stillborn ( NSB ) , and number of mummified piglets ( MUM ) . We examined a total of 504 carrier-by-carrier ( CxC ) and 5 , 992 carrier-by-noncarrier ( CxNC ) matings ( Table 2 ) . Interestingly , all five haplotypes show significant reduction in both TNB ( Table 2 , Fig 1B ) and NBA for CxC matings ( S2 Table ) . The reduction in TNB ranges from 15 . 1 to 21 . 6% which is somewhat smaller than the expected 25% assuming early lethality with complete penetrance for homozygotes ( Table 2 ) . No significant increase in number of stillborn ( NSB ) or mummified piglets ( MUM ) was found , suggesting that homozygotes die very early in pregnancy ( S3 Table ) . Together the 504 CxC matings cause a loss of 1 , 261 piglets over the last 5 years ( comparing average litter size of CxC and CxNC matings ) , affecting 2 . 9% and 0 . 92% of all litters in the Landrace and Duroc population , respectively ( Table 2 ) . None of the five regions were previously reported to be associated with reduced TNB [21] . To find causal mutations , we analysed WGS ( Landrace: 167 , Duroc: 119 ) and RNA-seq ( Landrace: 34 , Duroc: 25 ) data available from the populations under study ( S4–S5 Tables ) . The data was mapped to the latest Sscrofa11 . 1 reference build and functionally annotated using the Variant Effect Predictor ( VEP ) [22] . Next , we focussed on variants likely causing embryonic lethality ( EL ) in homozygous state , examining the impact of individual variants on the proteins . First , we selected loss-of-function ( LoF ) variants ( frameshift , stop-gained , splice-site ) and predicted deleterious missense variants within each population [23] . The predicted LoF and deleterious mutations show clear patterns of purifying selection , as observed from generally lower allele frequencies ( S6 Fig ) , an enrichment of inframe indels ( S7 Fig ) , and an enrichment of LoF mutations in the N-and C-terminal end of the gene ( S8 Fig ) . The likelihood of carriers being present in even a small random sample of pigs is high due to the relatively high carrier frequency of the candidate haplotypes found in this study . The candidate haplotypes could therefore be identified in pigs of the same populations for which WGS data ( LA1: 21 , LA2: 17 , LA3: 7 , LA4: 9 , DU1: 9 ) or RNA-seq data ( LA1: 4 , LA2: 3 , DU1: 3 ) was available . For each of the five haplotypes we used criteria of physical distance and co-segregation ( see Methods for details ) to select candidate causal mutations . A single strong candidate mutation was identified for all haplotypes , except LA4 ( Table 3 ) . We assessed whether the splice mutations in TADA2A and POLR1B are subject to nonsense-mediated mRNA decay , a surveillance pathway eliminating transcripts that contain premature stop-codons [32] . We assessed the expression of the wild-type and mutant transcripts in carrier animals for both genes . The abundance of both the mutant TADA2A , and POLR1B transcripts are significantly lower ( 2 . 5- to 5 fold ) compared to the wild-type transcripts , supporting that the mutant transcripts are likely subject to nonsense mediated decay ( S12 Table ) . We genotyped the complete litters of three LA1 , and one LA2 CxC mating for the predicted causal mutations , and confirmed the carrier status of both parents for each litter ( S13–S14 Tables ) . The three LA1 litters produced 38 piglets , 14 were homozygous for the wild-type allele ( 36 . 8% ) , 24 were heterozygous carriers ( 63 . 2% ) , and no homozygous mutants were found ( P<0 . 005 , Table 4 ) . The LA2 litter produced 13 piglets , 3 homozygous wild-type , 10 heterozygous carriers of the deletion , and no homozygous del/del mutants ( P = 0 . 076 ) . These results are in line with the 1:2 genotype ratio expected for CxC litters , supporting the recessive lethality of the candidate causal mutations . The current frequency of the embryonic lethals raises the question how population frequencies of lethal alleles have developed over the past years . Interestingly , we observe that overall the recessive lethal alleles are maintained at relative stable frequency over the past seven years ( 2012–2018 , Fig 1C ) , despite ongoing selection on littersize in these populations . The impact of individual lethal recessive alleles largely depends on its frequency . For example , assuming random matings , we estimate that about 1 . 8% of the population litters are CxC matings for LA1 , while only 0 . 21% of the litters are CxC matings for the LA4 allele ( Table 2 ) . The four Landrace lethals combined affect 2 . 9% of the litters within the population , responsible for the death of 0 . 52% of the total population of embryos , which causes an average reduction of 0 . 073 TNB in the population ( Table 2 ) . Next , we investigated whether ELs could contribute to the heterosis effect for fertility ( TNB ) observed in the crossbreds . The current Landrace population is mostly crossed with a Large White ( LW ) population to generate a commercial F1 population . F1 litters in Landrace sows produce on average 0 . 20 piglet larger litters compared to purebred Landrace ( LR = 14 . 18 , LR/LW = 14 . 38 ) ( S17 Table ) . All three identified mutations ( LA1-LA3 ) are not segregating in the LW population , suggesting no homozygous affected individuals in the F1 population . Therefore , part of the TNB difference is likely caused by the four recessive lethals affecting the purebred litters ( given the average reduction of 0 . 073 TNB as a result of the four ELs ) . Nevertheless , other heterotic effects will contribute to the increased litter size as well .
In this study we report five embryonic lethal haplotypes that segregate with carrier frequencies in the range of 4 . 6–13 . 4% in two commercial pig populations . We show that the use of large-scale genotype data within single populations provides the power to find lethal alleles with low frequencies . The inclusion of over 28 thousand individuals from the Landrace population , for instance , allowed us to detect the LA4 haplotype that has an allele frequency of only 2 . 3% . For three of the five recessive lethal haplotypes no homozygous carrier individuals were found , suggesting complete LD with a causal , recessive lethal variant . However , none of the recessive embryonic lethal haplotypes resulted in the theoretically expected 25% reduction in piglets born ( range observed is: 15 . 1–21 . 6% ) . The most likely explanation is that the number of embryos frequently exceeds the uterine capacity of the sow . Hence , by reducing the number of embryos by 25% , fewer wildtype/wildtype and wildtype/mutant embryos are eliminated [33] . This compensatory effect could be especially relevant if homozygous affected zygotes fail to develop or embryos die very early on . Especially if they die prior to implantation in the uterus , other viable healthy zygotes can compensate ( i . e . take their place in the uterus ) of the lethal effect in homozygous zygotes . A compensatory effect is particularly likely for LA1 homozygous affected embryos ( POLR1B ) , since in homozygous POLR1B knock-out mice embryos terminate development before implantation in the uterus is established [27] . Moreover , we did not observe an increase in mummified or stillborn piglets for CxC matings , again suggesting early termination ( i . e . prior to day 35 in gestation ) of homozygous animals in utero . All four genes affected by embryonic lethal alleles are involved in cellular housekeeping functions including transcription ( POLR1B , TADA2A ) , translation ( URB1 ) , and DNA damage repair ( PNKP ) , supported by the relative high expression of these genes within different tissue types [34] . The RNA-seq data from carrier animals confirmed the functional impact of the DU1 splice-donor and LA1 splice-site mutations , both resulting in truncated proteins caused by the skipping of complete exons . Interestingly , we show that a single splice-donor mutation can simultaneously cause exon skipping and intron retention , something described previously in human studies [35] , but not previously observed in pigs . Moreover , the alternatively spliced mRNAs are likely subject to nonsense-mediated decay , because the level of the mutant mRNA is significantly lower compared to the wild-type mRNA . All mutations ( except URB1 ) are located within parts of the genes predicted to be intolerant to LoF mutations observed from a negative subRVIS score [36] . Interestingly , embryonic lethality has been described in targeted mice null-mutants for POLR1B and PNKP [27 , 30] , but not for URB1 and TADA2A . In this study , however , we demonstrate that both URB1 and TADA2A are essential for normal embryonic development in pigs , likely to be similar in human . We did not find any coding variants or structural variants that are in high LD with the LA4 haplotype ( S18 Table ) . However , other type of variants ( e . g . small insertion elements ) could also induce embryonic lethality or genetic disease [37] , something not well explored in this study . We show that the frequency of the lethal haplotypes over time , at least over the past seven years , is stable , suggesting that there is no strong selection against these recessive lethal variants . The population genetic analysis indicates that the observed frequencies of recessive lethal alleles found in this study , can be the result of genetic drift alone . Moreover , the study on de novo mutations shows that the lethal mutations in the LA1 , LA2 , and DU1 haplotypes ( allele frequency > 4% ) likely arose over 25 generations ago ( assuming no heterozygote advantage ) . Genetic drift as a driving force for the observed frequencies is further supported by the lack of clear evidence for heterozygote advantage ( except for LA1 ) . Nevertheless , we cannot exclude that these alleles have been subject to genetic-hitchhiking in the past , resulting in heterozygote advantage due to LD with a beneficial allele that became fixed . Evidence for heterozygote advantage has been found for other highly detrimental variants that occur in higher frequencies than those observed here in wild and domesticated populations [8 , 9] . This could also be the case for the most frequent recessive lethal in our study , LA1 , for which a highly positive effect on mothering ability for heterozygous carriers was found . In sow lines , mothering abilities are among the most important selection traits . The favorable phenotype of heterozygous carriers ( mothering ability ) offsets the occasional lower litter size , as long as the carrier frequency does not become too high . Nevertheless , our simulations show that the allele frequency of recessive lethals can rise up to 10% as a result of genetic drift alone . At this frequency , the negative effects on fitness ( i . e . smaller litters and lack of homozygotes ) will prevent further increase in allele frequency . Recessive lethals , by definition , deviate from the Hardy-Weinberg equilibrium ( HWE ) . We analyzed whether our liberal HWE marker threshold might hampered the detection of high frequency ELs , but no new high frequency haplotypes were revealed ( S21 Table ) . Nevertheless , not all embryonic lethal variation currently present in the populations under study was identified . In fact , even if LD between recessive lethal causal variants and SNP-chip based haplotypes would be perfect , the minimum allele frequency that could be detected is around 2% for the Landrace population , and around 4% for the Duroc population . In addition , lethal recessives residing on more common haplotypes cannot be detected because the SNP density is likely too low to distinguish between the haplotype with and the haplotype without the lethal recessive . We estimated that ELs likely account for 1% of deaths in these pig populations , but the four identified Landrace lethals account for the loss of 0 . 52% of all newborn pigs per generation , showing that the remainder 0 . 48% is caused by yet to be identified EL mutations . In pigs , the crossbred production animals show clear signs of heterosis , especially for fertility related traits [14] . We provide compelling evidence that embryonic lethals contribute to the heterosis effect seen in the Landrace crossbred litters . Assuming that recessive lethal variation is generally occurring in a single breeding line only , crossbred products will only be heterozygous for the lethal recessive mutations . We show that at least 2 . 9% of the litters within a single pure breeding line ( Landrace ) are offspring of matings between carriers of lethal recessives identified in this study , and that the four identified lethal variants are responsible for a significant part of the total heterosis effect ( as measured in surviving piglets ) . The heterosis effect is caused by the suppression of recessive lethal alleles by dominant wildtype alleles in the crossbreds [15] , providing evidence that the impact of lethal recessives on fertility and heterosis in these commercial pig populations is likely underestimated . Nevertheless , other detrimental , but not lethal alleles , uniquely segregating in purebred pig populations likely contribute to the heterosis effect even more , although this has never been properly quantified . Our study shows high resolution and efficiency of combining large-scale genotype ( SNP chip ) , phenotype , whole-genome sequence , and RNA-sequencing data to identify deleterious mutations that confer early embryonic lethality in pigs . We report five relatively common embryonic lethal alleles with carrier frequencies between 4 . 7–13 . 4% . Four of the variants destroy the structure of essential genes involved in cellular housekeeping processes including mRNA transcription , translation , and DNA repair . Simulation shows that observed allele frequencies can be mainly explained as consequence of drift only and there is no clear evidence for heterozygote advantage for favourable traits . The large amount of phenotype and genotype data collected in modern breeding programs in combination with increasing genomic data provides excellent possibilities to monitor old and new detrimental mutations segregating in purebred livestock populations . Although , we provide compelling evidence that the identified embryonic lethal alleles contribute to the observed heterosis effect for fertility , only a small proportion of the overall heterosis can be explained by the effect of the EL alleles detected . Other factors contributing to heterosis remain to be detected .
Samples collected for DNA extraction were only used for routine diagnostic purpose of the breeding programs , and not specifically for the purpose of this project . Therefore , approval of an ethics committee was not mandatory . Sample collection and data recording were conducted strictly according to the Dutch law on animal protection and welfare ( Gezondheids- en welzijnswet voor dieren ) . The dataset consists of 28 , 085 and 11 , 255 animals from Norwegian Landrace and Duroc purebreds , respectively . The animals are genotyped on the ( Illumina ) Geneseek custom 50K SNP chip with 50 , 689 SNPs ( 50K ) ( Lincoln , NE , USA ) . The chromosomal positions were determined based on the Sscrofa11 . 1 reference assembly . SNPs located on autosomal chromosomes were kept for further analysis . Next , the SNPs were filtered using following requirements: Each marker had a MAF greater than 0 . 01 , and a call rate greater than 0 . 85 , and an animal call rate > 0 . 7 . SNPs with a p-value below 1x10-5 for the Hardy-Weinberg equilibrium exact test were also discarded . All pre-processing steps were performed using Plink v1 . 90b3 . 30 [38] . After quality control , the final dataset contained 43 , 375 and 42 , 706 markers for Landrace and Duroc populations , respectively . We used BEAGLE version 4 . 1 genetic analysis software to phase both populations separately [39] . Haplotypes exhibiting missing or deficit homozygosity were identified using an overlapping sliding window approach from 0 . 5 to 5 MB . Within each window individual haplotypes ( with a frequency > 0 . 5% ) were evaluated for missing or deficit homozygosity . The expected number of homozygotes was estimated using two methods: ( 1 ) Estimation based on haplotype frequency , using the Hardy-Weinberg principle , ( 2 ) Estimation based on haplotype information from both parental haplotypes with the formula described by Fritz et al . , 2013 [20] . An exact binomial test was applied to test the number of observed homozygotes with the number of expected homozygotes . Haplotypes were considered significant if P < 5×10−3 . We examined the wild-type haplotypes for each LA1 carrier animal to identify recombinant individuals . We used PyVCF [40] to gather both haplotypes for all carriers animals within the LA1 genomic region from the BEAGLE phased VCF file . Next , we divided the LA1 haplotype in 5 shorter sub-haplotypes ( length = 1Mb ) . Next , we examined whether the sub-haplotypes were carried in homozygous state in the group of LA1 carrier animals . Homozygous sub-haplotypes were excluded to carry the causal mutation . We examined phenotypic records for TNB , NBA , NSB , and MUM to verify the lethality of the detected haplotypes . We listed all CxC and CxNC matings available and used a Welch t-test to assess if the phenotypes from the CxC matings significantly differ from CxNC matings . A P-value < 0 . 05 was considered significant . The order of CxNC matings does not reflect the sex of the parent animal and is both carrier boar and carrier sow combined Sequence data was available for 167 ( Landrace ) and 119 ( Duroc ) animals from paired-end 100 bp reads sequenced on Illumina HiSeq [41] . The sequenced samples are frequently used boars born between 2003 and 2017 , selected to capture as much of the genetic variation present in the Landrace and Duroc populations . The majority of the sequenced animals were also represented in the 50K genotype dataset ( Landrace = 161 , Duroc = 72 ) . The coverage ranges from 6 . 65 to 21 . 46 , with an average coverage of 12 . 70 ( S19 Table ) . Sickle software was used to trim the sequences [42] . BWA-MEM ( version 0 . 7 . 15 , [43] ) was used to map the WGS data to the Sscrofa11 . 1 reference genome . Samtools dedup was used to discard PCR duplicates [15] . GATK IndelRealigner was used to perform local realignments of reads around indels [44] . Freebayes variant calling software was used to call variants with following settings: min-base-quality 10—min-alternate-fraction 0 . 2—haplotype-length 0—ploidy 2—min-alternate-count 2 [45] . Post processing was performed using bcftools [46] . Variants with low phred quality score ( <20 ) , low call rate ( <0 . 7 ) and variants within 3 bp of an indel are discarded . Next , genotype calls are filtered for sample depth ( min: 4 , max: AvgDepth *2 . 5 ) leaving a total of 18 , 118 , 052 , and 15 , 857 , 077 post-filtering variants for Landrace and Duroc population , respectively . The average variant call rate is 95 . 4% ( Landrace ) and 96 . 4% ( Duroc ) , and the average transition / transversion ( TS/ TV ) ratio is 2 . 42 and 2 . 27 , respectively , in concordance with previous findings in pigs [47] . The Smoove pipeline ( https://github . com/brentp/smoove ) was used to call SVs . Smoove uses various software to call and filter SVs taking the alignment BAM files , and the Sscrofa11 . 1 reference genome as input . First , Lumpy software is used to call SVs [48] . Next , Svtyper is used to genotype SVs [49] . To further filter SV calls , Mosdepth is used to remove high coverage regions , and Duphold to annotate depth changes within and on the breakpoints of SVs . We performed variant ( SNPs , Indels , and SVs ) annotation using Variant Effect Predictor ( VEP , release 90 ) [22] . The variant effect prediction in protein altering variants was performed using SIFT [23] . The following variant classes were considered potentially causing LoF: splice acceptor , splice donor , inframe indels , frameshift , stop loss , stop gained , and start lost variants . LoF and deleterious missense variants were selected within each population that met following criteria . The variant is found in a maximum of 1 homozygous individual , allowing one false genotype assignment . Next , the variant is annotated in a gene that is a 1-to-1 ortholog with cattle to minimize the effect of off-site mapping of sequence reads , which can be particularly problematic for large gene families . Finally , the list of EL candidates was manually validated for possible sequencing and alignment artefacts . Further functional support was obtained from the MGI database release 6 . 10 ( i . e . phenotypes from null-mutant mice ) to predict the relative impact on the phenotype [26] . To identify candidate causal mutations for the haplotypes exhibiting missing homozygosity we applied the following criteria: 1 ) The mutation is located within 5 Mb of the haplotype boundaries . 2 ) The mutation is carried in heterozygote state by the haplotype carriers and no homozygous individuals are observed . 3 ) The mutation is absent from non-haplotype-carrier animals . 4 ) . The mutation is in high LD with the candidate lethal recessive haplotype ( R2 > 0 . 7 ) . LD analysis was performed using Plink v1 . 90b3 . 30 [38] with following settings:—chr-set 18 , —r2 , ld-window-r2 0 . 7 . The impact of the splice mutations on the expression of the gene was assessed using RNA-seq data . The animals sequenced are frequently used artificial insemination boars selected based on extreme phenotypes all present in the genotyping data [50] . The phenotypes are based on high and low sperm DNA fragmentation index , a measure of well packed double-stranded DNA vs single-stranded denatured DNA , which is an important indicator of boar fertility . We mapped the RNA-seq data to the Sscrofa11 . 1 reference genome using STAR [51] and called transcripts and FPKM expression levels using Cufflinks [52] . To test for nonsense mediated decay , we examined the transcript expression level of both the mutant and wild-type transcript identified by Cufflinks . The predicted effect on the mRNA was further evaluated by manually inspecting alignments using the JBrowse visualization software [53] . Variants were called on the RNA using Freebayes v1 . 1 . 0 [45] to examine if the genes are subject to genomic imprinting , heterozygous coding variants are listed in S20 Table . We tracked four recent CxC litters and sampled the complete litter including parent animals . The complete litter and parents were genotyped for the candidate causal variants using matrix-assisted laser desorption/ionization time-of-flight mass spectroscopy ( MALDI-TOF MS ) assays . The candidate mutations were fitted into the same assay and the assay was designed using MassARRAY Assay Design software ( Agena Biosciences , Hamburg , Germany ) . The genotyping was done using the IPLEX protocol according to manufacturer’s instructions . The difference in the expected and observed Mendelian genotype ratios was tested using a Chi-Square test . We analyzed the frequency of the haplotypes harboring embryonic lethals per half-year starting from 01-jan-2012 and assessed the frequency on the total population ( live animals ) on each time point . We then examined the proportion of carrier and non-carrier animals to obtain the carrier frequencies for each time point . The percentage of affected litters was estimated by taking the product of the carrier frequency , and we examined the piglet loss using the phenotypic records available within the breeding program in the last seven years ( 2012–2018 ) . To test whether the EL alleles contribute to the heterosis effect for fertility in the crossbred litters in purebred Landrace sows , we made following assumptions: First , we expected no EL litters in the crossbreds ( heterozygotes ) . Second , we assumed 2 . 9% EL litters in purebred Landrace from the four identified lethal alleles . Third , we calculated the percentage of population deaths for each of the recessive lethals individually by taking the product of affected litters and the litter reduction . Combined , the four lethals account for 0 . 52% of population deaths , and the overall piglet reduction was calculated as the product of the average TNB ( 14 . 17 ) and the population deaths caused by EL litters in the Landrace population . We simulated changes in allele frequency across multiple populations under the model of Wright [54] . Each simulation was performed with different start frequencies , corresponding to the frequencies of the identified haplotypes . We selected a population Ne of 150 , and population size of 2050 ( 50 boars , and 2000 sows ) . Each genotype has an associated fitness value , and we set the fitness to zero for homozygous lethal allele carriers , and fitness 1 ( no negative fitness effect ) to heterozygotes and non-carriers . We assume constant population size through time , and matings are simulated randomly at each generation . Changes in allele frequencies are calculated using the R package driftR ( https://github . com/cjbattey/driftR ) . The simulation calculates allele frequencies from a random draw of a binomial distribution with a probability of success equal to the post-selection expected frequency for each generation and each population . The results are plotted in R using the package ggplot2 [55] . The frequency of de novo mutations was estimated based on a population size of 2050 , accounting for a de novo mutation allele frequency equal to = 1/4100 = 0 . 024% . We used a human and cattle based per generation de novo mutation rate equal to 1 . 2e-08 per nucleotide per generation [56 , 57] . The product of the genome size ( in nucleotides ) and mutation rate is used to calculate the number of de novo mutations per individual ( 4915 . 82 Mb * 1 . 2e-08 = 59 ) . Considering a replacement rate of approximately 50% , we estimate that 60 , 475 de novo mutations will arise each generation ( 1 , 025*59 ) . We used the same model from Wright [54] to simulate changes in allele frequency across multiple populations for de novo mutations . To test whether carriers of lethal haplotypes show signs of heterozygote advantage on important traits in the breeding goal , we performed association analyses between all lethal haplotypes found in this study and a total of 25 traits ( S15–S16 Tables ) included in the breeding goal of the evaluated populations . Estimated breeding values ( EBV ) were used as a response variable for each trait under study . The EBV of each animal was obtained from the routine genetic evaluation by Topigs Norsvin using an animal model . Association analyses were performed using the software ASREML [58] applying the following model: EBVij=μ+Hi+aj+eij where EBVij is the observed EBV for the animal j , μ is the overall EBV mean of the population , Hi is the number of copies ( 0/1 ) of the lethal haplotype i , aj is the additive genetic effect estimated using a pedigree-based relationship matrix , and eij the residual error . A p-value below 1 × 10−5 was considered significant . | Lethal recessives are mutations that cause early lethality in homozygous state that usually occur at very low frequency in wild and domestic populations . In livestock , however , those mutations might become more prevalent as a result of inbreeding . In this study , we report five such recessive lethal haplotypes that cause embryonic lethality in homozygous state in pigs . The causal mutations are of different type but all destroy the structure of essential genes involved in cellular housekeeping processes , essential for embryonic development . The lethal recessives have substantial impact on the population fitness affecting up to 3% of the population litters , causing the death of 0 . 52% of the total population of embryos . Moreover , these 'natural knockouts' can increase understanding of gene function within the mammalian clade . Together , our study will allow monitoring , and facilitate the purging and partial elimination of recessive lethal mutations in frequently used pig breeds . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"population",
"genetics",
"vertebrates",
"animals",
"mammals",
"alleles",
"genetic",
"mapping",
"mutation",
"population",
"biology",
"frameshift",
"mutation",
"swine",
"genetic",
"loci",
"genetic",
"drift",
"haplotypes",
"heterosis",
"eukaryota",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"amniotes",
"evolutionary",
"processes",
"organisms"
] | 2019 | Loss of function mutations in essential genes cause embryonic lethality in pigs |
During prophase of the first meiotic division ( prophase I ) , chromatin dynamically reorganises to recombine and prepare for chromosome segregation . Histone modifying enzymes are major regulators of chromatin structure , but our knowledge of their roles in prophase I is still limited . Here we report on crucial roles of Kdm5/Lid , one of two histone demethylases in Drosophila that remove one of the trimethyl groups at Lys4 of Histone 3 ( H3K4me3 ) . In the absence of Kdm5/Lid , the synaptonemal complex was only partially formed and failed to be maintained along chromosome arms , while localisation of its components at centromeres was unaffected . Kdm5/Lid was also required for karyosome formation and homologous centromere pairing in prophase I . Although loss of Kdm5/Lid dramatically increased the level of H3K4me3 in oocytes , catalytically inactive Kdm5/Lid can rescue the above cytological defects . Therefore Kdm5/Lid controls chromatin architecture in meiotic prophase I oocytes independently of its demethylase activity .
Meiosis I differs from the second meiotic and mitotic divisions in that homologous chromosomes are segregated , not sister chromatids . Accurate chromosome segregation in meiosis I is preceded by dramatic changes in prophase chromatin organisation . Eventually , homologous chromosomes are mechanically held on the metaphase spindle via chromosome arm cohesion and the crossover-induced physical linkage of chiasmata [1] . This structural configuration is brought about by recombination , in the majority of organisms the driving force behind the alignment , or pairing of homologues , guiding their subsequent close association , or synapsis . Notably , in C . elegans and Drosophila , pairing and synapsis occur without double strand breaks ( DSBs ) and are thus recombination-independent [2 , 3] . A global feature of meiosis , however , is the instalment of a chromosome axis on each synapsed bivalent , which serves as a platform to enable the activities of the recombination complexes and , importantly , link the homologues along their entire length [4] . This connection is mediated by the synaptonemal complex ( SC ) , a highly conserved proteinaceous structure with lateral and transverse elements . The full joining of SC elements along the homologues results in synapsis . Although the exact role of the SC in recombination is not fully dissected and may vary among organisms , it is thought to have a promoting effect on the maturation of homologue-exchange crossover events [5] . Thus , it establishes the prerequisite for chiasmata and therefore a suitable chromosome configuration for homologous segregation upon meiotic anaphase I . The chromosome axis is comprised of structural proteins like the cohesin complex , composed of the subunits SMC1 and SMC3 and two non-SMC subunits [6 , 7] . Enrichment of cohesins leads to shortening of sister chromatids along their longitudinal axis . Overall , these dramatic chromatin changes along with SC loading are key features of meiotic chromosome architecture . For the formation of crossovers , a role for chromatin structure rearrangements–nucleosome accessibility and histone modifications–has been described [8] . Histone trimethylation at H3 ( H3K4me3 ) is located at DSB sites in hotspots in mammals and yeast , along with the H3K4 methyltransferase PRDM9 [9–12] . However , the exact mechanism on the chromosome axis for this higher order DNA structural feature is not known . Another mechanism to alter nuclear chromatin organisation , to enable pairing , is the coupling of homologous centromeric regions in early meiosis [13] . Drosophila has adopted an extreme version of coupling , where after the breaking of somatic pairing , homologous and non-homologous centromeres are clustered even before the germline cells enter meiosis [14–17] . This clustering is maintained throughout oogenesis , post SC disassembly , when all homologous chromosomes have compacted into a spherical structure in the oocyte nucleus , the karyosome [18] . In the context of karyosome formation , a link has been shown between the underlying chromatin configuration via histone modifications and the global reorganisation of chromatin-associated protein complexes like the SC or condensin [19] . However , further regulators of chromatin architecture implicated in these meiotic reorganisations remained to be identified . Here we show that the histone demethylase Kdm5/Lid controls chromatin architecture in prophase I oocytes , including the SC , pairing of homologous centromeres , and the karyosome . Interestingly , we found that although Kdm5/Lid is a major demethylase of H3K4me3 , its demethylase activity is dispensable for prophase I chromatin architecture in Drosophila oocytes .
In wild-type Drosophila prophase I oocytes , meiotic chromosomes cluster together to form a spherical structure , the karyosome , after completion of recombination ( [18]; Fig 1A and 1C ) . Through an RNAi screen of nuclear and chromatin-binding proteins [20] , we identified Kdm5/little imaginal discs ( lid ) encoding a histone demethylase as a gene important for normal karyosome morphology . Expression of short hairpin RNA ( shRNA ) targeting Kdm5/lid in female germlines throughout oogenesis resulted in sterility and abnormal morphology of the karyosome in most oocytes ( Fig 1C ) . To exclude the possibility of off-target effects , another shRNA targeting Kdm5/lid was used and showed a similar karyosome defect ( Fig 1C ) . Furthermore , a semi-lethal Kdm5/lid mutant ( lid10424/k06801; [21] ) showed a similar karyosome defect in oocytes , confirming that Kdm5/Lid is required for proper karyosome morphology ( Fig 1C ) . Kdm5/lid is known to genetically interact with another histone demethylase gene , lsd1 , by antagonising each other in position effect variegation [22] . To test for a genetic interaction in meiosis , we co-depleted this protein together with Kdm5/Lid . We did not see enhancement or rescue of the Kdm5/lid RNAi prophase I defect by co-RNAi of lsd1 ( S1 Fig ) . Kdm5/Lid and Kdm2 are two histone demethylases thought to remove a methyl group from H3K4me3 in Drosophila [23] . To determine the contribution of Kdm5/Lid activity in oocytes , we probed the level and distribution of H3K4me3 in the oocyte nucleus at each oogenesis stage by immunostaining using a specific antibody against this modification . The area of ovarioles in which an oocyte is determined ( called the germarium ) is subdivided into regions 1 , 2a , 2b and 3 , according to the morphology and developmental events ( Fig 1A; [24] ) . In region 1 , a cystoblast produced by a germline stem cell undergoes four rounds of pre-meiotic mitosis to generate a cyst containing 16 cells . Up to four cells in each cyst initiate meiosis in region 2a and , among them , two ( pro-oocytes ) maintain the meiotic state in region 2b , and one cell is finally selected as the oocyte in region 3 . Subsequently , oogenesis is divided into stages 1–14 ( stage 1 corresponds to region 3 ) . In control meiotic nuclei , H3K4me3 signals were observed on chromosomes except for a DAPI-intense region which corresponds to pericentromeric heterochromatin ( [25]; Fig 1B ) . The total signal intensity was gradually decreased as oogenesis progressed . In Kdm5/lid RNAi , intense H3K4me3 signals were observed on chromosome arms excluding pericentromeric heterochromatin , and the level of H3K4me3 was much higher than seen in the control RNAi throughout oogenesis ( Fig 1B ) . Therefore , Kdm5/Lid provides a major H3K4 demethylase activity during meiosis . H3K4me3 is associated with DNA double strand breaks ( DSBs ) induced by Spo11 to initiate the recombination process in mice and yeast [10 , 11] . Furthermore , it is known that persistent DSBs lead to a karyosome defect through activation of meiotic recombination checkpoint signalling in Drosophila oocytes [26] . Therefore , increased H3K4me3 level may result in persistent DSBs which activate the meiotic checkpoint signalling and in turn induces karyosome defects . To test this possibility , DSBs were detected by immunostaining using an antibody [27] against phosphorylated H2Av ( the H2AX homologue ) that is associated with DSBs ( Fig 1D ) . In both control and Kdm5/Lid-depleted meiotic nuclei , DSBs were observed in region 2a and 2b , and nearly disappeared by region 3 as previously reported ( [28]; Fig 1D ) . Therefore , it indicates that the karyosome defect observed in Kdm5/Lid-depleted oocytes is not due to a checkpoint-triggering delay in the repair of DSBs . Furthermore , the main karyosome defect in Kdm5/lid RNAi or mutant is disruption of its spherical morphology . In contrast , DSB-repair mutants typically show chromatin association with the nuclear envelope [27] . To test whether the karyosome defect in Kdm5/lid RNAi is mediated by the meiotic recombination checkpoint , the checkpoint signalling was suppressed by a mutation in the checkpoint kinase Chk2/Mnk . The karyosome defect caused by a mutation in a DNA repair gene ( Rad51/spnA ) was rescued by the mnk mutation as previously shown ( [29]; Fig 1E ) . In contrast , the karyosome defect in Kdm5/lid RNAi was not rescued by the mnk mutation ( Fig 1E ) . Furthermore , co-depletion of an Mnk effector protein , p53 [26 , 30] , did not rescue the karyosome defect in Kdm5/lid RNAi ( Fig 1E ) . These results demonstrated that the karyosome defect caused by Kdm5/lid RNAi is independent of the meiotic recombination checkpoint , in accordance with the timely DSB repair observed ( Fig 1D ) . To determine the chromosome organisation within the karyosome , the location of centromeres was assessed by immunostaining of the centromeric protein CenpA/Cid ( Fig 2A–2C ) . In control germaria , clustering of centromeres started around the pre-meiotic 8-cell stage ( Fig 2A–2C ) . By region 2a , centromeres were clustered to one or two closely located foci in nearly all meiotic nuclei ( Fig 2A–2C ) , which is consistent with previous observations [14–17] . In Kdm5/lid mutant or RNAi germaria , centromeres were clustered in region 1 with the normal timing , but showed a clustering defect in region 2a or later ( Fig 2A–2C; S2A–S2C Fig ) . This indicates that Kmd5/Lid is required for maintenance of centromere clustering during meiosis , but not for establishment during pre-meiotic stages . There are three levels of centromere association observed in control oocytes , which are cohesion of sister-centromeres , pairing of homologous centromeres , and clustering of non-homologous centromeres ( Fig 2D; [13] ) . To determine which level of association is disrupted in Kdm5/lid RNAi , fluorescence in situ hybridisation was carried out using the pericentromeric dodeca satellite , specific to chromosome 3 , as a probe . Control oocytes have only one focus in the nucleus , indicating that sister-centromeres were joined by cohesion and homologous centromeres were paired properly ( Fig 2E–2G ) . About half of the Kdm5/lid RNAi oocytes showed two separate foci , the remainder having one focus . Importantly no oocytes had more than two foci ( Fig 2E–2G ) . We further examined pairing of pericentromeric regions of chromosome 2 and X . The pericentromeric AACAC satellite specific to chromosome 2 showed a pairing defect similar to that seen using the chromosome 3 pericentromeric probe ( S2D Fig ) . In contrast , pericentromeric 359-bp repeats specific to the X chromosome did not show a pairing defect ( S2E Fig ) . This is in agreement with previous reports showing that X-chromosome pairing is mediated by a different mechanism from chromosome 2 and 3 [16 , 17] . Single RNAi of lsd1 encoding another demethylase did not display a centromere clustering defect or pairing defect of the pericentromeric satellites of chromosome X or 3 ( S1A , S1C and S1D Fig ) . These results showed that cohesion of sister-centromeres was maintained , but pairing of homologous pericentromeric regions of chromosomes 2 and 3 was disrupted by Kdm5/Lid depletion . The SC is an elaborate proteinaceous structure formed in early prophase I ( prior to karyosome formation; Figs 1A and 3A ) that stabilises pairing of homologous chromosomes and promotes crossover between them . Loss of the transverse protein C ( 3 ) G , which bridges two homologues in the SC is known to result in a loss of homologous centromere pairing [31 , 14] . In the premeiotic 8-cell stage in the germarium , SC components accumulate at centromeres [14 , 32 , 17] . Once the 16-cell stage enters meiotic zygotene in region 2a , the SC is visible in punctae in up to four nuclei , which form a full-length SC upon entering pachytene in region 2a . In region 2b , fully assembled filamentous SC is restricted to the two pro-oocytes and subsequently , only the oocyte retains a full SC in region 3 [33] . The SC persists on the karyosome in the oocyte until stage 6 , when the filaments appear mostly broken and in short threads [33 , 34] . To determine whether Kdm5/Lid is important for the integrity of the SC , we immunostained the transverse protein C ( 3 ) G in ovaries expressing the control or Kdm5/lid shRNA ( Fig 3; S3 Fig ) . In control , filamentous C ( 3 ) G accumulated on chromosome arms in multiple nuclei at the earliest meiotic region 2a ( S3A Fig ) . By region 2b , the filamentous C ( 3 ) G structure was formed fully along chromosome arms in one of two pro-oocytes ( Fig 3A–3C ) . This fully formed filamentous structure in the oocyte started fragmenting at stage 3 , and had fragmented or disassembled in the majority of oocytes by stage 6 , except in the centromeric regions where C ( 3 ) G remained associated ( Fig 3A , 3C and 3D; S3B Fig ) . In region 2a , filamentous C ( 3 ) G structures were observed in less than half of the Kdm5/lid RNAi nuclei accumulating C ( 3 ) G , but by region 2b , these filamentous structures were fragmented ( Fig 3B and 3C ) . By stage 4 , all C ( 3 ) G structures were disassembled and dissociated from chromosome arms , much earlier than in control oocytes ( Fig 3C and 3D ) . However , C ( 3 ) G remained associated with centromeric regions , as seen in control oocytes ( Fig 3D ) . These observations point to a role for Kdm5/Lid in stabilising or maintaining the SC along chromosome arms , although it is dispensable for association of C ( 3 ) G with centromeric regions . Defective C ( 3 ) G filaments in meiotic nuclei depleted of Kdm5/Lid may be caused by defects in the underlying chromosome cores , filamentous cohesin-containing structures visible under a microscope due to shortening and compaction of chromosome arms ( Fig 3A; [6] ) . To determine whether the chromosome core is disrupted by Kdm5/lid RNAi , the cohesin subunit SMC1/3 was visualised by immunostaining ( Fig 4; S4 Fig ) . In control meiotic nuclei , the filamentous structure of cohesin was mostly assembled by region 2a , and fully by region 2b ( Fig 4A and 4B ) . These filamentous structures started disassembling at stage 4 ( Fig 4B and 4C ) . In meiotic nuclei depleted of Kdm5/Lid , the filamentous structure of cohesin was assembled partially by region 2a in a similar way to control , but started being disassembled at region 2b and was mostly disassembled by stage 4 ( Fig 4 ) . However , we noticed that the cohesin signal remained associated with centromere regions ( Fig 4C ) , much like our observation of persisting C ( 3 ) G staining at centromeres ( Fig 3D ) . These results suggest chromosome cores are partially assembled , but are either unstable or precociously disassembled from chromosome arms in the absence of Kdm5/Lid . Li et al . ( 2010 ) previously proposed that the demethylase activity of Kdm5/Lid is not required for viability and fertility [23] , suggesting a demethylase-independent function of Kdm5/Lid . To test whether the demethylase activity is essential for the nuclear organisation of prophase I , the wild-type gene and the Kdm5/lid gene with the mutated catalytic site ( JmjC*; H637A , E639A; [35] ) under its own promoter were obtained from the authors . Transgenic flies were generated and tested for the ability to rescue the phenotype of the lid10424/k06801 mutation . As expected , the Kdm5/lid mutant carrying a catalytically inactive Kdm5/lid transgene ( JmjC* ) showed high levels of H3K4me3 in oocytes , similar to the Kdm5/lid mutant without a transgene , while the Kdm5/lid mutant carrying a wild-type Kdm5/lid transgene had low levels of H3K4me3 ( Fig 5A ) . This confirmed that the mutation indeed abolishes the demethylase activity . The Kdm5/lid mutant without a transgene showed a similar phenotype as Kdm5/lid RNAi in terms of karyosome defects ( Fig 5B ) , unclustering of centromeres ( Fig 5C ) and a failure to maintain the SC along the arms ( Fig 5D ) . Introduction of a wild-type Kdm5/lid transgene rescued all of these defects in the Kdm5/lid mutant ( Fig 5B–5D ) . Remarkably , the Kdm5/lid mutant carrying the catalytically inactive transgene ( JmjC* ) showed largely normal karyosome morphology ( Fig 5B ) , centromere clustering ( Fig 5C ) and SC morphology ( Fig 5D ) . High H3K4me3 levels and normal SC morphology were observed in the same germarium , demonstrating that catalytically inactive Kdm5/Lid rescued the synaptonemal complex defects of the Kdm5/lid mutant without changing the H3K4me3 levels ( Fig 5D ) . Taken together , we conclude that the demethlyase activity of Kdm5/Lid is dispensable for chromatin organisation in meiotic prophase I . To see the effects of Kdm5/Lid depletion in chromosome segregation at later stages , mature oocytes that mainly arrest in metaphase I were subjected to fluorescence in situ hybridisation using the pericentromeric dodeca satellite probe specific to chromosome 3 , and further stained for DNA and α-tubulin ( Fig 6A and 6B ) . Control oocytes had a cluster of meiotic chromosomes associated with a bipolar spindle . Two pericentromeric foci were observed at the edge of the chromosome cluster ( Fig 6A and 6B ) . This represents the configuration in which sister centromeres were tightly associated together , but homologous centromeres were pulled towards the opposite poles . Homologous chromosomes are linked through chiasmata which are the result of crossovers in prophase I . A significant proportion of Kdm5/lid RNAi oocytes had two separate clusters of meiotic chromosomes ( sometimes of unequal sizes ) located closer to the poles of the bipolar spindle . In some cases , both clusters contained one centromere 3 signal each , while in the other cases , one of the chromosome clusters contained both centromere 3 signals ( Fig 6A and 6B ) . Therefore , Kdm5/Lid is crucial for chromosome positioning and orientation in metaphase I oocytes . Such chromosome positioning near the spindle poles resembles achiasmatic chromosomes in mutants which do not undergo meiotic recombination [36] . To gain further insights into whether Kdm5/Lid depletion affects crossing-over , we examined the localisation of the newly discovered Vilya protein [37] . In wild type , Vilya has been shown to associate weakly with the SC , and to further accumulate strongly at most DSBs in region 2a and at foci which are suggested to be crossover sites in region 2b . It was challenging to confidently distinguish Vilya accumulation at crossover sites from Vilya localisation to fragmented SC commonly seen in Kdm5/lid RNAi pro-oocytes in region 2b . Therefore we counted the number of Vilya foci in region 2a and 2b meiotic nuclei and measured the signal intensity of each focus ( Fig 6C and 6D ) . In region 2a , the number and intensity of the foci was nearly identical between Kdm5/lid RNAi and control ( Fig 6C and 6D ) , indicating Kdm5/Lid does not affect DSB formation consistent with our previous observations of phospho-H2Av . In region 2b , the intensity of most foci was much weaker in Kdm5/lid RNAi pro-oocytes than control ( Fig 6C and 6D ) . These weak foci may represent crossover sites that accumulate Vilya at a low level , or reflect reduction of Vilya signals due to underlying SC defects . However , we favour the alternative interpretation in which most of these weak foci represent Vilya localisation to fragmented SC , and that fewer crossovers are present in Kdm5/lid RNAi pro-oocytes . This is in good agreement with our observation that Kdm5/lid RNAi oocytes show abnormal metaphase I chromosome positioning similar to achiasmatic chromosomes , and with a proposed role of the SC in promoting crossing-over between homologues in various organisms [5] .
In this article we show that the histone demethylase Kdm5/Lid controls maintenance of both the chromosome axis and the synaptonemal complex ( SC ) . Loss of Kdm5/Lid leads to an increase of H3K4me3 levels on chromosome arms , demonstrating major demethylase activity for Kdm5/Lid in the female germline . Euchromatic , but not centromeric cohesin and SC proteins are lost in the absence of Kdm5/Lid , with a pronounced unclustering of centromeres and disruption of karyosome morphology throughout oogenesis . Meiotic DNA double strand breaks were formed and repaired normally . Unexpectedly , the demethylase activity of Kdm5/Lid is not required for maintaining the SC , centromere clustering or the proper formation of the karyosome . Our data lead us to propose Kdm5/Lid to act as a regulator of early meiotic chromatin reorganisation independently of its demethylase activity . In the absence of Kdm5/Lid , fewer strong foci of crossover-associated Vilya are found and consequently alignment of bivalents in meiosis I is compromised . Consistent with its role in maintenance of the SC , our data suggest that Kdm5/Lid promotes crossing over and therefore ensures accurate chromosome segregation . Our crucial finding is the requirement of Kdm5/Lid for chromatin organisation of meiotic prophase I nuclei . Chromatin architecture dramatically changes during meiosis . The roles of structural components , such as cohesins and SC components , have been well studied . In contrast , how the chromatin architecture is regulated is not well understood , except for an involvement of master-regulator kinases such as Plk , NHK-1 or the chromosomal passenger complex [19 , 38–41] . Histone modifying enzymes are good candidates to connect chromatin structure to chromosomal regulator complexes . For example , there is clear evidence for H3K4me3-enrichment at meiotic recombination sites in several organisms , and in human and mice this is promoted by DNA-binding of PRDM9 , to specify “hotspots” of DSB-mediated recombination initiation [9–12 , 42] . The mutant phenotype of Kdm5/lid , which is phenocopied by RNAi , is distinct from other known meiotic mutants affecting prophase I complexes in Drosophila . We found that the SC and its chromosome core only partially formed and failed to be maintained along chromosome arms , compromising synapsis . Interestingly , judged from the accumulation of the transverse SC element C ( 3 ) G and the cohesin core subunits SMC1/3 , both complexes are established and maintained at centromeric regions . In mutants for the Drosophila-specific cohesin proteins ORD , SOLO and SUNN , despite similarities in disassembling the core and SC , centromeric SMC is not maintained [6 , 43–45] . Thus , our study , together with the previous data , support the idea that cohesin at centromeres and chromosome arms is controlled differently [6] . Strikingly , premeiotic clustering of centromeres is established in the absence of Kdm5/Lid , and only when entering meiosis , the nuclei displayed unclustering of centromeres , which is distinct from SC mutants which show unclustering of centromeres [16 , 17] . This raises the possibility that centromere clustering in premeiotic mitosis and meiosis may be regulated independently . Another interesting observation is that despite persistent accumulation of C ( 3 ) G and SMC1/3 at centromeres , homologous centromeres of chromosome 2 and 3 have become unpaired without Kdm5/Lid . As expected , the X chromosome , which appears to employ a separate mechanism of pairing [16 , 17] , is not affected . Sister-centromeres are still held together , which is a similar phenotype to c ( 3 ) G mutants but distinct from cohesion defects seen in other mutants [14 , 15 , 44–47] . It was thought that the persistent pairing of homologous centromeres until spindle formation is important for accurate chromosome segregation , and that this is mediated by centromere localisation of SC components . Our results suggest either that Kdm5/Lid function is important for making these centromeric components functional , or that maintaining the SC at chromosome arms is important for maintaining homologous centromere pairing . Furthermore , Kdm5/Lid is important for karyosome formation . The karyosome is a compact spherical body made of clustered chromosomes formed within the oocyte nucleus after recombination has completed [18] . A similar structure is also found in oocytes of other animals , including humans [48] . The karyosome defect is likely to be separable and independent from the SC defect , as mutants defective for SC formation do not show karyosome defects [14] , and the SC is disrupted before the karyosome is formed . This suggests that Kdm5/Lid is a more general regulator of the chromatin architecture in oocytes . It is known that a failure of DSB repair would result in abnormal karyosome morphology and a delay in SC disassembly through activating meiotic recombination checkpoint signalling [25 , 27] . However , the two following observations argue against the possibility that a delay in DSB repair causes the karyosome defects seen in Kdm5/lid RNAi . Firstly , formation and repair of DSBs take place in a timely fashion without Kdm5/Lid . Secondly , suppression of the meiotic recombination checkpoint did not rescue the karyosome defects in Kdm5/lid RNAi . What would be the consequence of these defects ? In a proportion of metaphase I oocytes lacking Kdm5/Lid , chromosomes are abnormally located closer to the spindle poles with or without separation of homologous chromosomes . Location of two chromosome masses , with occasionally unequal sizes , close to the poles is similar to observations in ord mutants , which fail to maintain chiasmata and to arrest at metaphase I [46] . Additionally , an error in bi-orientation may be caused by such an unpairing of homologous centromeres , as homologous centromeric pairing is proposed to promote disjunction of homologues at anaphase I ( reviewed in [13] ) . Indeed , pachytene nuclei lacking Kdm5/Lid have fewer strong foci of Vilya which has been suggested to mark crossover sites [37] . Although the interpretation is tricky , it is tempting to speculate that a failure in maintaining the SC may lead to a reduced number of crossovers . This in turn would result in metaphase chromosomes lacking chiasmata , and thus failing to maintain a physical link on the metaphase plate . We showed that Kdm5/Lid provides a major demethylase activity towards H3K4me3 in prophase I oocytes , as its loss dramatically increased levels of H3K4me3 , in accordance with previous observations of global increases after Kdm5/Lid depletion [49] . A potential contribution of the second demethylase , Kdm2 , remains to be determined . The H3K4me3 epigenetic marker is often associated with transcriptionally active genes [50] . However , we found that catalytically inactive Kdm5/Lid can perform its function in SC maintenance , centromere clustering/pairing and karyosome formation in oocytes . It was previously shown that Kdm5/Lid has transcriptional roles independent of its catalytic activity in mammals and Drosophila [35 , 23 , 51] , but specific functions had not been identified at the cellular level . Importantly , the rescue of Kdm5/Lid loss by catalytically inactive Kdm5/Lid is not mediated through a removal of the methyl group from H3K4me3 , for example by activation of the other demethylase Kdm2 or by residual Kdm5/Lid , as H3K4me3 was abnormally high in the presence of the demethylase-dead transgene , comparable to levels seen upon loss of Kdm5/Lid . If Kdm5/Lid's catalytic activity is not essential , how does Kdm5/Lid regulate chromatin architecture ? Previous studies have already hinted at such mechanisms . Kdm5/Lid has been shown to form a complex with other proteins including the histone deacetylase Rpd3 , and to negatively regulate activity of Rpd3 [52] . An increase in histone acetylation is observed in mouse oocytes reaching mature stages [53] , and the inhibition of Rpd3’s deacetylation of target genes by Kdm5/Lid could possibly contribute to chromatin accessibility and structure in meiosis . In line with this , in cultured cells , Kdm5/Lid overexpression increases acetylation levels of H3K9 , an Rpd3 target site [54 , 55] , whereas Kdm5/Lid-loss reduces H3K9Ac [55] . Thus , we assessed if co-depleting Rpd3 from ovaries restores the observed defects upon Kdm5/Lid depletion . Unfortunately , double RNAi of Rpd3 and Kdm5/Lid resulted in tiny ovaries which we were not able to analyse . It is further possible that Kdm5/Lid may be required for the activity or localisation of other subunits in this complex , which is essential for chromatin organisation in prophase I oocytes . In mammals the Kdm5 family is linked with multiple processes , including cellular senescence , cell differentiation and mitochondrial biogenesis [56–61] . The Kdm5 functions are often assumed to be mediated by transcriptional regulation , as H3K4me3 is thought to be a transcription-active epigenetic marker . However , in most cases , requirement of the catalytic activity has not been demonstrated or tested . Our finding emphasises the chromatin regulating functions of Kdm5 independent from its catalytic activity , over the conventional view of Kdm5 as an "eraser" of the epigenetic mark . Finally , the Kdm5 family of proteins function in diseases including tumorigenesis and mental retardation [62–71] . Our findings may provide a new insight into how malfunctions in this family of histone demethylases causes disease at the molecular and cellular levels .
Standard fly techniques were followed [72] . All fly stocks have been cultured at 25°C in standard cornmeal media . For RNAi in ovaries , P{Gal4::VP16-nos . UTR}MVD1 were crossed with the following RNAi TRiP lines ( Harvard Medical School ) : ctrl RNAi ( white; GL00094 ) , lid RNAi ( GLV21071 , GL00612 ) , lsd1 RNAi ( HMS00638 ) , p53 RNAi ( GL01220 ) , lid mutant lines with a P-element insertion in the 5’UTR ( y1 w67c23; P{lacW}lidk06801/CyO and cn1 P{PZ}lid10424/CyO; ry506 ) have been obtained from Bloomington Drosophila Stock Center . Meiotic recombination checkpoint suppression was obtained by a heterozygous mutation , mnkp6 ( DmChk2; [73] ) . A DSB repair mutation , spnA1 [74] , was used as control . Flies carrying both spnA1 and mnkp6 mutations as well as flies carrying mnkp6 , Gal4::VP16-nos . UTR and RNAi construct were obtained by standard successive genetic crosses . Vilya3XHA transgenic flies ( w; pUASp-CG2709-HA2/CyO ) have been kindly provided by R . S . Hawley [37] . Standard molecular techniques were used throughout [75] . For generating rat anti-Cid antibodies , the full-length Cid coding region was introduced into the Gateway ( Invitrogen ) entry vector pENTR , and then into destination vector pGEX4T1-Gateway ( Amersham ) for GST- fusion . GST-Cid was expressed in BL21/pLysS , and purified from the insoluble fraction after it was run on a SDS gel . For generating a rat anti-C ( 3 ) G antibody , full-length C ( 3 ) G has been fused with GST , and the protein has been expressed in BL21/pLysS and purified from a soluble fraction using glutathion beads . These purified proteins were used for generating antisera by Diagnostic Scotland . pCasper4-based plasmids carrying wild-type lid and catalytically inactive lid[JmjC*] driven by the native promoter [23] were kindly provided by J . Secombe , and used to generate transformants ( The Best Gene ) . The lid[JmjC*] mutant contains changes of His637 and Glu639 to Ala [35] . Flies containing both RNAi constructs for lid and lsd1 or lid and p53 were identified by genotyping using wing PCR . DNA preparations from fly wings have been done according to [76] . Fluorescent in situ hybridisation ( FISH ) of ovaries was carried out as previously described [77 , 78] . The 359-bp repeat was obtained by PCR amplification from Drosophila genomic DNA , using primers designed from the published sequence [79] . An oligonucleotide corresponding to the dodeca satellite labelled at the 3’ end has been used as a pericentromere 3 FISH probe [77] , while an oligonucleotide corresponding to the AACAC repeat has been used as a pericentromere 2 FISH probe . To quantify the level of Kdm5/lid mRNA , RT-qPCR was carried out as previously described [80] except 7–10 pairs of ovaries from adult females ( matured for 3–5 days at 25°C ) were used . Three biological replicates were carried out for Kdm5/lid RNAi and Kdm5/lid mutant with or without transgenes , except for two biological replicates of Kdm5/lid mutant flies carrying a lid[JmjC*] transgene ( S5 Fig ) . Actin5C was used as a control for normalization . Ovaries and mature oocytes were immunostained and analysed according to [81 , 82] , respectively . C ( 3 ) G and SMC1 immunostaining was carried out as previously described [33] . Primary antibodies were used as follows: mouse monoclonal anti-Lamin antibody ( 1/100; ADL67 . 10; Developmental Studies Hybridoma Bank ) , rabbit anti-γH2Av antibody ( 1/100; [27] ) , rat anti-C ( 3 ) G antibody ( 1/100; this study ) , rat anti-SMC1 antibody ( 1/100 , kindly provided by C . Sunkel; [83] ) , guinea pig anti-SMC1 and anti-SMC3 ( 1/2 , 000 and 1/1 , 000 respectively , kindly provided by S . E . Bickel; [6] ) , rat anti-Cid antibody ( 1/100; this study ) , rabbit anti-Cid antibody ( 1/800 , Active Motif ) , rabbit anti-H3K4me3 antibody ( 1/100; Active Motif ) , rat monoclonal anti-HA antibody ( 1/100; 3F10; Roche ) , mouse monoclonal anti-Hts antibody ( 1/100; 1B1; Developmental Studies Hybridoma Bank ) and mouse monoclonal anti-α-tubulin antibody ( 1/250; DM1A; Sigma ) . Secondary antibodies conjugated with Cy3 , Alexa488 or Cy647 ( Jackson Lab or Molecular Probes ) were used at a 1/250 dilution . DNA was counterstained with DAPI ( 0 . 4 μg/ml; Sigma ) . Immunostained samples were examined by an Axioimager ( Zeiss ) attached to a confocal laser scanning head , LSM5 Exciter ( Zeiss ) . Either one Z-plane or a maximum intensity projection of selected Z-planes is shown in the figures after contrast and brightness were adjusted uniformly across the field using ImageJ ( NIH ) . To quantify the intensity of H3K4me3 levels , ImageJ software was used . The area and the Z-planes containing the karyosome were selected , and the sum of H3K4me3 signal intensity in the karyosome area of all selected Z projections was measured . The background level in an area in the nucleoplasm was estimated by measuring the sum signal intensity of the Z planes . The total karyosome-specific H3K4me3 signal was estimated to be the sum signals in the karyosome area subtracted by the background signal of the equivalent size . To quantify the intensity of Vilya3XHA foci , the areas of Vilya3XHA foci and nucloplasmic background were selected , and maximum signal intensity of the background area was subtracted from the maximum signal intensity of the foci area . Fisher's exact test and t-test were used for categorical and parametric data , respectively . | Accurate transmission of chromosomes carrying genetic materials from generation to generation is essential for life . Cell divisions that generate gametes , such as eggs and sperm , are critical , as chromosomes inherited from both parents recombine and are accurately sorted into gametes . Errors in these cell divisions often result in infertility , miscarriages or birth defects such as Down syndrome in humans . During these divisions , chromosomes undergo dramatic reorganisation but the molecular mechanisms are not well understood . Chromosome organisation is known to be regulated by various epigenetic marks , which are chemical marks on chromatin crucial for regulating gene expression . We found that an enzyme ( Kdm5/Lid ) that erases a mark linked to active gene expression regulates multiple aspects of meiotic chromatin organisation in oocytes , including stability of the recombination machinery . Unexpectedly , this function does not require its enzymatic activity . Our findings provide novel insights into how chromosomes are reorganised during reproduction and prompt re-evaluation of the role of this eraser enzyme . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"homologous",
"chromosomes",
"reproductive",
"system",
"rna",
"interference",
"chromosome",
"structure",
"and",
"function",
"centromeres",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"germ",
"cells",
"oocytes",
"epigenetics",
"chromatin",
"prophase",
"genetic",
"interference",
"animal",
"cells",
"chromosome",
"biology",
"gene",
"expression",
"biochemistry",
"rna",
"ovaries",
"cell",
"biology",
"nucleic",
"acids",
"ova",
"anatomy",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"chromosomes"
] | 2016 | Kdm5/Lid Regulates Chromosome Architecture in Meiotic Prophase I Independently of Its Histone Demethylase Activity |
H2A . Z is a histone H2A variant conserved from yeast to humans , and is found at 63% of promoters in Saccharomyces cerevisiae . This pattern of localization suggests that H2A . Z is somehow important for gene expression or regulation . H2A . Z can be acetylated at up to four lysine residues on its amino-terminal tail , and acetylated-H2A . Z is enriched in chromatin containing promoters of active genes . We investigated whether H2A . Z's role in GAL1 gene regulation and gene expression depends on H2A . Z acetylation . Our findings suggested that H2A . Z functioned both in gene regulation and in gene expression and that only its role in gene regulation depended upon its acetylation . Our findings provided an alternate explanation for results that were previously interpreted as evidence that H2A . Z plays a role in GAL1 transcriptional memory . Additionally , our findings provided new insights into the phenotypes of htz1Δ mutants: in the absence of H2A . Z , the SWR1 complex , which deposits H2A . Z into chromatin , was deleterious to the cell , and many of the phenotypes of cells lacking H2A . Z were due to the SWR1 complex's activity rather than to the absence of H2A . Z per se . These results highlight the need to reevaluate all studies on the phenotypes of cells lacking H2A . Z .
In addition to their role in genome packaging , histones also play a role in the functional organization of eukaryotic genomes . Clear causal relationships have been established between some specific modifications of histones at specific loci and the subsequent events that occur at these loci . Histones are modified by enzymes that couple acetyl , methyl , phosphoryl , ubiquitin , or sumo moieties to specific locations either on histone tails , which extend outward from the nucleosome core , or at positions in the core , such as acetylation of H3 lysine 56 , near where the DNA helix enters and leaves the nucleosome [1] . Modified histone tails serve in some cases as docking sites for protein complexes . Thus , in principle , a particular collection of modifications on the nucleosomes of a locus can recruit specific complexes to that locus to achieve a particular outcome [2]–[7] . In addition to histone modifications , nucleosomes can also be specialized by virtue of the presence of histone variants . Saccharomyces encodes three histone variants: H2A . Z , which is conserved from yeast to humans; a variant of H2B called H2B2 , conserved among yeasts; and Cse4 , an H3 variant , which functions at the nucleosomes at centromeres [8] . Like Cse4p , H2A . Z is also localized to specific chromosomal locations with specialized functions . In S . cerevisiae , H2A . Z is incorporated into nucleosomes near , but not at , centromeres , at the borders of heterochromatic domains , and near the promoters of 63% of genes [9]–[12] . H2A . Z is incorporated into chromatin by the SWR1 complex ( SWR1-Com ) a multi-subunit enzyme whose catalytic subunit , Swr1 , is a member of the Swi2/Snf2 family of chromatin remodeling enzymes [13]–[15] . H2A . Z's localization at promoters suggests that it plays an important role in gene expression . Yet genome-wide micro-array analyses indicate that H2A . Z affects the steady-state mRNA levels of only 5% of S . cerevisiae's genes [16] . Interestingly , most of the genes downregulated in cells lacking H2A . Z were near the boundaries of SIR-silenced heterochromatin . This observation revealed that H2A . Z functions as part of the boundary separating euchromatin and heterochromatin [16] . H2A . Z is acetylated at up to four positions on its N-terminal tail by the NuA4 and SAGA histone-acetyltransferase complexes [17]–[19] . Moreover , H2A . Z's heterochromatin-boundary function depends on its acetylation [17] . Promoter-proximal H2A . Z is also acetylated and , as measured on a cell population , the level of acetylation correlates with the gene's expression level [19] . Recent work suggests that acetylated-H2A . Z promotes transcription of adjacent genes . Specifically , H2A . Z at the promoters of the oleate-responsive genes CTA1 , POX1 , POT1 , and FOX2 is acetylated on Lys14 . Cells with a mutant form of H2A . Z that cannot be acetylated at this position are defective in induction of these genes [20] . H2A . Z's contribution to gene-induction was first explored in the context of the GAL1 , GAL7 , and GAL10 genes [21] , [22] , which are induced in medium containing galactose , repressed in medium containing glucose , and expressed at a basal uninduced level by cells grown in medium with nonfermentable carbon sources [23]–[26] . Because the galactose regulon is one of only a handful of thoroughly studied regulated genes in yeast , it has provided many fresh insights into gene regulation . Hence results on , and claims about , this regulon take on special importance in the field . Induction of GAL1 , GAL7 , and GAL10 occurs more rapidly when S . cerevisiae cells are grown in a nonrepressing , noninducing carbon source ( such as raffinose ) and then shifted to inducing conditions ( galactose ) than when cells are grown in repressing conditions ( glucose ) and then transferred into inducing conditions [23]–[26] . The one exception to this pattern involves a phenomenon known as transcriptional memory . S . cerevisiae cells grown in inducing conditions prior to short-term growth in repressing conditions are able to reinduce GAL- gene expression upon induction as rapidly as cells grown continuously in nonrepressing conditions [27]–[29] . This “memory” of recent inducing conditions is reported to be H2A . Z dependent [27] , although other explanations have been offered [29] . The role of H2A . Z in galactose induction extends beyond its role in GAL1 transcriptional memory . Cells that are grown in nonrepressing conditions prior to galactose induction require H2A . Z for the rapid induction of GAL1 [21] , [22] . H2A . Z promotes the rapid induction of GAL1 by recruiting the Mediator complex to the GAL1 promoter [30] , [31] . The work presented in this paper was aimed at testing the potential role of H2A . Z acetylation in gene induction and transcriptional memory . We found no evidence for a role for H2A . Z in GAL1 transcriptional memory , discovered a role for H2A . Z acetylation in gene induction , and discovered a confounding influence of SWR1-Com on gene regulation in cells lacking H2A . Z .
Upon galactose induction , cells previously grown long-term in repressing conditions induce GAL1 expression more slowly than cells previously grown in noninducing-nonrepressing conditions . The conclusion that H2A . Z is essential for GAL1 transcriptional memory was based on the following two observations . First , when transferred to inducing conditions from long-term growth in repressing conditions HTZ1 and htz1Δ cells induce GAL1 slowly and at a similar rate [27] . Second , when transferred to inducing conditions from short-term growth in repressing conditions , HTZ1 cells induce GAL1 transcription rapidly , but htz1Δ cells are reported to not induce GAL1 any more rapidly than htz1Δ cells that had been grown long term in repressing conditions prior to galactose induction [27] . We reasoned that if H2A . Z acetylation were required exclusively for transcriptional memory , then cells carrying an unacetylatable allele of HTZ1 , htz1-K3 , 8 , 10 , 14R , would exhibit defective GAL1 induction following short-term growth in glucose , but exhibit normal GAL1 induction following long-term growth in glucose . To determine first whether H2A . Z-acetylation had any role in galactose expression , GAL1 mRNA levels were evaluated by quantitative reverse transcriptase ( Q-RT ) PCR in HTZ1 ( JRY7971 ) , htz1Δ ( JRY9001 ) , and htz1-K3 , 8 , 10 , 14R ( JRY7983 ) cultures grown in long-term repressing conditions prior to galactose induction . Cells grown continuously in glucose medium were transferred to galactose medium and GAL1 induction was evaluated at 2-h intervals for 14 h . One characteristic shared between all three strains' GAL1 induction phenotypes was an approximately 3-h lag period with little to no GAL1 expression . Quantitative analysis suggested that neither htz1Δ nor htz1-K3 , 8 , 10 , 14R cultures exhibited substantially different lag periods prior to the onset of GAL1 expression than those exhibited by HTZ1 cultures ( Figure 1A; Table 1 , column A ) . These results suggested that neither H2A . Z nor its acetylation influenced how rapidly the cultures exited glucose repression and began GAL1 transcription . Other than their lag periods the two mutant cultures exhibited significantly different GAL1 induction phenotypes than those of HTZ1 cultures . Cultures of the two mutant strains had lower steady-state GAL1 expression levels than HTZ1 cultures ( Figure 1A; Table 1 , column C ) . Quantitative analysis suggested that htz1Δ and htz1-K3 , 8 , 10 , 14R cultures required 54 . 7% and 60 . 2% more time , respectively , than HTZ1 cultures to reach half steady-state GAL1 expression levels ( Figure 1A; Table 2 , column E; note that half steady-state levels were used instead of half-maximum levels because the level of expression during induction typically overshot the induced steady-state level ) . These values , however , underplayed the severity of the htz1Δ and htz1-K3 , 8 , 10 , 14R cultures' GAL1-transcription rate phenotypes because all three strains spent the majority of time that was required to reach half-steady-state levels in the lag period prior to GAL1 activation ( Figure 1A; Table 1 , columns A and E; note that half steady-state levels were used instead of half-maximum levels because the level of expression during induction typically overshot the induced steady-state level ) . To accurately compare the GAL1 transcription rates of the three strains it was necessary to determine the amount of time that cultures of these strains required to reach half-steady-state levels of GAL1 expression from the time of GAL1 activation . These values were determined for each culture by subtracting its GAL1 activation time from the time required to reach the half steady-state level of GAL1 expression . This analysis revealed that once they had begun expressing GAL1 , htz1Δ and htz1-K3 , 8 , 10 , 14R cultures required 503% and 625% of the time required for HTZ1 cultures , respectively , to express GAL1 at half-steady-state levels ( Table 2 , column G ) . Thus , both H2A . Z and its acetylation contributed to the rate of GAL1 expression in cultures grown under long-term glucose repression prior to galactose induction . Because the expression of GAL1 in htz1Δ and htz1-K3 , 8 , 10 , 14R strains was similar , the role of H2A . Z in GAL1 expression was presumably dependent upon its acetylation . To determine whether H2A . Z acetylation affected the level of H2A . Z at the GAL1 promoter , chromatin immunoprecipitation experiments were performed with qPCR to quantitate the level of enrichment . Both acetylatable and unacetylatable H2A . Z were present at approximately equal levels at GAL1 ( Figure 1b ) . Therefore , acetylation of H2A . Z was important for GAL1 induction at some point after H2A . Z's incorporation at the GAL1 promoter . To determine whether H2A . Z acetylation had a role in transcriptional memory , GAL1 mRNA levels were evaluated in HTZ1 ( JRY7971 ) , htz1Δ ( JRY9001 ) , and htz1-K3 , 8 , 10 , 14R ( JRY7983 ) cultures that were grown short-term in repressing conditions prior to galactose induction . Cells grown in galactose medium prior to short-term growth in glucose medium ( 12 h ) were transferred to galactose medium and GAL1 induction was evaluated for 14 h in inducing conditions . None of the three strains exhibited a significant lag in GAL1 expression ( Figure 1C; Table 1 , column B ) . Quantitative analysis of these data suggested that all three strains , when grown short-term in repressing conditions , expressed GAL1 in half the time , or less , than when the same strains were induced following long-term growth in repressing conditions ( Table 3 , column D ) . The combined effect of near-zero onset times and increased GAL1 transcription rates was that all three strains reached half steady-state GAL1 expression levels in 90% less time than was required for the same strains to reach this level when they were grown long-term in repressing conditions prior to galactose induction ( Table 3 , column C ) . Thus , all three strains exhibited transcriptional memory with respect to GAL1 transcription . Importantly , relative to the HTZ1 strain , the two mutant strains exhibited less severe phenotypes when they were grown short-term in repressing conditions prior to induction than when they were grown long term in repressing conditions prior to induction ( Table 2 , compare column G with H ) . Thus , neither H2A . Z nor its acetylation played an important role in GAL1-transcriptional memory . Because the results described above differed substantially from ostensibly equivalent experiments [27] , we obtained the strains used in the previously published experiments , HTZ1 ( CRY 1 ) and htz1Δ ( DBY 50 ) , and attempted to reproduce the previously published results . Just as described above , both HTZ1 ( CRY 1 ) and htz1Δ ( DBY 50 ) cultures exhibited a similar lag period before GAL1 mRNA was detectable ( Figure 2A; Table 1 , column A ) . As before , when grown under long-term repressing conditions prior to galactose induction , galactose-induced HTZ1 ( CRY1 ) cells had both higher steady-state GAL1 mRNA levels and faster GAL1 transcription rates than htz1Δ ( DBY 50 ) cells ( Figure 2A; Table 1 , columns D and G ) . Quantitative analysis suggested that once both cultures had begun expressing GAL1 , the htz1Δ ( DBY 50 ) cultures required about 3 . 5× more time than HTZ1 ( CRY 1 ) cultures to reach half-steady-state GAL1 expression levels ( Table 2 , column G ) . Additionally , as was the case with the other set of strains , both HTZ1 ( CRY 1 ) and htz1Δ ( DBY 50 ) cultures induced GAL1 expression significantly more rapidly when grown short-term ( 12 h ) in repressing conditions prior to galactose induction than when the same cultures were grown long term in repressing conditions prior to galactose induction ( Figure 2B; Table 1 , column B ) . Cultures of both strains also required significantly less time to accumulate half-steady state levels of GAL1 mRNA when grown short term rather than long term in repressing conditions prior to galactose induction: HTZ1 ( CRY 1 ) and htz1Δ ( DBY 50 ) cultures required 88% and 69% less time , respectively , under these conditions to accumulate half-steady levels of GAL1 mRNA transcripts ( Table 3 , column C ) . Thus , as before , both HTZ1 and htz1Δ cultures exhibited transcriptional memory of prior GAL1 induction . Therefore , H2A . Z was important for GAL1 induction regardless of whether cells were induced from short-term or long-term growth in repressing conditions prior to induction . Two factors contribute to the GAL1 expression level in a culture of cells: the proportion of cells that are expressing GAL1 , and the level of GAL1 expression in the fraction of cells in which it is expressed . S . cerevisiae regulates GAL1 expression in response to different growth conditions both by increasing the number of GAL1-expressing cells and by increasing the level of GAL1 expression . Both parameters respond independently to different aspects of growth conditions [32] . To determine whether htz1Δ and htz1-K3 , 8 , 10 , 14R cultures' GAL1- expression defects were attributable to decreased proportions of GAL1-expressing cells , or to decreased GAL1 expression level per cell , flow cytometry was used to monitor galactose induction of a fusion protein containing the entire GAL1 coding sequence , with a C-terminal fusion to green fluorescent protein ( GFP ) , in htz1-K3 , 8 , 10 , 14R , htz1Δ , and HTZ1 cells ( Figures 3 and S6 , S7 , S8 ) . If H2A . Z were to contribute to the probability that a cell enters the galactose-induced state per unit of time , but not to the expression level in those induced cells , then htz1Δ cultures should have a smaller proportion of GFP-positive cells at each postinduction time point than HTZ1 cultures , but the GFP-positive cells should have similar fluorescence intensities to those in HTZ1 cultures . However , if H2A . Z were important for achieving high expression levels but did not influence the probability of induction per se , then htz1Δ and HTZ1 cultures should have similar proportions of GFP-positive cells , but the GAL1-GFP-expressing cells from htz1Δ mutant cultures would have lower GFP fluorescence than GAL1-GFP-expressing cells from HTZ1 cultures . The same logic would apply to the possible roles of H2A . Z acetylation . To compare the results of these experiments , a threshold value of GFP-intensity was used to classify cells as either GFP-positive or GFP-negative . This threshold was set so that between 1% and 2% of cells from noninduced HTZ1 cultures were classified as GFP-positive . On average htz1-K3 , 8 , 10 , 14R cultures had 33% fewer GFP-positive cells than HTZ1 cultures at all postinduction time points ( Figures 3 and 4A ) . Additionally , GFP-positive cells from htz1-K3 , 8 , 10 , 14R cells had , on average , 17% lower mean-GFP intensity than HTZ1 cultures ( Figures 3 and 4B ) . The simplest interpretation of these findings was that H2A . Z-acetylation influenced both the time required to induce GAL1-GFP expression and the rate at which Gal1-GFP accumulated once induced . Another possibility was that the differences between HTZ1 and htz1-K3 , 8 , 10 , 14R cells were due exclusively to differences in either the time required for induction or to the rate of Gal1-GFP accumulation . To distinguish between these two possibilities , the GAL1-induction times and Gal1-GFP accumulation rates were determined for both cultures by fitting a simple mathematical model of gene expression to the data for each culture ( the model is described in Materials and Methods; Figures 5 and S1 , S2 , S3 , S4 ) . The model simulated the galactose induction phenotype of a culture by estimating the distribution of activation times and expression rates of the measured cells . The model's parameters were fitted to the observed data for each strain by optimizing the fit to cell-specific measurements of GAL1-GFP levels . Each culture's average induction time and average accumulation rate are presented in Tables 4 and 5 , respectively . This analysis revealed that htz1-K3 , 8 , 10 , 14R cells induced GAL1-GFP expression 31% ( +/−3 . 3% ) more slowly than did HTZ1 cells ( Table 4 ) , and that induced cells in both HTZ1 and htz1-K3 , 8 , 10 , 14R cultures accumulated Gal1-GFP at similar rates ( Table 5 ) . Thus , with respect to GAL1 induction , H2A . Z-acetylation reduced the amount of time required to induce GAL1 , but did not influence the rate at which induced cells accumulated Gal1-GFPp . Interestingly , htz1Δ cells had a more severe defect in GAL1-GFP expression phenotypes than did cells with unacetylatable H2A . Z ( Figures 3 , 4A , and 4B; Tables 4 and 5 ) . On average , htz1Δ cultures had 28% fewer GFP-positive cells than htz1-K3 , 8 , 10 , 14R cultures at the 4-h and 6-h time points . At these same time points , the average GFP-intensity of GFP-positive cells in htz1Δ cultures was 46% lower than that of GFP-positive cells in htz1-K3 , 8 , 10 , 14R cultures . Moreover , htz1Δ cells induced GAL1-GFP 18 . 2% ( +/−3 . 8% ) later and accumulated Gal1-GFP 38 . 1% ( +/−5 . 6% ) more slowly than htz1-K3 , 8 , 10 , 14R cells ( Tables 4 and 5 ) . These results were surprising because htz1Δ and htz1-K3 , 8 , 10 , 14R cultures had similar GAL1 mRNA induction phenotypes ( Figure 1A ) . mRNA analysis revealed that the htz1Δ cultures used in these experiments accumulated GAL1-GFP transcripts , in contrasts to the GAL1 transcripts in Figure 1A , more slowly than either HTZ1 or htz1-K3 , 8 , 10 , 14R culture ( Figure 4C ) . These results suggested that htz1Δ cells accumulated Gal1-GFP more slowly than htz1-K3 , 8 , 10 , 14R cells because they produced GAL1-GFP mRNA more slowly than htz1-K3 , 8 , 10 , 14R cells . All of the mRNA measurements performed in this study were performed on bulk cultures , whereas the flow cytometry measurements were made on single cells within cultures . To determine whether the flow cytometry measurements of Gal1-GFP accumulation in HTZ1 , htz1Δ , and htz1-K3 , 8 , 10 , 14R strains corresponded well with each strain's GAL1-GFP mRNA accumulation phenotype , the average GFP intensity of each culture was determined ( Figure 4D ) . The galactose-induction phenotypes of all three strains , as measured by average GFP accumulation , were qualitatively similar to their galactose induction phenotypes as measured by GAL1-GFP mRNA accumulation . Thus , the flow-cytometry data in these studies reflected GAL1-GFP mRNA accumulation . At face value , the more severe galactose-induction phenotypes of htz1Δ than of htz1-K3 , 8 , 10 , 14R cells suggested that H2A . Z's role in GAL1 induction was only partially dependent on its acetylation . However , as described below , the more severe GAL1-expression defects in htz1Δ cells resulted from secondary complications that arose from the action of SWR1-Com in cells lacking H2A . Z . Acetylation of lys14 on H2A . Z is important for its role in FOX2 and POT1 induction [20] . To determine whether the acetylation of lys14 or other lysine residues of H2A . Z contributed to GAL1 induction , the GAL1 induction phenotypes of diploid cultures each with one null allele and individual lys-to-arg mutations as the other allele ( htz1-K3R/htz1Δ , htz1-K8R/htz1Δ , htz1-K10R/htz1Δ , and htz1-K14R/htz1Δ ) were determined using flow-cytometry . Surprisingly , none of the single acetylation-site mutants exhibited GAL1-GFP expression defects ( Figure 6; example FACS profiles are in Figures . S6 , S7 , S8 ) . Thus , H2A . Z's role in GAL1 induction depended on its acetylation , but did not depend exclusively on the acetylation of any single tail-lysine residue . These results were surprising given the focus on the acetylation of H2A . Z lys14 in previous studies in S . cerevisiae [18] , [19] , but they are consistent with discoveries made in Tetrahymena . In Tetrahymena , acetylation of H2A . Z's tail lysines contributes to H2A . Z's function simply by decreasing the positive charge of H2A . Z's tail and thus all sites of acetylation function equally well in this respect [33] . SWR1-Com deposits H2A . Z into chromatin in a two-step process , removing H2A from nucleosomes and subsequently replacing it with H2A . Z [13] . We hypothesized that if H2A . Z were not available , then SWR1-Com might still perform the first step of this mechanism , disrupting the structure of nucleosomes at those positions at which H2A . Z would normally reside , and that this disruption could affect normal promoter function . Thus , the phenotype of cells lacking H2A . Z might be a composite of two different defects: the lack of H2A . Z's function per se , and SWR1-Com's nucleosome-disrupting activity in the absence of H2A . Z . If this hypothesis were correct , then a subset of htz1Δ's phenotypes should be suppressed in cells lacking SWR1-Com function . Indeed as predicted by this model , strains with the htz1Δ mutation in combination with a null mutation in any gene encoding an important component of the SWR1 complex ( SWR1 , SWC2 , SWC3 , SWC5 , and SWC6 ) exhibited less severe mutant phenotypes than htz1Δ single-mutant strains on medium containing compounds that each cause a different type of stress ( Figure 7 ) . To determine if the htz1Δ mutant's galactose-induction was more defective than that of the unacetylatable H2A . Z mutant for a similar reason , the GAL1 expression phenotypes of both swr1Δ HTZ1 ( JRY9005 ) and swr1Δ htz1Δ ( JRY9006 ) double-mutant cultures were determined using flow cytometry . Prior to induction , htz1-K3 , 8 , 10 , 14R , swr1Δ HTZ1 , and swr1Δ htz1Δ cultures had similar proportions of GFP-positive cells , and fewer GFP-positive cells than htz1Δ cultures ( Figures 8 and 9A ) . Thus , the swr1Δ mutation completely suppressed the htz1Δ mutant's apparent glucose-repression defect . At every postinduction time point , swr1Δ HTZ1 and swr1Δ htz1Δ cultures had similar proportions of GFP-positive cells to htz1-K3 , 8 , 10 , 14R cultures and significantly higher proportions of GFP-positive cells than htz1Δ cultures ( Figures 8 and 9A ) . The swr1Δ HTZ1 and swr1Δ htz1Δ cells induced GAL1 expression as rapidly as htz1-K3 , 8 , 10 , 14R cells and significantly earlier than htz1Δ cells ( Table 4 ) . Thus , the severity of the htz1Δ mutant's delayed GAL1-induction phenotype was suppressible by the swr1Δ mutation and therefore likely resulted from the SWR1 complex's activity in the absence of H2A . Z . Furthermore , because htz1-K3 , 8 , 10 , 14R cells and htz1Δ swr1Δ cells needed approximately the same amount of time to induce GAL1 , H2A . Z's role in promoting rapid GAL1 activation completely depended on its acetylation . Interestingly , GAL1-expressing cells from swr1Δ HTZ1 and swr1Δ htz1Δ cultures had significantly higher average GFP intensities than those from htz1Δ cultures but they had significantly lower average GFP intensities than those in htz1-K3 , 8 , 10 , 14R cultures ( Figures 8 and 9B ) . Quantitative analysis revealed that GAL1-expressing cells from both swr1Δ HTZ1 and swr1Δ htz1Δ cultures accumulated Gal1-GFP 18 . 8% ( +/−5 . 3% ) more rapidly than htz1Δ cells and 23 . 8% ( +/−6 . 7% ) more slowly than htz1-K3 , 8 , 10 , 14R cells ( Table 5 ) . Thus the severity of the htz1Δ mutant's Gal1-GFP-accumulation-rate phenotype was suppressible by the swr1Δ mutation and therefore likely resulted from the activity of SWR1-Com in H2A . Z's absence . Moreover , our finding that swr1Δ HTZ1 and swr1Δ htz1Δ cells accumulated Gal1-GFP more slowly than htz1-K3 , 8 , 10 , 14R cells suggested that H2A . Z has an acetylation-independent role in increasing GAL1-expression rate .
In this work , we showed that H2A . Z , through its acetylation , contributed to induction of the GAL1 gene , a paradigmatic example of a highly inducible gene of Saccharomyces . Acetylated H2A . Z contributed to GAL1 induction both by increasing the fraction of cells that induced at each time point , and by increasing the level of expression per induced cell . Earlier work established that GAL1 induction has a property termed transcriptional memory , reflecting the ability of cells that were recently induced to be more easily reinduced following short incubations in repressing conditions than after extended incubations in repressing conditions [27]–[29] . Moreover , H2A . Z has been thought to be a key participant in this transcriptional memory [27] . The conclusion that H2A . Z is important for GAL1 transcriptional memory is based on experiments involving the induction of GAL1 as a function of its expression history: when induced from long-term growth in repressing conditions , both htz1Δ and HTZ1 cultures were reported to have induced GAL1 at similar rates . htz1Δ cultures were reported to have induced GAL1 at a similar rate regardless of whether they had been grown under repressing conditions for either short or long periods of time . However , HTZ1 cultures that were grown in repressing conditions for short periods of time were reported to induce GAL1 expression much more rapidly than those grown in repressing conditions for long periods of time [27] . Our work was originally directed at understanding the importance of H2A . Z acetylation to the role of H2A . Z in GAL1- transcriptional memory . To this end , we determined the GAL1- induction phenotypes of htz1-K3 , 8 , 10 , 14R cultures , which carry an unacetylatable allele of H2A . Z . Surprisingly , both htz1Δ and htz1-K3 , 8 , 10 , 14R cultures grown in inducing conditions prior to short-term growth in repressing conditions induced GAL1 expression more rapidly than those grown long-term in repressing conditions prior to galactose induction . Thus , both htz1Δ and htz1-K3 , 8 , 10 , 14R cells exhibited GAL1 transcriptional memory . Moreover , regardless of whether they were grown long-term or short-term in repressing conditions prior to induction , htz1Δ and htz1-K3 , 8 , 10 , 14R cultures induced GAL1 more slowly than HTZ1 cultures . These results indicated that H2A . Z was important for GAL1 induction regardless of a cell's growth conditions prior to induction . Thus , the galactose-induction defects that we observed for htz1Δ and htz1-K3 , 8 , 10 , 14R strains grown short-term in repressing conditions prior to galactose induction were reflective of H2A . Z and acetylated H2A . Z having a general role in GAL1 induction rather than a specific role in GAL1- transcriptional memory . If H2A . Z or H2A . Z acetylation had a specific role in transcriptional memory , then one would expect cells lacking H2A . Z or containing only an unacetylatable form of H2A . Z to exhibit more severe phenotypes when reinduced than during the primary induction . However , quantitative analysis of the GAL1 induction phenotypes of the htz1Δ and htz1-K3 , 8 , 10 , 14R strains indicated that the difference between the two mutant strains' GAL1 induction phenotypes were less severe , with respect to the HTZ1 strain's GAL1 induction phenotypes , when cultures of these strains were reinduced rather than induced . Thus , neither H2A . Z nor acetylated H2A . Z contributed to GAL1-transcriptional memory , other than in the general processes of GAL1 transcription , at least under the conditions of these experiments . To be completely clear , our data did not discount the existence of what has been referred to as transcriptional memory of GAL1 induction . A better explanation for memory has been provided by the discovery that the Gal1p protein itself has both galactokinase activity that is crucial for galactose metabolism , as well as Gal3 activity , which is also encoded by the separate GAL3 gene . Gal3p activates the GAL4-encoded activator of GAL1 induction . Thus GAL1-transcriptional memory can be explained by a positive feedback loop in which GAL1 induction leads to the synthesis of a protein that is both an enzyme and an autoinducer , as shown by others [29] , [34] . Our contribution was limited to discounting a role for H2A . Z in this memory . This understanding of GAL1-transcriptional memory suggests a possible explanation for why previously published experiments concluded that htz1Δ cultures lack GAL1- transcriptional memory [27] . The model presented above posits that a cell's ability to reinduce GAL1 expression rapidly following short-term repression requires the persistence of Gal1p in the cytoplasm . Thus the amount of time that dividing cells retain the ability to rapidly reinduce GAL1 expression following repression is a function of both the stability of Gal1p and its abundance prior to glucose repression . The abundance of Gal1p in a cell prior to glucose repression is important because of its dilution with cell division , and thus at some number of cell divisions , the amount of Gal1p will not meet the threshold level required for its role in GAL1 reinduction . We observed that htz1Δ cultures had a nearly 20% lower steady-state GAL1 expression level than HTZ1 cultures and that when grown long-term in repressing conditions prior to galactose induction htz1Δ cultures did not reach this level of expression until galactose induction had proceeded for more than 14 h . If the previously published experiments did not allow galactose induction to occur for a sufficient period of time , then the htz1Δ and HTZ1 cultures used in these experiments would not be directly comparable with respect to Gal1p levels . Thus cells within htz1Δ cultures would be less likely than those in HTZ1 cultures to have sufficiently high Gal1p levels to allow for the rapid reinduction of GAL1 . The discrepancies between the previously published data [27] and those presented here , concerning the role of H2A . Z in primary inductions of GAL1 , have a straightforward explanation . The conclusion that H2A . Z was not important for primary galactose inductions was based upon htz1Δ cells having induced GAL1 expression less well than HTZ1 cells after a 2-h induction following short-term growth in repressing conditions , whereas HTZ1 and htz1Δ cells induced GAL1 equally well following long-term growth in repressing conditions . Our observations were quantitatively similar . However , the critical point is that the magnitude of induction at this early time point was negligible in both htz1Δ and HTZ1 cultures . At all longer periods of galactose induction , htz1Δ cells induced GAL1 expression significantly less well than HTZ1 cells . We believe the earlier conclusions were based upon inadequate induction periods in some experiments . The original work implicating H2A . Z in transcriptional memory of GAL1 also reached the same conclusion for INO1 . However , the data offered in support of these conclusions are weaker than those offered in support of H2A . Z's role in GAL1 induction memory . First , these studies fail to establish that S . cerevisiae exhibits transcriptional memory of INO1 in the same way that it exhibits transcriptional memory of GAL1 . Unlike GAL1 , cells that are grown short term in repressing conditions prior to induction induce INO1 more slowly and at lower levels than cells that had been grown long term in repressing conditions prior to induction [27] . Thus , transcriptional memory of INO1 functions in the opposite way of how it functions in GAL1 transcription—decreasing rather than increasing a cell's response to inducing conditions . Second , since INO1-transcriptional memory results in slower INO1 reinductions , cells lacking INO1-transcriptional memory should induce INO1 more rapidly than cells that have INO1- transcriptional memory . These studies show that htz1Δ cells both induce and reinduce INO1 more slowly than HTZ1 cells [27] . Therefore , htz1Δ cells do not seem to lack transcriptional memory of INO1 , rather they seem to exhibit defective INO1 transcription regardless of whether they had recently induced INO1 expression . Because SWR1-Com catalyzes a two-step reaction removing H2A from nucleosomes and replacing it with H2A . Z , we considered the possibility that SWR1-Com's function , in the absence of H2A . Z , might leave those nucleosomes normally destined to receive H2A . Z compromised in some way . Thus the overall phenotype of htz1Δ would be a composite of those consequences due to the lack of H2A . Z , and those due to uncoupled H2A removal from nucleosomes . Two lines of evidence supported this hypothesis . First , the severity of htz1Δ cells' sensitivities to various agents with different mechanisms and targets were substantially suppressible by mutations in genes encoding subunits of SWR1-Com . Second , the difference between GAL1 induction in htz1Δ cells and in cells with unacetylatable H2A . Z was largely suppressed by the swr1Δ mutation , creating the less severe phenotype of the unacetylatable H2A . Z mutant . This model is further supported by the observation that htz1Δ cells have chromatin that is in the partially open configuration at the PHO5 promoter under noninducing conditions [21] . We predict that this partially open configuration is a physical manifestation of the mischief wrought by the Swr1-Complex in the absence of H2A . Z . The benomyl-sensitivity phenotype of the swc5Δ htz1Δ double mutant suggests another possible explanation for why SWR1-Com is dangerous for cells that lack H2A . Z . Unlike the swr1Δ , swc2Δ , swc3Δ , and swc6Δ mutations that strongly suppressed the htz1Δ mutant's benomyl sensitivity phenotype , the swc5Δ mutation only weakly suppressed this phenotype . In vitro studies have shown that SWR1-Com complexes lacking Swc2p , Swc6p , Swc4p , Yaf9 , or Arp6 bind nucleosomes less well than complete SWR1-Com complexes . In contrast , SWR1-Com complexes that lack Swc5p bind nucleosomes better than complete SWR1-Com complexes [35] . Since Swc5p is required for SWR1-Com's function , the simplest model for why the swc5Δ mutation does not strongly suppress the htz1Δ mutant's benomyl sensitivity is that mutant SWR1-Com complexes lacking Swc5 may persist in chromatin , perhaps removing H2A , but be unable to replace it with H2A . Z . Our observation that swr1Δhtz1Δ cells required more time to induce GAL1 expression , and expressed GAL1 more slowly once induced , suggested that H2A . Z had two distinct roles in GAL1 expression—one allowing efficient induction of GAL1 , and another to increase the rate of GAL1 expression . That H2A . Z had a role in GAL1 induction was not surprising given H2A . Z's enrichment at the GAL1-promoter . However , that H2A . Z had a role in increasing GAL1's expression rate , as inferred from our model , was unexpected . There are two lines of evidence that H2A . Z may be important for the expression , per se , of actively transcribed genes . First , even though H2A . Z predominantly localizes to promoters , it is not completely absent from open reading frames ( ORFs ) . The ACT1 and PRP8 ORFs , two loci that have been historically considered nonenriched for H2A . Z , are slightly enriched for H2A . Z relative to no-tag controls ( Figure S5 ) . Second , the htz1Δ mutant is sensitive to 6-azauracil , a toxic compound that slows the growth rate of cells that are defective in mRNA transcript elongation [36] . Thus , it is possible that H2A . Z plays a direct role in transcript elongation . Recent reports raise that possibility further , showing that H2A . Z may aid expression by suppressing antisense transcripts [37] . In summary , our results established that H2A . Z plays no significant role in GAL1- transcriptional memory . In contrast H2A . Z , and its acetylation contributed to both the induction of the gene and to its expression per se , adding valuable new insights into one of the best-studied examples of eukaryotic gene regulation . In addition , we showed that SWR1-Com caused defects in gene expression and induction in the absence of H2A . Z , presumably due to nucleosome disruption , that force a reevaluation of all previously described phenotypes of cells lacking H2A . Z .
All of the strains used in this study are presented in Table 6 . All of these strains were from the W303 background . One-step integration of knockout cassettes has been previously described [38] . JRY9001 was constructed by transforming the KanMX cassette into JRY7754 . To generate KWY2512 , the DNA sequence encoding GFP was inserted before the stop codon of the GAL1 open reading by transforming a HIS3-marked construct encoding the GFP protein . JRY9002 , JRY9003 , JRY9004 were segregants from crosses of JRY7972 , JRY7983 , and JRY9001 to KWY2512 , respectively . JRY9005 and JRY9006 were segregants from crosses of JRY7752 to JRY9002 and JRY9004 , respectively . JRY9011 , JRY9012 , JRY9013 , JRY9014 , JRY9015 , and JRY9016 were segregants from crosses of JRY9000 to JRY7972 , JRY7983 , JRY9007 , JRY9008 , JRY9009 , and JRY9010 , respectively . MKY1028/MKY1029 , MKY1030/MKY1031 , MKY1032/MKY1033 , MKY1034/MKY1035 , and MKY1036/MKY1037 were created by disrupting SWR1 , SWC2 , SWC3 , SWC5 , and SWC6 respectively in MKY1038 using a SpHIS5MX knockout cassette that was amplified from pFA6a-His3MX6 [38] . Yeast media were as defined [39] . Seed culture density affected GAL1 induction phenotypes , so precautions were taken to ensure that seed cultures of all strains had similar growth histories . Specifically , seed cultures for all experiments were grown in YP-Dextrose ( D-glucose , 2% ) except DBY50 and CRY1 , which were grown in CSM-Dextrose ( D-glucose , 2% ) . 50 ml seed cultures were inoculated with cells from a single colony and grown overnight with shaking at 30°C to OD 0 . 2 , and were then harvested by centrifugation at 2 , 060g for 1 min . The cells were then washed with 25 ml of prewarmed 30°C YP-galactose and resuspended in 50 ml of 30°C YP-galactose , except in experiments performed with DBY50 and CRY1 , in which CSM-Galactose was used instead of YP-galactose for both washing and resuspending in order to follow precisely the procedures of others [27] . The volume of culture removed for each time point was replaced with the same volume of 30°C YP-galactose . Both determination of mRNA levels by quantitative reverse-transcriptase ( Q-RT ) PCR and ChIP were performed as described [17] except that SYBR GreenER ( Invitrogen ) PCR reagents were used . H2A . Z-3Flag , and H2A . Z-K3 , 8 , 10 , 14R-3Flag were immunoprecipitated using the αFlag M2 resin ( Sigma ) . Cells were harvested by centrifugation , fixed in a 4% paraformaldehyde/3 . 4% sucrose solution for 10 min at room temperature and then stored overnight at 4°C in a 1 . 2 M sorbitol solution with KPO4 buffer at pH 7 . 5 . GFP expression data were collected for each sample using the FC-500 ( Beckman-Coulter ) flow cytometer and analyzed using the Flow-Jo software package . The GAL1-GFP expression status of individual cells within cultures on a cell-by-cell basis in each culture was determined by plotting flow-cytometry measurements as a histogram of GFP fluorescence ( y-axis number of cells; x-axis Log GFP intensity relative to GFP-negative values ) . The threshold of GFP intensity was set so that between 1% and 2% of glucose-grown HTZ1 cultures would be classified as GFP-positive . Cells that had GFP-intensity greater than this threshold value were counted as GFP positive ( GAL1-GFP expressing ) . The level of GAL1 expression in different populations was calculated by determining the geometric mean GFP intensity . We developed a simple mathematical model to analyze the dynamics of GAL1 mRNA expression levels . This model allowed us to robustly quantify the onset time of GAL1 induction , steady state GAL1 mRNA level , and the time needed to reach half of the steady-state level . The model is based on three parameters , which we optimized to maximize the fit of the model to the measured GAL1 mRNA levels . These include: ( 1 ) the time x when induction of GAL1 mRNA begins; ( 2 ) the rate α at which GAL1 mRNA is produced; and ( 3 ) the rate δ at which GAL1 mRNA molecules are being degraded . According to the model , the relative amount of GAL1 mRNA at time t , M ( t ) , follows the ordinary differential equation ( ODE ) :Namely , GAL1 is not being expressed at all until time point x , from which point it is produced at a fixed rate α , and being degraded at a fixed ratio δ , until it reaches the steady state equilibrium:Given the model parameters , and starting from zero M ( 0 ) = 0 , we can solve the ordinary differential equation using the Runge-Kutta method ( as implemented in MATLAB 7 . 6 ) , and estimate the mRNA level of GAL1 at every time point t . We optimized the three parameters x , α , and δ for every culture to minimize the root-mean-square deviation ( RMSD ) between the experimental measurements and the modeled values . The values that were used in each of the best-fit models are presented in Table S1 . We constrained the parameters x , α , and δ to non-negative values , and used the active-set optimization algorithm ( FMINCON function in MATLAB 7 . 6 ) . For the memory experiments , the optimized values of the GAL1 expression onset times , for all cultures , were very close to zero , and practically below the time resolution of the model and data . We therefore simplified the model , and explicitly set x to zero . Finally , to estimate the half steady-state time point , we used the optimized parameters for each culture to find the steady state level α/δ , and to solve the ordinary differential equation and identify when GAL1 levels reach half of the steady state level . To analyze the flow cytometry data , the time-course measurements of single-cell Gal1-GFPp intensities were transformed into GAL1-GFP induction times and Gal1-GFP accumulation rates . To do this , a simplified model of GAL1-induction was developed , and its six parameters fitted to the measured data for each culture . For every cell , this model assumes that GAL1 is completely repressed until its induction time ti , when cellular Gal1-GFPp begins to accumulate at a fixed rate xi . We therefore model Ei ( t ) , the Gal1-GPFp content of the ith cell at time t as:where: The estimated expression is added to a stochastic noise term εi , drawn from a Normal distribution with parameters ( μ , σ2 ) , to simulate a basal level of GAL1 expression . The model was used to simulate a population of 100 , 000 cells , whose GAL1-GFP-induction times ti's and accumulation rates xi's were sampled independently from two Gamma distributions: ti ∼ Gamma ( kt , θt ) , and xi ∼ Gamma ( kx , θx ) , and their stochastic noise terms sampled from a Normal distribution: εi ∼ Normal ( μ , σ2 ) . Given a set of six parameters ( kt , θt , kx , θx , μ , σ2 ) this model sampled activation times , accumulation rates , and noise terms for each of the 100 , 000 cells in the simulation , and computed the cellular Gal1-GFPp levels Ei ( t ) for each of the four times points that were measured ( 0 , 2 , 4 , and 6 h following induction ) , which allowed for the simulation of flow-cytometry outputs . Activation times and accumulation rates were sampled from a stochastic distribution rather than being fixed at specific values to account for the natural variability among cells because of biological variables like cell size , position in the cell cycle , cell age , and other factors that were not treated as variables in the model . Gamma distributions were used due to their non-negativity property . The parameters of the model were optimized by minimizing the root-mean-squared deviation between the measured data ( average of triplicates ) and the model predictions , summed over the four measured time points ( 0 , 2 , 4 , and 6 h . ) To optimize these parameters , genetic algorithms were used ( as implemented in the GA function in MATLAB 7 . 6 ) followed by a derivative-free optimization using the simplex algorithm ( FMINSEARCH function in MATLAB 7 . 6 ) . These optimization steps were repeated with 200 random starting points for each strain , and the optimal set of parameters were then selected ( Tables S2 and S3 ) . The error in our estimation of each strain's induction time and accumulation rate was calculated by determining the range of values for each parameter that were used in the top 50 best-fit simulations for each strain . The models that were determined for each strain's Gal1-GFP expression phenotype were used as a proxy to quantitatively compare the GAL1-activation times and Gal1-GFPp accumulation rates of HTZ1 , htz1-K3 , 8 , 10 , 14R , htz1Δ , swr1Δ HTZ1 and swr1Δ htz1Δ cells . | Transcriptional memory is the well-documented phenomenon by which cells can “remember” prior transcriptional states . A paradigmatic example of transcriptional memory is found in the yeast Saccharomyces . S . cerevisiae remembers prior transcription of the galactose metabolism gene GAL1 . When a gene is transcribed , the DNA must first be at least partially relieved of its packaging into chromatin by histone proteins . Previous research had suggested that S . cerevisiae used a chromatin modification , the incorporation of the histone variant H2A . Z into the region surrounding the GAL1 promoter , to remember the previous status of GAL1 transcription . Not all H2A . Z molecules are the same , however . For example , it has recently been discovered that H2A . Z can be acetylated on the four lysine residues in its N-terminal tail region . In an attempt to determine whether H2A . Z acetylation is required for GAL1 transcriptional memory , we unexpectedly discovered that , although both H2A . Z and H2A . Z acetylation are important for strong and rapid GAL1 induction , neither H2A . Z nor H2A . Z acetylation plays an important role in GAL1 transcriptional memory . We propose that the discrepancy between our conclusions and those in prior publications arise from the prior analysis of insufficiently short periods of GAL1 induction or from complications arising from the comparison of the phenotypes of wild-type yeast strains to those of htz1Δ mutants ( carrying the null mutation of the gene encoding H2A . Z ) mutants . In the current work we show that the htz1Δ mutant's phenotype does not simply reflect the absence of H2A . Z in chromatin but instead also reflects the pleiotropic effects of the Swr1 chromatin remodeling complex that is responsible for H2A . Z deposition into chromatin . In the absence of H2A . Z the Swr1 complex itself causes cell damage . In this paper we show that swr1Δ htz1Δ double mutants have substantially less severe mutant phenotypes than htz1Δ mutants . Thus , studies using the swr1Δ htz1Δ mutant offer more detailed insight into the consequences of the absence of H2A . Z in chromatin than do studies performed on single htz1Δ mutants , and our results help to clarify the role of H2A . Z in the regulation of GAL1 induction and transcriptional memory . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology/gene",
"expression",
"genetics",
"and",
"genomics/gene",
"expression"
] | 2010 | Roles for H2A.Z and Its Acetylation in GAL1 Transcription and Gene Induction, but Not GAL1-Transcriptional Memory |
The case-fatality rate of severe leptospirosis can exceed 50% . While prompt supportive care can improve survival , predicting those at risk of developing severe disease is challenging , particularly in settings with limited diagnostic support . We retrospectively identified all adults with laboratory-confirmed leptospirosis in Far North Queensland , Australia , between January 1998 and May 2016 . Clinical , laboratory and radiological findings at presentation were correlated with the patients’ subsequent clinical course . Medical records were available in 402 patients; 50 ( 12% ) had severe disease . The presence of oliguria ( urine output ≤500 mL/24 hours , odds ratio ( OR ) : 16 . 4 , 95% confidence interval ( CI ) : 6 . 9–38 . 8 , p<0 . 001 ) , abnormal auscultatory findings on respiratory examination ( OR 11 . 2 ( 95% CI: 4 . 7–26 . 5 , p<0 . 001 ) and hypotension ( systolic blood pressure ≤100 mmHg , OR 4 . 3 ( 95% CI 1 . 7–10 . 7 , p = 0 . 002 ) at presentation independently predicted severe disease . A three-point score ( the SPiRO score ) was devised using these three clinical variables , with one point awarded for each . A score could be calculated in 392 ( 98% ) patients; the likelihood of severe disease rose incrementally: 8/287 ( 3% ) , 14/70 ( 20% ) , 18/26 ( 69% ) and 9/9 ( 100% ) for a score of 0 , 1 , 2 and 3 respectively ( p = 0 . 0001 ) . A SPiRO score <1 had a negative predictive value for severe disease of 97% ( 95% CI: 95–99% ) . A simple , three-point clinical score can help clinicians rapidly identify patients at risk of developing severe leptospirosis , prompting early transfer to referral centres for advanced supportive care . This inexpensive , bedside assessment requires minimal training and may have significant utility in the resource-limited settings which bear the greatest burden of disease .
Leptospirosis is a zoonotic infection with a global distribution [1 , 2] . Although most infections are mild and self-limiting , the disease is believed to kill almost 60 , 000 people every year [1] . Severe disease–manifesting as pulmonary haemorrhage , acute kidney injury ( AKI ) or multiorgan failure–develops in 5–15% of cases . The case-fatality rate of severe leptospirosis is as low as 6% if there is prompt access to vasopressors , renal replacement therapy ( RRT ) and mechanical ventilation [3] , but it can rise to greater than 50% if the delivery of this supportive care is delayed [4] . However , identifying the patients who are at risk of developing severe disease can be difficult . Different studies have suggested that the presence of a variety of clinical features , laboratory investigations and imaging and electrocardiography findings can help [5–10] . While these approaches may be helpful in well-resourced settings where there is access to advanced laboratory and radiology support , they may have less utility in low and middle-income countries ( LMIC ) , which bear a disproportionate burden of the disease [1] . Leptospirosis is endemic in tropical northern Australia , and the state of Queensland has one of the highest reported incidences in the developed world [11] . Most of the cases in Queensland occur in relatively remote locations where there is limited access to diagnostic support . Accordingly , given the potential for patient deterioration , if there is clinical uncertainty about a patient’s prognosis , they are often transferred–sometimes great distances–to a tertiary centre for continuing care . Not only is this frequently unnecessary , it is inconvenient for patients and their families , and expensive for the health system . To improve the triage of patients with leptospirosis , and identify patient characteristics that predict severe disease , we reviewed the presentation of adults with confirmed leptospirosis in Far North Queensland and correlated their clinical findings and laboratory and imaging results with their subsequent clinical course . Our aim was to produce a simple score that could be used to quickly identify the patients at greatest risk of deterioration , expediting their referral for intensive care unit ( ICU ) support . We also hoped that the score could predict which patients could be safely managed without transfer , providing reassurance for local clinicians and reducing costs for the health system . Recognising that leptospirosis has a significant burden in LMIC–and in remote locations in high-income countries–it was also hoped that the score that might be applicable where access to diagnostic support is limited .
This retrospective study was performed at Cairns Hospital , a 531-bed , tertiary referral hospital in tropical , northern Australia that–with 16 smaller community hospitals–provides medical services to a population of approximately 280 , 000 people across an area of 380 , 000km2 . The local electronic pathology reporting system ( AUSLAB ) was used to identify all leptospirosis cases in the region between January 1998 and May 2016 . Adult patients ( ≥16 years of age ) were defined as having confirmed leptospirosis if they met one or more of the following criteria: ( 1 ) Leptospires isolated from blood culture; ( 2 ) Microscopic agglutination test ( MAT ) single titre of ≥ 1:400; ( 3 ) Fourfold rise in MAT antibody titres; ( 4 ) Detection of Leptospira in blood by polymerase chain reaction ( PCR ) . Medical charts were reviewed at the hospital of first presentation and at Cairns Hospital if a patient required inter-hospital transfer . It was recognized that a proportion of the medical records would be unavailable as the health service has a policy of destroying the paper medical record if there have been no new patient encounters for ten years . Patient characteristics at the time of presentation to medical attention were reviewed . The World Health Organization was undertaking enhanced surveillance of leptospirosis in the region and a case report form was in use for much of the study period . Clinical findings , haematology , biochemistry , urinalysis , chest x-ray and electrocardiogram results were recorded . The following cut-offs–based on the literature , reference ranges and everyday clinical practice–were used to characterise any association severe disease: hypotension ( systolic blood pressure ≤100 mmHg ) , anaemia ( haemoglobin ≤100 g/L ) , severe thrombocytopenia ( platelets ≤50 x 109/L ) , acidosis ( bicarbonate ≤22 mmol/L ) , AKI ( creatinine ≥2 mg/dL ) , jaundice ( bilirubin ≥3 mg/dL ) and C-reactive protein ≥200 mg/L . The quick Sequential Organ Failure Assessment ( qSOFA ) and the quick National Early Warning Score ( qNEWS ) scores were calculated for the patients with sufficient clinical information [12 , 13] . Severe disease was defined as the development of pulmonary haemorrhage , ICU admission , or a requirement for RRT , intubation or vasopressor support . Pulmonary haemorrhage was said to be present if there was frank haemoptysis or if blood was present on tracheal aspirate . The study obtained approval from the Far North Queensland Human Research Ethics Committee ( HREC/16/QCH/37 – 1043LR ) . As per the approval , this retrospective study used anonymized patient data and did not obtain individual patient consent . This study reviewed human patients only; no animals were involved in any aspect of the study . Data were entered into an electronic database ( Microsoft Excel ) and analysed using statistical software ( Stata 14 . 0 ) . Groups were analysed using the Kruskal-Wallis and chi-squared tests . Multivariate analysis was performed using backwards linear and logistic regression . For the multivariate analysis , only variables with an area under the receiver operating characteristic ( AUROC ) curve of >0 . 7 in univariate analysis were selected .
There were 738 cases of laboratory-confirmed leptospirosis during the study period . Medical charts were available in 429 cases; 402 ( 94% ) were adults . Their median ( interquartile range ( IQR ) ) age was 33 ( 23–45 ) years; 362 ( 90% ) were male . Nearly all the cases ( 397/402 ( 99% ) ) were acquired locally , 273/397 ( 69% ) occurred during the region’s November-April wet season , and 355/397 ( 89% ) occurred in a region of high-intensity banana and dairy cattle farming situated approximately 100km south of Cairns . In the 384 in whom an occupation was documented , 327 ( 85% ) had the potential for occupational exposure . There were 50 ( 12% ) patients who developed severe disease , including two ( 0 . 5% ) who died ( Fig 1 ) . In 331/402 ( 82% ) cases , clinicians included leptospirosis in the differential diagnosis at the time of presentation . Leptospires were isolated from blood culture in 275/402 ( 68% ) , MAT was diagnostic in 178/402 ( 44% ) and PCR was positive in 151/402 ( 38% ) . Serovars could be determined in 353/402 ( 88% ) ; Australis and Zanoni were the commonest serovars , and the most likely to cause severe disease ( Table 1 ) . The median ( IQR ) duration of symptoms was 4 ( 3–6 ) days in the patients who developed severe disease compared with 3 ( 2–4 ) days in those that did not ( p = 0 . 0001 ) . Comorbidities were documented in 395 patients and were more common in the patients who had severe disease ( 15/49 ( 31% ) than those that did not ( 29/346 ( 8% ) , p<0 . 0001 ) . Clinical findings were similar to those in the published literature , although only 9% had a serum bilirubin ≥3mg/dL and conjunctival suffusion was documented in only 109/397 ( 27% ) who were assessed . The symptoms , signs and laboratory tests and their association with severe disease are presented in Tables 2 and 3 . In univariate analysis , the symptoms of diarrhoea , dyspnoea and bleeding were most associated with the development of severe disease ( Table 2 ) . The presence of renal impairment and thrombocytopenia were the laboratory tests most associated with the development of severe disease ( Table 3 ) . In the 50 patients with severe disease , 45 ( 90% ) required ICU admission , 27 ( 54% ) developed pulmonary haemorrhage , 27 ( 54% ) required vasopressor support , 18 ( 36% ) required RRT and 24 ( 48% ) required mechanical ventilation . APACHE III scores were available for 39/45 ( 87% ) patients admitted to ICU; the median score was 84 ( range 27–169 ) . There were only two deaths in the study . The first was an 80-year-old man with a history of diabetes mellitus and 5 days of symptoms; he developed multiorgan failure and died one day after presentation despite ICU support . The second , a 73-year-old man with a history of cardiovascular disease , chronic lung disease and connective tissue disease , had four days of symptoms; he also had multiorgan failure and died within one day of presentation . Multivariate analysis identified four independent variables associated with severe disease; three of these variables were clinical signs–abnormal auscultatory findings on respiratory examination ( odds ratio ( OR ) ( 95% CI ) : 8 . 9 ( 3 . 5–22 . 4 ) , p<0 . 0001 ) , oliguria ( OR ( 95% CI ) : 8 . 2 ( 3 . 2–21 . 2 ) p<0 . 0001 ) and hypotension ( OR ( 95% CI ) : 3 . 8 ( 1 . 5–9 . 9 ) , p = 0 . 006 ) and one was a laboratory variable ( creatinine ≥2 mg/dL ) ( OR ( 95% CI ) : 7 . 0 ( 2 . 7–18 . 1 ) p<0 . 0001 ) . The three clinical findings–awarded one point each–were used to generate a three-point SPiRO score ( Systolic blood Pressure ≤100 mmHg , Respiratory auscultation abnormalities , Oliguria , Table 4 ) . The risk of severe disease increased incrementally with the SPiRO score . A score of zero had a negative predictive value ( NPV ) for severe disease of 97 . 2% ( 95% CI: 94 . 6–98 . 8% ) . A score greater than one had a positive predictive value ( PPV ) ( 95% CI ) for severe disease of 77 . 1% ( 59 . 9–89 . 6 ) , while a score of three had a PPV of 100% ( 66 . 4–100 ) ( Table 5 and Fig 2 ) . The predictive ability of the SPiRO score was compared with the qSOFA and the qNEWS scores . In the 379/402 ( 94% ) patients in whom the scores could be calculated , the AUROC of the SPiRO score ( 0 . 87 ( 95% CI 0 . 81–0 . 9 ) was higher than that of the qSOFA ( 0 . 76 ( 95% CI 0 . 70–0 . 83 ) score ( p = 0 . 003 ) . The difference between the AUROC of the SPiRO score and the qNEWS score ( 0 . 81 ( 95% CI 0 . 74–0 . 87 ) failed to reach statistical significance ( p = 0 . 053 ) .
In adults with leptospirosis , a simple three-point clinical score–the SPiRO score–appears to reliably identify patients at risk of severe disease . The score could be used anywhere that leptospirosis is seen , but as a rapid and inexpensive assessment , which can be performed at the bedside by even junior health care workers , it has significant appeal for a disease that has its greatest impact in resource-poor settings . An absence of hypotension , oliguria or abnormal auscultatory findings–a SPiRO score of zero–was particularly helpful in identifying low-risk patients . The score could therefore determine which patients can be safely managed in remote locations , avoiding unnecessary and expensive transfer . The likelihood of severe disease rose incrementally as the score increased , facilitating recognition of the high-risk patient , expediting the initiation of supportive treatment and prompting consideration of transfer to referral centres . Although many variables have been shown to predict severe leptospirosis , a simple scoring system to quantify the relative risk of severe disease has proven elusive [5] . As in this series , older age has been associated with severe leptospirosis and worse outcomes in multiple countries including India [4] , Brazil [14] and Turkey [15] . Other predictors of severe leptospirosis or leptospirosis-attributable mortality have consisted of a combination of clinical features , laboratory findings and interpretation of imaging and electrocardiography . It is notable that pulmonary involvement is associated with worse outcomes in almost every published series , while renal involvement and hypotension also have also been shown to have significant prognostic utility . In the French West Indies , dyspnoea , oliguria , white blood cell count >12 , 900/mm3 , alveolar infiltrates on chest X-rays and repolarization abnormalities on electrocardiograms were independently associated with death [10] . In Brazil , pulmonary involvement , oliguria , creatinine >3 mg/dL and platelets <70 , 000/mm3 were independent predictors of mortality , with pulmonary involvement being the strongest prognostic factor [16] . Similarly , in India [17 , 18] , Indonesia [19] , and Greece [20] , pulmonary involvement was associated with increased mortality . In Thailand , pulmonary rales , oliguria , hypotension and hyperkalaemia were all independently associated with death [21] . While an elevated serum creatinine , white cell count and thrombocytopenia were also associated with severe disease in our series , it is important to remember that laboratory support may be limited in the rural and remote settings of LMIC where most cases of leptospirosis are seen . Even where there is access to laboratory support in these locations , results are not always available promptly . Similarly , while an abnormal chest X-ray had prognostic utility in our series , there may not be access to radiology support in leptospirosis-endemic areas and even when there is , accurate interpretation of imaging findings requires high quality images and significant medical training . Finally , although electrocardiography is inexpensive and relatively easy to perform , the identification of repolarization abnormalities also requires some expertise . These issues may also be relevant in high-income settings like Australia . In rural locations where most cases of leptospirosis are diagnosed , it can take up to 24 hours for the processing , transport and analysis of even simple haematology and biochemistry tests such as platelet count or serum creatinine . Other laboratory tests that have been linked to severe leptospirosis , such as the quantification of leptospires in blood , are unlikely to be routinely available in the foreseeable future [6 , 22 , 23] . Imaging is also not necessarily accessible , and patients are usually reviewed initially by junior staff . The entirely clinical SPiRO score therefore has significant appeal . It is simple to perform , reproducible , requires little medical training and addresses the renal impairment , pulmonary involvement and hypotension that have repeatedly been shown to be associated with the worst clinical outcomes [10 , 16–21] . AKI in leptospirosis occurs due to direct leptospire invasion resulting in tubulointerstitial nephritis [24] . Renal biopsy most frequently reveals a mononuclear cellular infiltration and interstitial oedema , although an immune-complex glomerulonephritis may also be present [25 , 26] . While leptospirosis has traditionally been thought to cause non-oliguric AKI [27] , oliguria is an early clinical marker of AKI that is less likely to respond to rehydration and more likely to require RRT [28] . Hypotension in leptospirosis is usually due to vasodilatory mediators and proinflammatory cytokines released in response to the infection; this results in reduced renal blood flow , further exacerbating renal injury [29] . Pulmonary involvement–perhaps the most serious manifestation of severe disease–is frequently overlooked [23 , 30] . Leptospirosis impairs the fluid handling of alveolar epithelial cells resulting in pulmonary oedema which can trigger acute respiratory distress syndrome [31–33] . Pulmonary haemorrhage–the most feared respiratory manifestation–is thought to occur from a direct effect of leptospiral proteins or toxic cellular components on multiple components of the alveolocapillary membrane [34] . As larger areas of haemorrhage coalesce , symptoms worsen and clinical signs are more likely to be apparent on auscultation [35] . As pulmonary haemorrhage progresses , pulmonary vascular resistance also increases , further contributing to systemic hypotension [29] . Severe disease was common in our series , but the case-fatality rate was very low with both deaths occurring in elderly patients with significant comorbidities . The wide variation in case-fatality rates reported in the literature has been attributed to differing definitions of severe leptospirosis , although our definition was conservative and the patients’ APACHE III scores were high . It is possible that the local case-fatality rate is higher than we have reported as patients with rapidly fatal leptospirosis may have a negative MAT test early in their disease . However , culture and PCR are used widely locally with only 18% of cases were diagnosed by MAT alone and accordingly , the number of unrecognized , fatal cases is probably small . The excellent outcomes are likely to be the result of early disease recognition and access to prompt ICU support . There is an extensive medical retrieval network in Australia which ensures people living in rural and remote services have access to sophisticated healthcare . However , given the country’s great expanse , these services are costly , frequently relying on a combination of road ambulances , helicopters and aeroplanes . While the coordination of retrieval services is centralised and efficient , organising safe and appropriate medical evacuation is time-consuming . A simple scoring system that facilitates recognition of patients with severe leptospirosis could help guide clinicians to identify which patients are most likely to require critical care support and early referral to retrieval services . Conversely , the SPiRO score could help prevent unnecessary medical evacuation , which would be welcome for patients and reassuring for the clinicians involved in their care . Our study was retrospective and the SPiRO score requires prospective validation to ensure its applicability in other geographical settings . However , in other locations including Brazil [14 , 36] , Thailand [21] , Moldova [37] , Greece [20] , and Réunion [38] , hypotension , oliguria or abnormal respiratory auscultation have been identified previously as independent predictors of severe disease . Indeed , in a much smaller series from French Polynesia the same three clinical parameters were found to be the only independent variables in predicting severe disease [39] . The SPiRO score therefore has potential global utility . While many counties with a high incidence of leptospirosis do not have access to medical retrieval services or advanced ICU support , the score may still identify those who may benefit from closer monitoring and might be expected to improve outcomes . The clinical findings of leptospirosis have been linked to the infecting serovar , with serogroup Icterohaemorrhagiae particularly associated with severe disease [6 , 35] . This serogroup was uncommon in our series , occurring in only two cases , while severe disease was most commonly associated with serovars Australis and Zanoni . Variations in the prevalence of different serogroups and serovars have the potential to limit the generalizability of our findings , however , as previously noted , the clinical phenotype seen in our cohort was remarkably similar to that seen in the published literature [3 , 10 , 35] . Evidently , the SPiRO score can only be applied to patients with a diagnosis of leptospirosis , a condition whose prompt diagnosis remains challenging . While clinical findings can inform the clinician , they are non-specific and may not differentiate leptospirosis from other tropical infections including rickettsial disease , malaria and dengue . PCR is rarely available where the disease is endemic , and even in well-resourced settings like Australia , results take several days . Point-of-care tests have the greatest potential to facilitate diagnosis in both resource-poor and rich countries , but although their sensitivity and specificity is improving , these tests are not currently in routine use [40 , 41] . If reliable point-of care tests can be developed and coupled with a validated simple predictive tool , the early recognition and management of leptospirosis is likely to improve significantly . That being said , clinicians in leptospirosis-endemic areas often recognize the disease–in our series , leptospirosis was in the initial differential diagnosis in over 82% of cases . Furthermore , even when the diagnosis of leptospirosis cannot be confirmed , a patient presenting to a remote clinic with hypotension and evidence of pulmonary and renal disease is likely to require referral for more sophisticated care whatever the aetiology . The retrospective nature of our study meant that documentation was sometimes incomplete , and investigations were not standardised . However , clinicians working in the area have a high index of suspicion for the disease and a leptospirosis pro forma was in use for most of the study period . As a result , the clinical features on presentation were generally well documented . Data are now being collected prospectively to confirm these preliminary observations . The antibiotic therapy , its timing , route and duration were available in most cases and almost all patients received at least one appropriate agent . However , the enormous variety of antibiotic regimens prescribed precluded meaningful analysis of their relative efficacies . In conclusion , a simple three-point clinical based scoring tool appears to help clinicians identify people at risk of developing severe leptospirosis . The score requires prospective validation in other geographical locations , but it has the potential to improve the care of people with leptospirosis , particularly in resource-limited settings where the disease has its greatest clinical burden . | Leptospirosis , a neglected tropical disease with a global distribution , is estimated to kill 60 , 000 people every year . Predicting those at risk of developing severe disease is challenging , and a simple scoring system to quantify the risk of severe disease has proven elusive . Identifying the high-risk patient is important , as it might expedite the initiation of life-saving supportive care . This review of 402 adult patients with leptospirosis in tropical Australia determined that three clinical variables identified at presentation independently predicted severe disease ( a subsequent requirement for Intensive Care Unit admission , intubation , vasopressor support , renal replacement therapy or the development of pulmonary haemorrhage ) . These three variables ( abnormal auscultatory findings on respiratory examination , hypotension and oliguria ) were used to generate a simple , three-point clinical score which can be determined rapidly and reliably at the bedside by health care workers with minimal training . This simple score may help the clinical management of patients with leptospirosis , particularly in lower and middle-income countries that bear the greatest burden of disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"clinical",
"laboratory",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"respiratory",
"infections",
"tropical",
"diseases",
"geographical",
"locations",
"australia",
"biomarkers",
"pulmonology",
"bacterial",
"diseases",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"hypotension",
"infectious",
"diseases",
"zoonoses",
"clinical",
"laboratories",
"blood",
"pressure",
"leptospirosis",
"creatinine",
"people",
"and",
"places",
"biochemistry",
"diagnostic",
"medicine",
"oceania",
"hemorrhage",
"biology",
"and",
"life",
"sciences",
"vascular",
"medicine"
] | 2019 | A simple score to predict severe leptospirosis |
Natively unstructured or disordered regions appear to be abundant in eukaryotic proteins . Many such regions have been found alongside small linear binding motifs . We report a Monte Carlo study that aims to elucidate the role of disordered regions adjacent to such binding motifs . The coarse-grained simulations show that small hydrophobic peptides without disordered flanks tend to aggregate under conditions where peptides embedded in unstructured peptide sequences are stable as monomers or as part of small micelle-like clusters . Surprisingly , the binding free energy of the motif is barely decreased by the presence of disordered flanking regions , although it is sensitive to the loss of entropy of the motif itself upon binding . This latter effect allows for reversible binding of the signalling motif to the substrate . The work provides insights into a mechanism that prevents the aggregation of signalling peptides , distinct from the general mechanism of protein folding , and provides a testable hypothesis to explain the abundance of disordered regions in proteins .
The biological function of many proteins is determined by their native , three-dimensional structure and unfolded ( or incorrectly folded ) copies of such proteins tend to be inactive , if not outright dangerous . However , many proteins contain large regions ( >30 amino acids ) that are disordered in their natural physico-chemical environment [1]–[4]; some proteins are even entirely disordered [5] , [6] . As more peptide sequences are being studied , it is becoming increasingly clear that natively-disordered sequences are far more common than previously thought . Disordered sequences have been found on a large number of eukaryotic genes ( >30% ) [2] , [5] , [7] , [8] . Moreover , the number of genes on a genome with disordered regions appears to increase with the complexity of the species [2] , [5] , [7] , [8] . Despite a lack of stable structure in the native form of the protein , disorder is strongly associated with specific cellular functions , most significantly with cell signalling and regulatory processes [9]–[14] . Several suggestions have been made about the possible benefits of disordered regions in a protein: they could be more malleable , have a large binding surface , bind to diverse ligands , bind with high specificity and make the binding process reversible [1] , [12] , [15] , [16] . Indeed , there exist numerous examples of natively disordered proteins that form a more defined structure upon binding to a ligand [17] , implying that the protein loses conformational entropy on binding . Disordered regions ( peptide sequences that are generally unfolded ) and natively unstructured binding regions ( sequences that only take a specific structure upon binding ) have some general features . Disordered regions contain fewer hydrophobic , more hydrophilic , more charged amino acids and more repeats in their sequence as compared to natively structured proteins [6] . On the other hand interfacial regions between a natively unstructured binding region and a rigid protein contain relatively more hydrophobic and fewer charged contacts , as compared to rigid-rigid interfaces [18] . In general , only a small ( hydrophobic ) motif of the disordered region is involved in the actual binding and this binding motif remains in an extended configuration even upon binding and ‘folding’ [19]–[21] . As a consequence , the exposed binding area per residue is relatively large [15] , [18] ( see Figure 1 ) . Recent studies have revealed that many small ( linear ) binding motifs are surrounded by disordered regions [22] , [23] . A typical linear binding motif contains some 6 residues and is surrounded by approximately 20 residues that are natively unstructured [23] . The binding motifs are typically more hydrophobic than the flanking residues . Since the binding regions are relatively small , they are unlikely to form fully folded ( or specific ) structures in solution when not bound to a substrate . In this study we focus on the steric effects of the disordered regions adjacent to small hydrophobic binding motifs . As the presence of disordered regions near small binding motifs appears to be generic , it seems justified to use a generic model . The nature of the coarse-grained model allows us to simulate the specificity , steric hindrance , configurational and translational entropy of the peptide chain . Each residue of the peptide chain occupies a single point on a cubic lattice . The lattice makes efficient movements in the peptide chain possible so that many different configurations of the chain can be sampled with a Monte Carlo algorithm . Residues on neighbouring lattice points interact in a pairwise manner . Each of the 20 amino acids has a specific interaction energy with each of the other amino acids [24] , [25] . For example , two neighbouring hydrophobic amino acids lower the internal energy and are thus attracted to each other . The large number of possible interactions and sequences enables the design of amino acid sequences that fold into a specific structure [26] , [27] . Using these designed peptide sequences it is possible to describe the folding mechanism of highly specific folding [26] , [27] or binding [16] , [28] . However , due to its coarse-grained nature , the model would be unsuited to represent the structure or binding site of a specific , naturally occurring protein . We use this coarse-grained model to investigate how the binding free energy of a short binding motif depends upon its structural environment: we simulate binding to a substrate for a flexible binding motif , a flexible motif embedded in an unstructured chain and a rigid binding motif embedded in a rigid structure ( see Figure S1 ) . The model of the substrate and binding region embedded in disordered flanks have been designed to contain the general features associated with disordered regions and natively unstructured binding regions , viz . an extended binding conformation , a large binding surface , hydrophobicity of the binding region and hydrophilic flanks . We find that the binding motif embedded in a rigid structure unbinds at higher temperatures than either the flexible binding motif or the binding motif in a longer disordered region . The latter two binding free energies are very similar over the range of temperatures simulated . However , we show that even at low concentrations the ( hydrophobic ) binding motif aggregates with itself , and that the ( hydrophilic ) disordered flanks prevent such aggregation at temperatures relevant for reversible binding .
To investigate how the binding free energy of a short binding motif depends upon its structural environment , a binding motif was designed to specifically bind in a groove of a rigid substrate ( Figure 1 ) . The amino acid sequence ( Arg , Trp , Tr , Leu , Tyr ) of this motif is predominantly hydrophobic , but contains a single charged amino acid . In our coarse-grained model , neighbouring hydrophobic residues attract each other , whereas amino acids of the same charge repel each other . The binding of this binding motif was simulated embedded in three different structures: as a single flexible binding motif ( BM ) , as a single flexible binding motif with disordered flanks of 15 Threonine residues on each side ( BM disorder ) and embedded in a rigid structure of Threonine residues ( BM rigid ) , see Figures S1 and S2 . Threonine is a hydrophilic amino acid . In our model contacts involving Threonine do not contribute to the internal energy of the configuration so that the internal energy of the binding motif bound to the substrate is the same for all three structures ( see Methods ) . The binding and unbinding process was simulated at different temperatures , while the concentration of the substrate and peptide are kept constant . Figure 2 shows that at low temperatures ( T<0 . 25 ) the average degree of binding ( 〈Pb〉 ) is high , i . e . the binding motif is nearly always bound to the substrate , and at high temperatures ( T>0 . 45 ) the average degree of binding is low . The flexible peptides ( BM and BM disorder ) are unstructured in the unbound state ( see Figure S2 ) . There is a transition between the bound and unbound state at which reversible binding is possible . This transition can also be observed by the peak in the heat capacity ( Cv ) . Similar peaks in heat capacity are found at folding transitions of both simulated and real proteins ( e . g . , [29] , [30] ) . The sharpness of the heat-capacity curve also indicates that the binding motif binds with high specificity to the substrate . Binding of an aspecific motif to the substrate would result in a much broader heat-capacity peak . In nature binding motifs typically have a signalling function , implying that the peptide should be able to bind as well as unbind in the relevant temperature range . Figure 2 shows that the binding motif binds reversibly to the substrate for approximately 0 . 2<T<0 . 3 . Interestingly , Figure 2 shows that the disordered flanks have little effect on the binding free energy: the average amount of binding and heat capacity are similar over the entire temperature range for both flexible peptides ( BM and BM disorder ) . Additional simulations showed that even with a much larger substrate the difference in binding free energy between the binding motif and the motif embedded in disordered flanks remains small . However , as previously reported [16] , the flexibility of the binding motif itself lowers the difference in free energy between the bound and unbound state , since conformational entropy is lost upon binding to the substrate . Figure 2 shows that the temperature range for reversible binding of flexible peptide chains is lower than for a rigid binding motif . Even though disordered flanks appear to contribute little to the binding free energy , the collective contribution of many such flanks may be important . We simulated 10 binding motifs without the substrate to investigate the collective behaviour of the peptides . Figure 3 shows that 10 binding motifs without flanks tend to aggregate whereas those with flanks do not at a temperature at which reversible binding is possible; the lowest free energy configuration for 10 binding motifs with flanks is as free chains or in very small clusters , whereas the binding motifs without flanks make many more external contacts . To investigate this phenomenon for a larger number of peptide chains , we simulated aggregation behaviour of the two types of binding motifs with a Grand Canonical Monte Carlo simulation , while keeping the free binding motifs at low concentration ( see Methods ) . First , simulations starting from a single chain in the simulation box were performed at different temperatures . Many more external contacts form for the binding motif than for the binding motif embedded in disordered flanks ( Figure 4 ) . Moreover , the aggregates form at higher temperatures for binding motifs without disordered flanks . From these simulations we selected aggregates of different cluster sizes . Each cluster of aggregates was simulated at different temperatures to determine the transition temperature , Ts , at which the aggregate would shrink rather than grow in size ( Figure 5 ) . Comparing Figure 2 with Figure 5 it can be observed that the binding motifs ( BM ) are in an aggregated state at temperatures within the reversible binding regime , whereas the binding motifs with disordered ( BM disorder ) are fully dissolved . Figure 2 also shows that with increasing aggregate size the aggregates formed by binding motifs without disordered flanks become more difficult to melt , indicating that once an aggregate is formed it will be difficult to dissolve . Binding motifs embedded in disordered domains , generally form micelle-like structures that do not grow larger than approximately 12 chains ( see Figure 4 ) . Decreasing the length of the disordered flanks , down to 5 residues on each side of the binding motif , does not have a strong effect on the melting temperatures . In that case the micelles formed are somewhat larger . The system also shows considerable hysteresis: the aggregated clusters melt at much higher temperatures than the ones at which they formed . Again , this effect is much smaller for binding motifs embedded in disordered flanks .
Our simulations suggest that the primary role of disordered flanks adjacent to small peptide binding motifs is to suppress aggregation in solution rather than to modify the binding strength to the substrate . This observation provides a rationale for the experimental observation that linear binding motifs are often found in disordered parts of a peptide chain [23] . In this work only a small difference in binding strength between binding motifs with and without disordered flanks is found . The model used here is based on the assumption that interactions between the disordered flanks and the substrate are of a steric nature . However our results do not preclude the possibility that the binding strength changes significantly if the disordered flanks have additional interactions with the substrate , for example through charged residues or a second binding motif . Our work focuses on the physical effect of disordered flanks that have no specific interaction with the substrate . The isolated binding motifs described in the present paper would aggregate due to hydrophobic interactions . We suggest that such motifs , without hydrophilic flanks , are toxic . There is indeed increasing evidence that hydrophobic aggregation is correlated with toxicity for the cell [31] . Of course , the model calculations that we present here are highly simplified . The degree of hydrophobicity in real binding motifs varies , although it is typically higher than that of disordered proteins or that of the surface of globular proteins . There is , therefore , a great need for experiments to quantify the difference in aggregation behavior of signalling peptides with and without disordered flanks . Aggregated proteins can form different structures: ordered beta sheet fibers ( amyloids ) or non-specific hydrophobic aggregates . Human diseases , such as Alzheimer and Parkinson disease , are mostly associated with the former . The work presented here is most closely related to the latter mechanism . Nevertheless , there is increasing evidence that the two mechanisms are connected and that hydrophobic pre-fibrillar aggregates may be causing the toxicity in amyloid forming proteins [32] , [33] . Insights in ( the prevention of ) protein hydrophobic aggregation may therefore be important for further understanding of both aggregation types . Of course , there could be other ways to suppress hydrophobic aggregation . For instance , aggregation would be strongly inhibited if the binding motif were embedded in a rigid structure [34] . However , a flexible binding motif has the advantage that it can combine the ability to bind reversibly with high specificity: this feature is important for regulatory motifs . As such , it would not be surprising to find that disordered flanks have evolved to suppress aggregation . There are several other biological examples of evolutionary pressure against aggregation [34] . For example: there exist very few proteins with beta-strands on the edge of protein structures–a feature that might induce amyloid formation by edge-to-edge aggregation of beta-sheets [35] . Another example is the ‘end-capping’ of sequence regions in globular proteins that would otherwise exhibit a high amyloid-forming propensity by charged or structure-disrupting residues [36] . The stabilising effect of disordered flanks is closely related to steric stabilisation of colloids by polymers . Indeed , steric stabilization has been exploited extensively in material and drug design to stop colloids aggregating [37] or to increase the lifetime of hydrophobic drugs by attaching the drug to block copolymers with a hydrophobic middle and hydrophilic flanks [38] . The latter experiments show that steric stabilisation of hydrophobic moieties is highly relevant in biological systems but , as is often the case , evolution “discovered” this effect first . The present work provides a testable hypothesis for the abundance of disordered regions in proteins: it suggests that disordered flanks adjacent to hydrophobic motifs can suppress aggregation of the hydrophobic peptides in solution . The hypothesis that we put forward gives a basis for in vitro or in vivo experiments into the effect of hydrophilic disordered flanks on the aggregation , solvability and toxicity of hydrophobic peptides . Confirmation of our predictions in a biological context may lead to new methods that could increase the bioavailability of hydrophobic peptides .
We use a coarse grained representation of a peptide chain where each residue occupies a single point on a cubic lattice [26] . Neighboring residues that would be covalently bound in a peptide chain are required to be on neighbouring lattice sites ( Figure 1 ) . Residues interact when residing on neighbouring sites . The internal energy of a configuration is given by: ( 1 ) where A ( i ) gives the amino acid at residue i , Ci , j = 1 when residues i and j interact and Ci , j = 0 otherwise . The interaction matrix M gives the pairwise interactions between all 20 amino acids and is based on the occurrence of amino acids in close proximity in experimentally determined protein structures [24] , [25] . The interaction matrix is normalised with respect to Threonine [25] , so that all pairwise interaction energies of Threonine are set to zero . We use this in our simulations to observe the purely entropic contributions of the disordered flanks . The interaction matrix used here is based on structural proteins , while pairwise interactions in unstructured regions may have slightly different propensities . One may expect that hydrophobic residues in unstructured peptide sequence may be some what less hydrophobic due to the exposed backbone . In this case it may be that the number of hydrophobic residues needed for peptide aggregation is slightly higher than in the current work , but we expect that the qualitative effects of the aggregation remain similar . We use a Monte Carlo simulation technique where trial steps are accepted according to: ( 2 ) where T is the simulation temperature , kb is the Boltzmann constant and −ΔE is the difference in energy between the new and old configuration of the model . Trial moves are either internal moves , changing the configuration of a chain ( end move , corner flip , crank shaft , point rotation ) , or rigid body moves , changing the position of the chain relative to other objects ( rotation , translation ) , see ref . [27] for more details . At each iteration a single local trial move is performed and a global trial move move ( including point rotations ) is performed with the probability ( Pglobal = 0 . 1 ) . In the binding simulations , only rigid body moves are applied to ‘rigid’ binding motifs , whereas the configurations of the flexible binding motifs are sampled with both internal and rigid body moves . The volume of the simulation box ( 60×60×60 lattice points ) was kept constant , yielding a concentration for the peptide that is higher than that typical of signalling peptides in a cell ( approximately 10–1000 times higher ) . However , the cytosol will contain other signalling peptides that , if not properly protected , could participate in aggregation . Moreover , as argued in the Supplementary Material ( Text S1 ) , the peptide solutions in our model are still sufficiently dilute to make it possible to extrapolate our findings to the typical concentrations that prevail inside a cell . Parallel tempering , or temperature replica exchange , was used to converge more rapidly to sampling of equilibrium configurations . Multiple simulations at different temperatures were run in parallel , while trying to swap temperatures every 50000 moves with 10000 trial temperatures swaps in each simulation . A trial swap between the temperatures of two replicas was accepted with a probability [39]–[41]: ( 3 ) The design of binding interface ( i . e . the contacts between the binding motif and the binding groove ) was achieved through a Monte Carlo algorithm that interchanges amino acids , while optimising the total energy of the bound state and keeping the variance of the amino acids high , see [27] , [28] for more details . In order to estimate the probability distribution P ( x ) ( where x is an “order parameter” , such as Cext , the number of external contacts ) , we use both configurations of accepted and rejected trial moves weighted by the Boltzman factors of each configuration [42] . The amount of binding of the binding motif to the substrate is tracked by comparing the number of ( non-covalent ) contacts Ci , j in a configuration to the contacts present in the fully bound state . Then the total number of native binding contacts is defined as: ( 4 ) where N is the total number of residues in the binding motif ( excluding the flanking regions ) . Tracking aggregation of multiple binding motifs is done by considering the total number of external contacts Cext: ( 5 ) where M is the total number of chains in the simulation box and is a contact between residue i in chain k and residue j in chain l . Note that Threonine-Threonine contacts do not contribute to Cext . The amount of binding is given by: ( 6 ) The constant volume heat capacity is calculated as: ( 7 ) Ensemble averages for an order parameter x are given by: ( 8 ) where P ( x ) is estimated as before . A grand canonical Monte Carlo simulation was performed to investigate the aggregation behaviour of binding motifs at a constant ( low ) concentration of these peptides . Trial insertions and deletions were performed with a probability of Pinsert = Pdelete = 0 . 005 per move . Trial insertion of new chains ( with an identical sequence ) were accepted with: ( 9 ) and deleted with: ( 10 ) where , N is the number of free chains in the simulation box before the move , V is the volume of the box , and μ the chemical potential . The volume was kept constant at 30×30×30 lattice points and exp ( μβ ) was kept constant at 3·10−6 in all simulations . A single peptide chain was simulated in a separate box , at the same temperature , to generate new configurations for insertion into the main simulation box . Only free chains were inserted and removed , i . e . no chains that make an external contact with another chain . Since the chains were simulated at very low density , moves are likely that remove the only peptide chain from the simulation box . At such an event the number of trial insertion moves ( Mi ) to re-entrance was taken as: ( 11 ) where U is a random , uniformly distributed variable on the interval [0 , 1] . The total number of sampling steps is given by the total number of trial moves ( S ) : ( 12 ) The order parameters and internal energy are all zero for the empty simulation box . Images in Figures 1 and 4 were produced using the UCSF Chimera package [43] . | In their natural cellular environment proteins are dissolved in a concentrated aqueous solution of biomolecules . Even under such crowded conditions , proteins must not clump together or aggregate; otherwise their biological functions may be compromised , and the cell could die . Diseases such as Parkinson and Alzheimer are thought to be caused by aggregation of specific proteins . Evolutionary pressure generally ensures that proteins do not aggregate in their natural biochemical environment . A well-known mechanism to prevent aggregation is the folding of proteins , where the hydrophobic ( attractive ) part of the protein is buried inside the protein . Here we report a different mechanism that can prevent the aggregation of proteins . Recently , it was discovered that many proteins contain regions that are disordered ( not folded ) in their natural environment . We show with coarse-grained simulations that aggregation of small hydrophobic binding motifs can be prevented by embedding the motifs in disordered regions: the disordered regions of different proteins obstruct or sterically hinder the formation of aggregates . Moreover , our simulations show that the disordered regions have no adverse effect on the biological function of the binding motifs , because they do not obstruct the binding and folding of the binding motif on its specific substrate . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biophysics/protein",
"folding",
"biophysics/theory",
"and",
"simulation",
"biophysics/biomacromolecule-ligand",
"interactions",
"computational",
"biology",
"chemical",
"biology/protein",
"chemistry",
"and",
"proteomics"
] | 2008 | Disordered Flanks Prevent Peptide Aggregation |
Exosomes are secreted organelles that have the same topology as the cell and bud outward ( outward is defined as away from the cytoplasm ) from endosome membranes or endosome-like domains of plasma membrane . Here we describe an exosomal protein-sorting pathway in Jurkat T cells that selects cargo proteins on the basis of both higher-order oligomerization ( the oligomerization of oligomers ) and plasma membrane association , acts on proteins seemingly without regard to their function , sequence , topology , or mechanism of membrane association , and appears to operate independently of class E vacuolar protein-sorting ( VPS ) function . We also show that higher-order oligomerization is sufficient to target plasma membrane proteins to HIV virus–like particles , that diverse Gag proteins possess exosomal-sorting information , and that higher-order oligomerization is a primary determinant of HIV Gag budding/exosomal sorting . In addition , we provide evidence that both the HIV late domain and class E VPS function promote HIV budding by unexpectedly complex , seemingly indirect mechanisms . These results support the hypothesis that HIV and other retroviruses are generated by a normal , nonviral pathway of exosome biogenesis .
Exosomes are secreted organelles that have the same topology as the cell and a diameter of approximately 50–150 nm , though larger exosomes have also been reported [1–3] . Exosome content varies considerably , but commonly includes tetraspanins , integrins , major histocompatibility complex proteins , cytosolic chaperones , cholesterol , and glycosphingolipids [2 , 4] . Many animal cell types secrete exosomes , and exosome-mediated signaling has been implicated in antigen presentation , morphogenesis , sperm maturation , and cancer–host interactions [2 , 4–6] . Less is known about the mechanisms of exosome biogenesis . Cells can generate exosomes by either of two modes , immediate or delayed . The immediate mode of exosome biogenesis occurs at the cell surface and involves outward vesicle budding ( outward is defined as away from the cytoplasm ) from endosome-like domains of the plasma membrane , domains we refer to as ELDs [3] . In contrast , the delayed mode of exosome biogenesis begins with outward vesicle budding at the limiting membrane of endosomes , generating vesicle-laden endosomes , typically referred to as multivesicular bodies ( MVBs ) [2 , 4] . If an MVB fuses with the plasma membrane , its internal vesicles are released as exosomes . However , MVBs can also fuse with lysosomes , leading to vesicle and cargo destruction [7] . Some cell types , such as T cells , prefer to make exosomes by the immediate mode , whereas other cell types , such as macrophages , prefer to make exosomes via the delayed mode . Interestingly , human immunodeficiency virus ( HIV ) particles bud from these two cell types at the same sites as exosomes , have the same topology as exosomes , have a similar size as exosomes , and are enriched in the same molecules as exosomes [3 , 8–13] . These and other observations indicate that there might be a mechanistic relationship between retrovirus budding and exosome biogenesis [11] . A major step in understanding the biogenesis of any organelle is to identify and characterize the cis-acting signals that target proteins to the organelle . Here we show that higher-order oligomerization and plasma membrane association target proteins to sites of exosome budding and into exosomes . Proteins can be directed into this exosomal protein sorting pathway by ( 1 ) exposing cell surface proteins to exogenous cross-linking agents ( e . g . , primary and secondary antibodies ) , ( 2 ) appending plasma membrane anchors to highly oligomeric cytoplasmic proteins , or ( 3 ) adding multiple homo-oligomerization domains to intracellular acylated proteins . The class E vacuolar protein-sorting ( VPS ) proteins are thought to drive outward vesicle budding [14] , but we find that inhibiting class E VPS function does not block exosome budding or the oligomerization-induced exosomal protein-sorting pathway . We also find that retroviral Gag proteins are sorted to ELDs and exosomes , that exosomal targeting information directs proteins to sites of HIV Gag budding and onto HIV virus–like particles ( VLPs ) , that higher-order oligomerization is a primary determinant of HIV Gag budding , that p6-deficient HIV can bud independently of class E VPS function , and that acquisition of exosomal sorting information is sufficient to induce the budding of a yeast long terminal repeat ( LTR ) retrotransposon . Taken together , these results support the hypothesis that retroviral budding is a form of exosome biogenesis [11] .
Jurkat T cells bud exosomes from ELDs [3] . ELDs are often clustered at a single pole in Jurkat T cells [3] , resembling the surface protein “caps” that form after antibody-induced oligomerization of leukocyte plasma membrane proteins [15 , 16] . To determine whether there is any relationship between exosomal protein sorting and surface protein capping , we marked ELDs by pulse-labeling Jurkat T cells with an exosomal lipid , N-Rh-PE [3] , chilled the cells to 4 °C , then incubated the cells with monoclonal ( bivalent ) immunogammaglobulin ( IgG ) antibodies specific for known T cell plasma membrane proteins , washed the cells , and incubated the cells with fluorescein isothiocyanate ( FITC ) -labeled polyclonal anti-mouse IgG antibodies , all on ice . Half the cells were fixed immediately , half were incubated at 37 °C for 1–2 h and then fixed , and both were examined by fluorescence microscopy ( Figure 1A–1X ) . At time 0 , the plasma membrane markers CD43 , CD45 , and CD59 showed no significant enrichment at ELDs . In contrast , cells incubated at 37 °C sorted these antibody–antigen complexes to ELDs . Similar results were observed for CD4 , CD5 , CD28 , CD31 , CD38 , CD55 , CD62L , CD98 , CD99 , and PrP ( unpublished data ) , as well as in cells expressing a different marker for ELDs , AIP1/VPS31-DsRED ( unpublished data ) [3] . Surface protein “capping” has been reported to require the oligomerization of oligomers [15 , 16] , which we refer to as higher-order oligomerization . To determine whether higher-order oligomerization was sufficient for targeting CD43 to ELDs and exosomes , N-Rh-PE–labeled Jurkat T cells were chilled to 4 °C , incubated with FITC-labeled monoclonal ( IgG ) anti-CD43 antibodies ( on ice ) , separated into two equal fractions , and incubated with either a mock solution or unlabeled polyclonal rabbit anti-mouse IgG antibodies , all on ice , followed by incubation overnight at 37 °C . Fluorescence microscopy of the two cell populations revealed that CD43 was sorted to ELDs only after the addition of polyclonal secondary antibodies ( Figure 1Y–1FF ) . Thus , higher-order oligomerization is both necessary and sufficient to target CD43 to ELDs . ELDs are sites of exosome budding , and we next tested whether higher-order oligomerization is sufficient to induce a protein's budding/exosomal sorting ( we define the term “budding” to mean the secretion of a molecule on sedimentable vesicles that have the general properties of exosomes/VLPs ) . N-Rh-PE–labeled T cells that had been exposed to only monoclonal anti-CD43 antibodies secreted N-Rh-PE–labeled exosomes that were mostly lacking the FITC-labeled CD43–antibody complex . In contrast , exosomes secreted by cells exposed to both the FITC-labeled monoclonal anti-CD43 antibody and the polyclonal anti-mouse IgG antibodies were approximately 30-fold more likely to carry detectable levels of the FITC-labeled CD43–antibody complex ( Figure 1GG–1KK ) . In paralogous experiments using unlabeled antibodies , immunoblot analysis of cell and exosome lysates revealed that higher-order oligomerization induces an approximately 10-fold increase in the amount of CD43–antibody complex secreted from the cell in exosomes ( Figure 1LL ) . This increase in exosomal CD43 did not appear to reflect a general increase in exosome biogenesis , because the levels of exosomal Lamp1 were unaffected by these manipulations . The hypothesis that higher-order oligomerization is sufficient to target plasma membrane proteins to ELDs and exosomes has a clear corollary: a plasma membrane anchor should be sufficient to target highly oligomeric , cytoplasmic proteins to ELDs and exosomes . To explore this possibility , we used the yeast protein TyA , which assembles into large oligomeric structures in the cytoplasm of yeast cells [17 , 18] . The suitability of TyA for these studies is enhanced by the fact that it is derived from the Saccharomyces cerevisiae Ty1 LTR retrotransposon , which accumulates in the cytoplasm , does not bud from cells , is not infectious , and replicates in an organism that is devoid of retroviruses . When expressed in Jurkat T cells , wild-type ( WT ) TyA ( tagged with green fluorescent protein [GFP] at its C-terminus ) accumulated in the cytoplasm and showed no enrichment at ELDs ( Figure 2A–2D ) . In contrast , AcylTyA-GFP , which contains a 10 amino acid–long acylation tag at its N-terminus ( designed to confer myristoylation at Gly2 , palmitoylation at Cys3 , and targeting of the protein to the plasma membrane [19] ) , co-localized at ELDs with the exosomal markers N-Rh-PE and surface CD63 ( Figure 2E–2L ) . A mutant form of AcylTyA-GFP lacking the putative acylation sites , Acyl ( G2A , C3A ) TyA-GFP ( containing alanine residues at positions 2 and 3 of the tag ) , was not sorted to ELDs and instead accumulated in the cytoplasm of Jurkat T cells ( Figure 2M–2P ) . To determine whether the sorting to ELDs correlated with secretion in exosomes , we collected exosomes from N-Rh-PE–labeled Jurkat T cells expressing either TyA-GFP or AcylTyA-GFP , bound them to glass , and examined them by fluorescence microscopy . Jurkat T cells failed to secrete TyA-GFP from the cell in exosomes ( Figure 2Q and 2R ) , but did secrete AcylTyA-GFP from the cell in N-Rh-PE–containing exosomes ( Figure 2S and 2T ) . In a separate experiment , we generated cell and exosome lysates from Jurkat T cells either mock-transfected or transfected with plasmids designed to express HIV Gag-GFP ( which is sufficient for budding [20] and buds from the cell in exosomes [3] ) , TyA-GFP , AcylTyA-GFP , and Acyl ( G2A , C3A ) TyA-GFP , and then subjected these to immunoblot analysis . Using a constant ratio of cell lysate:exosome lysate , we observed that AcylTyA-GFP was selectively secreted from the cell in exosomes ( Figure 2U ) , as was HIV Gag-GFP . We further purified exosomes from these cells by sucrose density flotation gradient centrifugation and assayed fractions across the gradient by immunoblot using antibodies specific for GFP and for CD63 , a known exosomal marker . AcylTyA-GFP co-fractionated with CD63 , providing further evidence that it was secreted from the cell in exosomes ( Figure 2V ) . Exosomes collected from these cells were also subjected to protease protection experiments . AcylTyA-GFP was degraded far more extensively in the presence of trypsin and Triton X-100 than when exposed to trypsin alone ( Figure 2W ) , indicating that AcylTyA-GFP was located in the lumen of the exosomes . Electron microscopy experiments provided additional evidence that the acylation tag is sufficient to induce the budding of TyA . TyA is known to form electron-dense protein complexes [17 , 18] , and thus , cells expressing an exosomal form of TyA would be expected to secrete exosomes that contain an electron-dense lamina under their membrane . Jurkat T cells normally secrete exosomes that lack an electron-dense lamina under their membrane [3] , and this was observed for exosomes secreted by cells expressing unmodified , non-exosomal TyA ( Figure 2X and 2Y ) . In contrast , cells expressing AcylTyA secreted exosomes that resembled retroviral VLPs in that they had an electron-dense lamina under their membrane ( Figure 2Z–2GG ) . These AcylTyA-containing exosomes varied significantly in size , from approximately 50-nm diameter to approximately 250-nm diameter , and were typically of spheroid morphology , though some possessed short membrane protrusions ( Figure 2CC–2EE ) . AcylTyA-containing exosomes also labeled for N-Rh-PE , shown here using 6-nm immunogold ( Figure 2FF and 2GG ) . Like the exosomal cargoes described above , the Gag proteins of the Orthoretroviridae ( true retroviruses ) are known to bind the plasma membrane and assemble into higher-order oligomeric complexes [21] . Therefore , if higher-order oligomerization and plasma membrane binding are sufficient for exosomal targeting , orthoretroviral Gag proteins should be sorted to ELDs and exosomes . Equine infectious anemia virus ( EIAV ) , human T-cell lymphotropic virus-1 ( HTLV-1 ) , murine leukemia virus ( MLV ) , Rous sarcoma virus ( RSV ) , Mason-Pfizer monkey virus ( MPMV ) , and human endogenous retrovirus-K ( HERV-K ) represent five major families of the Orthoretroviridae . The Gag protein from each of these viruses was expressed in Jurkat T cells as a C-terminally GFP-tagged protein , and in each case , the Gag-GFP protein was sorted to ELDs . This is shown here by their co-localization with surface CD63 , one exosomal marker ( Figure 3 ) , as well as by their co-localization with the ELD/exosome markers N-Rh-PE and AIP1-DsRED ( unpublished data ) . As a negative control , we followed the sorting of the Gag protein from simian foamy virus ( SFV ) , a representative of the Spumaviridae . The Spumaviridae are the viruses most closely related to the Orthoretroviridae , but SFV budding is mediated by its envelope protein rather than its Gag protein [21] . SFV Gag-GFP was not sorted to ELDs ( Figure 3CC–FF ) . To determine whether these Gag proteins were secreted from the cell in exosomes , exosomes were collected from Jurkat T cells that had been pulse labeled with N-Rh-PE and transfected with expression vectors for each Gag protein . Two days later , exosomes were collected , bound to glass , and visualized by fluorescence microscopy . Mock-transfected cells secreted exosomes that lacked GFP , as expected ( Figure 4A and 4B ) . In contrast , Jurkat T cells secreted each of the orthoretroviral Gag proteins in discrete particles , nearly all of which were also labeled with the exosomal lipid N-Rh-PE ( Figure 4C–4N ) . SFV Gag-GFP could not be detected in exosomes ( Figure 4O and 4P ) . Immunoblot analysis confirmed these observations ( Figure 4Q ) . If retroviral budding is a form of exosome biogenesis , then higher-order oligomerization and plasma membrane binding should target proteins to HIV VLPs ( the Gag-containing vesicles secreted by Gag-expressing cells ) . To test this prediction , we expressed HIV Gag-DsRED in Jurkat T cells and exposed these cells to FITC-labeled monoclonal anti-CD43 antibodies . The sample was then split in half , incubated with either mock solution or polyclonal anti-mouse IgG antibodies , incubated overnight at 37 °C , and examined by fluorescence microscopy . For cells exposed to only primary anti-CD43 IgG , the FITC-labeled antibody–CD43 complex showed little if any co-localization with HIV Gag-DsRED at ELDs ( Figure 5A–5D ) . In contrast , the higher-order oligomerization of CD43 induced by adding polyclonal anti-mouse IgG antibodies caused the co-localization of CD43–antibody complexes with HIV Gag-DsRED at ELDs ( Figure 5E–5H ) . Moreover , the Gag-DsRED-expressing cells that were exposed to both the FITC-labeled anti-CD43 IgG and polyclonal anti-mouse IgG antibodies secreted significant numbers of Gag-DsRED-containing VLPs that also carried FITC-labeled CD43–antibody complexes ( Figure 5I–5K ) . To test these conclusions in greater detail , we performed similar experiments on N-Rh-PE–labeled Jurkat T cells that expressed untagged , full-length HIV Gag . Using 6-nm immunogold to detect N-Rh-PE , 18-nm immunogold to detect CD43 , and the electron-dense Gag core to detect sites of VLP budding and VLPs , we quantified the amount of CD43 present on budding VLPs following CD43 dimerization ( monoclonal antibodies only ) or higher-order oligomerization . In cells exposed to only primary anti-CD43 antibodies , the levels of CD43–antibody complexes on emerging HIV VLPs were relatively low ( Figure 5L; CD43-gold detected on 7/110 Gag-containing vesicles , with an average of 0 . 10 ± 0 . 04 gold grains/Gag-containing vesicle ( ± standard error ) ) . In contrast , higher-order oligomerization of CD43 caused a nearly 10-fold increase in the amount of CD43–antibody complexes on budding HIV VLPs ( Figure 5M; CD43-gold detected on 70/119 Gag-containing vesicles , with an average of 0 . 94 ± 0 . 09 gold grains/Gag-containing vesicle ) . The greater labeling for N-Rh-PE than CD43 is likely due to the generally better labeling for small immunogold conjugates as well as the addition of a significant quantity of N-Rh-PE to the cells . The oligomerization-induced sorting of plasma membrane proteins to ELDs , exosomes , and VLPs led us to examine the sorting information in HIV Gag , the key budding factor in HIV . HIV Gag clearly has the expected properties of an exosomal cargo in that it assembles into highly oligomeric core particles of greater than 100 mDa ( up to 5 , 000 polypeptides/core ) , is anchored in the inner leaflet of the plasma membrane via an N-terminal myristoyl moiety [22–24] , and is sorted to ELDs and exosomes by Jurkat T cells [3] . Moreover , it is known that the budding of HIV Gag and virus is blocked by mutations that prevent its anchoring in the plasma membrane [24] or disrupt either of its two major oligomerization domains [25–30] , which are located in its capsid ( CA ) and nucleocapsid ( NC ) domains , respectively ( Figure 6A ) . On the other hand , the prevailing hypothesis is that Gag and virus budding is driven by ( 1 ) the actions of its late domain , the PTAP sequence located in the Gag p6 domain ( Figure 6A ) , and ( 2 ) the specific architecture of the homo-oligomeric Gag core complex , which is thought to promote budding by deforming the cell membrane [31] . This hypothesis is supported by several lines of evidence , most notably ( 1 ) reduced budding of HIV late domain mutants ( in certain cell types ) , ( 2 ) physical interaction between the late domain motif ( and other p6 sequences ) with certain class E VPS proteins , and ( 3 ) reduced HIV budding in cells with impaired class E VPS function [14 , 31–37] . To identify the budding/exosomal sorting information in HIV Gag , we followed the sorting and secretion of full-length and mutant Gag proteins by Jurkat T cells ( human T cells are a primary in vivo host for HIV ) . HIV Gag-GFP , which buds from cells in a manner similar to that of WT Gag [20] , was sorted to ELDs and exosomes ( Figure 6B–6I , 6R , 6S , and 6V ) , as we reported previously [3] . HIV Gag ( p49 ) -GFP lacks the entire p6 domain , including the late domain PTAP motif . HIV Gag ( p49 ) -GFP was also sorted to ELDs ( Figure 6J–6Q ) and secreted from the cell in exosomes ( Figure 6T–6V ) , and at levels similar to that of full-length HIV Gag-GFP . Thus , neither the HIV late domain nor the entire p6 domain were required for HIV Gag budding/exosomal sorting by Jurkat T cells . We next tested whether higher-order oligomerization was a primary determinant of HIV Gag budding . HIV Gag contains two major oligomerization domains , one in CA and one in NC , and it is already known that loss of either NC or CA blocks Gag budding [25–30] . HIV Gag ( p41 ) -GFP lacks all NC , p1 , and p6 sequences , and thus , lacks the interaction domain of NC but still retains one major oligomerization domain , in the C-terminal half of CA ( Figure 7A ) . HIV Gag ( p41 ) -GFP was not sorted to ELDs ( Figure 7B–7E ) and was not secreted from the cell in exosomes ( Figure 7N , lane 2 ) . However , HIV Gag ( p41 ) -GFP could be redirected to ELDs by addition of a heterologous dimerization domain ( a synthetic leucine zipper [LZ] ) : HIV Gag ( p41 ) -LZ-GFP , was sorted to ELDs and secreted from the cell in exosomes ( Figure 7F–7I and 7N–7P ) as well as HIV Gag-GFP . Thus , the primary budding/exosomal sorting signal in the NC-p1-p6 region of HIV Gag appears to be the oligomerization domain in NC ( also known as the I domain [20] ) , not the HIV late domain motif , at least in Jurkat T cells . This conclusion is supported by the inability of the p6 domain to rescue the budding/exosomal sorting defects caused by loss of NC-p1-p6 ( Figure 7J–7N ) . One concern in these experiments is that p6-independent budding of HIV Gag might be an experimental artifact caused by overexpression of HIV Gag . This does not appear to be the case , because control experiments demonstrated that the expression system that we used in our experiments actually drives lower levels of Gag expression in Jurkat T cells than an HIV provirus ( Figure 7Q ) . We next followed the sorting and secretion of a Gag protein that lacks both of its oligomerization domains . HIV Gag ( p39* ) is unable to oligomerize or bud from cells , has a point mutation in the CA oligomerization domain ( W184A ) , and lacks all p2 , NC , p1 , and p6 sequences [26] . As expected , HIV Gag ( p39* ) -GFP was neither enriched at ELDs nor secreted from the cell in exosomes ( Figure 8B–8E and 8V , lane 2 ) . Adding just one oligomerization domain to this protein failed to rescue its exosomal sorting , because neither HIV Gag ( p39* ) -LZ-GFP nor HIV Gag ( p39* ) -LZ-DsREDmonomer was sorted to ELDs or exosomes ( Figure 8F–8M and 8V , lanes 3 and 4 ) . However , addition of two independent oligomerization domains to HIV Gag ( p39* ) was sufficient to target HIV Gag ( p39* ) to ELDs and exosomes . Specifically , we observed that HIV Gag ( p39* ) -LZ-DsRED , which contains both an oligomeric form of DsRED [38 , 39] and the dimer-inducing leucine zipper , was sorted to ELDs ( Figure 8N–8Q ) and secreted from the cell in exosomes ( Figure 8V , lane 5 , 8W , and 8X ) to a similar extent as HIV Gag-DsRED ( Figure 8R–8U and 8V ) . Native gel electrophoresis confirmed that HIV Gag ( p39* ) -LZ-DsRED exists in a higher oligomeric state than Gag ( p39* ) -LZ-DsREDm ( Figure 8Y ) . The simplest conclusion from these results is that higher-order oligomerization , rather than any specific sequences in HIV Gag , is a primary determinant of HIV Gag budding/exosomal sorting . Such a conclusion , however , fails to account for the fact that other sequences in HIV Gag ( p39* ) might contribute to Gag budding , such as those that bind AP-3 [40] , perhaps phosphatidylinositol-4 , 5-bisphosphate [41] , etc . To address the possibility that these provide a unique and essential foundation for HIV Gag budding/exosomal sorting , we removed all Gag sequences from HIV Gag ( p39* ) -LZ-DsREDmonomer and HIV Gag ( p39* ) -LZ-DsRED , and replaced them with a plasma membrane anchor ( NH2-MGCINSKRKD-COOH [19] ) . The first of these proteins , Acyl-LZ-DsREDmonomer , showed little or no enrichment at ELDs and was a poor exosomal cargo ( Figure 9A–9D , 9I , 9J , and 9M ) . In contrast , Acyl-LZ-DsRED was efficiently secreted from the cell in exosomes ( Figure 9E–9H , 9K , 9L , and 9M ) . These results demonstrate two important points: first , that higher-order oligomerization and plasma membrane binding are sufficient to target a protein to exosomes , and second , that it is possible to generate a synthetic exosomal cargo based on this principle . The p6-independent , oligomerization-induced budding/exosomal secretion of HIV Gag is , from a virological perspective , most important if it also informs our understanding of HIV virus budding . To address this issue , we followed the budding of control and p6-deficient HIV viruses from human cells . It is well established that loss of p6 severely reduces HIV budding from human kidney-derived 293T cells [32 , 35 , 36] , and we therefore used these cells to assess the role of HIV p6 in HIV virus budding . 293T cells were transfected with a control HIV provirus , NL4 . 3ΔEnv::GFPkdel ( an Env-deleted , but otherwise replication-competent , variant of the HIV provirus NL4 . 3 [42] ) , and these cells budded significant levels of virus into the medium ( Figure 10A , left lanes ) . As expected from the literature , 293T cells transfected with a p6-deficient provirus , NL4 . 3ΔEnv::GFPkdel/p6L1ter , budded far less virus ( Figure 10A , middle lanes ) . ( The p6L1ter mutant selectively abrogates p6 expression [35] . ) At first glance , these results seem to indicate that the p6-independent budding of HIV Gag might not be very informative for the mechanism of HIV virus budding . However , the HIV provirus expresses several additional proteins , and these have the potential to influence the exosomal sorting of HIV Gag . In particular , HIV protease ( PR ) is a known inhibitor of HIV budding [43] and a predicted destroyer of exosomal sorting information in HIV Gag ( PR cleaves Gag between its matrix [MA] , CA , and NC domains [22] ) . Moreover , Huang et al [44] reported more than a decade ago that loss of the HIV late domain did not impair the budding of a PR-deficient HIV mutant . To test whether the budding defect of HIV late domain mutants is an indirect effect and caused primarily by HIV PR activity , we followed the budding of a p6-deficient , PR-deficient HIV provirus , NL4 . 3ΔEnv::GFPkdel/p6L1ter/PRD25A , which has both the p6L1ter mutation and an inactivating mutation [D25A] in PR . This p6-deficient , PR-deficient HIV virus showed no budding defect in 293T cells ( Figure 10A , right lanes ) . Consistent with these results , we observed that the HIV protease inhibitor saquinavir suppressed the budding defect of p6-deficient HIV in 293T cells ( Figure 10B ) . Previous reports have demonstrated that HIV budding is unaffected by PR inhibitors or PR mutations [24 , 44 , 45] . These results indicate that the p6 domain , although it plays an important role in HIV virus budding from 293T cells , is unlikely to play a direct , mechanistically essential role in the budding process . This general conclusion is also supported by studies of HIV budding in human T cells . Demirov et al . [35] previously demonstrated that loss of p6 , or the HIV late domain alone , “had little or no effect on particle release” from any of several human T cell lines and from primary human leukocytes ( peripheral blood mononuclear cells ) . We repeated these experiments in our own strain of Jurkat T cells and obtained similar results: our control and p6-deficient HIV viruses showed no significant difference in HIV budding from Jurkat T cells ( Figure 10C ) . Taken together , these results indicate that the p6-independent budding of HIV Gag we observed earlier ( Figures 6–8 ) has high relevance for HIV virus budding . The class E VPS proteins are thought to play direct roles in trafficking cargoes to MVBs , the formation of outward budding vesicles , and retrovirus budding [14 , 31 , 32] . However , recent studies have demonstrated that inhibition of class E VPS function does not block MVB biogenesis in animal cells [46–48] . To determine whether class E VPS function is required for exosome biogenesis , we took advantage of the fact that expression of an ATPase-defective form of VPS4B impairs class E VPS function [32 , 37 , 49] . Specifically , we generated a Jurkat T cell line that expressed DsRED-VPS4B/K180Q from a tetracycline-inducible promoter . The tetracycline-inducible expression of DsRED-VPS4B/K180Q is evident here in the fluorescence micrograph of two cells , one an uninduced cell that had been labeled with the plasma membrane marker PKH-67 ( green ) and fixed , the other a cell that had been exposed to 10 μg/ml tetracycline overnight , and then fixed ( Figure 11A–11C ) . Exosomes collected from cells incubated with or without tetracycline had similar levels of the exosomal markers CD63 and CD82 ( Figure 11D and 11E ) , indicating that inhibition of class E VPS function did not block exosome budding or the secretion of these proteins from the cell in exosomes . Class E VPS function is , however , required for cell growth and viability , and addition of tetracycline prevented cell growth ( Figure 11F ) . Exosomes secreted by cells expressing DsRED-VPS4B/K180Q had the size and spheroid morphology expected of exosomes , though they did seem prone to clustering ( Figure 11G and 11H ) . We next tested whether class E VPS proteins play the direct and essential role in retrovirus budding that is currently favored [14 , 31] . To explore this issue , we used a system established by Sundquist and colleagues for studying the role of class E VPS function in HIV budding [32 , 37 , 49] . This involves co-transfection of human cells with ( 1 ) plasmids designed to express WT or ATPase-defective forms of the AAA ATPase VPS4B ( DsRED-VPS4B/K180Q and DsRED-VPS4B/E235Q ) and ( 2 ) HIV proviruses or HIV Gag expression vectors , then measuring the amount of Gag released from the cells in sedimentable particles . Using the same VPS4B-expressing plasmids as Sundquist and colleagues , we too observed that inhibiting class E VPS function impaired the budding of HIV virus , in this case the budding of NL4 . 3ΔEnv::GFPkdel from Jurkat T cells ( Figure 12A ) . However , inhibiting class E VPS function had no detectable effect on the budding of the p6-deficient HIV virus NL4 . 3ΔEnv::GFPkdel/p6L1ter/PRD25A from Jurkat T cells ( Figure 12B ) .
Our data indicate that protein sorting to retroviral VLPs and viruses is mediated by the same signals that target proteins to exosomes . Our first evidence for this was the finding that higher-order oligomerization of CD43 is sufficient to induce its trafficking to HIV Gag VLPs . Other parallels between exosomal protein sorting and protein trafficking to VLPs/viruses include the sorting of diverse Gag proteins to ELDs and exosomes , the oligomerization-induced sorting of Gag to ELDs and exosomes , the ability to remove p6 without affecting release of HIV Gag VLPs or release of PR-deficient HIV virus , the fact that adding a plasma membrane anchor was sufficient to induce the budding of TyA ( the Gag-like structural protein of a yeast LTR retrotransposon ) , and the fact that all budding-competent proteins were sorted to ELDs and secreted from the cell in exosomes , whether of viral origin or not . An exosomal origin of HIV is also consistent with many prior observations [11] , including the secretion of amyloidogenic , exosomal proteins on retrovirus particles [56] and the interactions between HTLV-1 Gag and the exosomal proteins CD81 and CD82 [57] . The primary significance of these observations is that they reveal retrovirus budding to be a manifestation of a normal , cell-encoded exosome biogenesis pathway . This has important implications for the targeting of viral and nonviral proteins to sites of retrovirus budding and onto retrovirus particles , the identification and characterization of retrovirus budding factors ( and exosome biogenesis factors ) , and potentially for the targeting of antiretroviral agents to sites of budding and onto retroviral particles . An exosomal mechanism of retrovirus budding is also likely relevant to the evolutionary relationships between retroviruses and LTR retrotransposons [11] . For example , the ability to target TyA to ELDs and exosomes merely by adding an acylation tag indicates that acquisition of exosomal sorting information might be a critical step in the evolution of retroviruses from LTR retrotransposons , and that loss of membrane binding might mediate the reverse transition . This notion is consistent with the ability to convert the mouse LTR retrotransposon MusD from an intracellular , noninfectious transposon into a budding , infectious retrovirus merely by appending a retroviral MA domain to the N-terminus of its Gag-like protein [58] . The intracellular trafficking of HIV Gag has been studied closely . Some have concluded that retrovirus assembly initiates at endosomes and then proceeds to completion either there or at the plasma membrane [8 , 10 , 59–61] , whereas others report that retrovirus assembly initiates at the plasma membrane and proceeds to budding either there or at endosomes [62] . To us , these pathways are all consistent with an exosomal origin for HIV , for they are each compatible with the fact that exosome budding can occur at either ELDs or endosomes . Specifically , it appears that exosomes can bud from either endosomes or from ELDs , and that ELDs might form by lateral sorting in the plasma membrane as well as by endosome–plasma membrane fusion . Our results also have some relevance for exosome morphogenesis . The varied sizes and shapes of AcylTyA-containing exosomes indicate that these characteristics are not under strict mechanistic control . This suggests a model of exosome morphogenesis in which vesicle size and shape are mediated by the cargoes themselves . This would be consistent with the highly uniform morphology of retrovirus particles [13] , as well as the pronounced effects that certain Gag and PR mutations can have on virus morphology [24] . An exosomal model of retrovirus budding also offers new insights into why HIV budding is promoted by the HIV late domain . The most parsimonious interpretation of the data is that the late domain promotes virus budding indirectly . As for how , one possibility is that the late domain negatively regulates HIV PR until after budding , thereby ensuring the oligomerization-induced sorting of HIV Gag to exosomes . This hypothesis is supported by additional observations in the literature , such as the inhibitory effect of PR activity on budding [43] and the premature cleavage of the HIV Gag-Pol polyprotein in the context of an HIV late domain mutant [63] . The direct and indirect interactions between the HIV late domain and the class E VPS machinery indicate that they act together . Our data support this notion insofar as they show that class E VPS function appears to promote HIV budding via an indirect mechanism . It should be noted that our data do not exclude the possibility of more complex retrovirus budding mechanisms . For example , it is formally possible that the p6 domain and class E VPS proteins play direct , mechanistic , and essential roles in delivering HIV Gag to exosomes in some other cell types , or perhaps in a non-exosomal mechanism of budding . However , this last alternative offers no explanation for why HIV budding occurs at the same sites as exosome biogenesis or why HIV particles are enriched in exosomal markers [3 , 8 , 9 , 11 , 12 , 56 , 64–66] .
Commercial sources were used for the acquisition of N-Rh-PE , N-F-PE ( Avanti Polar Lipids , http://www . avantilipids . com ) , PKH-67 ( Sigma , http://www . sigmaaldrich . com ) , HIV protease inhibitor saquinavir ( Moravek , http://www . moravek . com ) , and antibodies ( Santa Cruz Biotechnology , http://www . scbt . com; Chemicon International , http://www . chemicon . com; Pharmingen , http://www . bdbiosciences . com; Abcam , http://www . abcam . com; and Jackson ImmunoResearch Laboratories , http://www . jacksonimmuno . com ) . Rabbit anti-HIV Gag p24 antibodies were from James Hildreth ( Meharry Medical College , Nashville , Tennessee ) . HIV Gag-GFP and AIP1/VPS31-DsRED were described previously [3] , the WT and mutant VPS4B expression vectors were from Wes Sundquist ( University of Utah ) , pNL4 . 3ΔEnv::GFPkdel was from Robert Siliciano ( Johns Hopkins University ) , and pDsRed-monomer-N1 was from Clontech ( http://www . clontech . com ) . HIV Gag mutants were amplified using primers designed to append an Asp718 site ( GGTACC ) immediately upstream of the start codon and a BamHI site ( GGATCC ) at the 3′ end of the open reading frame ( ORF ) . Following cleavage with Asp718 and BamHI , each PCR product was inserted between the Asp718 and BamHI sites of pcDNA3-GFP , pcDNA3-DsRed , or pcDNA3-DsREDm , in each case generating a continuous ORF encoding an in-frame fusion between the Gag ORF and the fluorescent protein ORF . To generate ORFs encoding LZ fusions , a fragment encoding NH2-LQRMKQLEDKVEELLSKNYHLENEVTRLKKLVGE-COOH was inserted between the Gag and fluorescent protein ORFs . Protein sequences for EIAV Gag , HTLV-1 Gag , MLV Gag , RSV Gag , MPMV-Gag , HERV-K Gag , SFV Gag , and TyA were used to design and synthesize human codon-optimized ORFs with Asp718 and BamHI sites flanking the ORF and inserted between the Asp718 and BamHI sites of pcDNA3-GFP . AcylTyA-GFP and Acyl ( G2A , C3A ) TyA-GFP were generated by amplifying the TyA ORF with primers designed to append codons for NH2-MGCINSKRKD-COOH or the G2A , C3A mutant version to the 5′ end of the ORF , which was also cloned upstream of and in frame with the GFP ORF in pcDNA3-GFP . Plasmids pNL4 . 3ΔEnv::GFPkdel/p6L1ter and pNL4 . 3ΔEnv::GFPkdel/p6L1ter/PRD25A were generated by amplifying the ApaI-SbfI fragment of pNL4 . 3ΔEnv::GFPkdel with primers designed to introduce the p6L1ter mutation ( change of nucleotides 2188–2190 from CTT to TGA ) or both the p6L1ter mutation and the D25A mutation of protease ( change of nucleotides 2379–2381 from GAT to GCC ) and then inserting these fragments into pNL4 . 3ΔEnv::GFPkdel . All experiments were performed with sequence-confirmed plasmids . Jurkat cells and K562 cells ( James Hildreth , Meharry Medical College , Nashville , Tennessee ) were maintained in serum-free Aim V medium ( GIBCO BRL , http://www . invitrogen . com ) , and our derivative of the Tet-on Jurkat T cell line ( Trex; Invitrogen , http://www . invitrogen . com ) was maintained in RPMI/10% fetal calf serum , with induction in 1 μg/ml tetracycline . Cells were pulse labeled with N-Rh-PE , N-F-PE , or PKH-67 as described [3] . Surface proteins were oligomerized by incubating cells with mouse monoclonal antibodies ( 1:100 ) for 45 min , washing four times with 1× PBS , incubating with secondary antibodies ( 1:100 ) for 35 min , washing three more times with 1× PBS , all at 4 °C . Cells were then either fixed or incubated at 37 °C for the time indicated and then fixed . Cells were transfected by mixing 1 × 107 cells with 5–10 μg of plasmid DNA ( 2:1 ratio for co-transfections ) at room temperature for 15 min , followed by electroporation at 300 V , 24 Ω , 800 μF . Images were collected and processed as described [3] . Immunogold surface labeling of cells was performed as described [3 , 67] . All images showing co-localization of proteins at ELDs were from experiments in which co-localization was detected in the majority of relevant cells , with the sole exception being the experiments with Acyl-LZ-DsRED and RSV Gag-GFP , in which co-localization was detected in approximately 10% of relevant cells . Exosomes were collected , analyzed by immunoblot and fluorescence microscopy , and purified by density gradient as described [3] . Protease protection experiments were performed by incubating exosomes with trypsin in the presence or absence of 0 . 1% Triton X-100 . For native gel electrophoresis , cells were lysed in 50 mM Tris HCl ( pH 7 . 5 ) , 150 mM NaCl , 1 mM EDTA , 1% Triton X-100 , 1× protease inhibitor cocktail ( Boerhinger-Ingelheim , http://www . boehringer-ingelheim . com ) , rotated at 4 °C for 2 h , clarified by centrifugation at 15 , 000×g for 15 min , separated by electrophoresis through 4%–20% gradient native gels ( Invitrogen ) , and processed for immunoblot .
The GenBank ( http://www . ncbi . nlm . nih . gov/Genbank ) accession numbers for the protein sequences discussed in this paper are as follows: EIAV Gag ( M16575 ) , HERV-K Gag ( Y17833 ) , HTLV-1 Gag ( D13784 ) , MLV Gag ( J02255 ) , MPMV-Gag ( M12349 ) , RSV Gag ( J02342 ) , SFV Gag ( U04327 ) , and TyA ( M18706 ) . The GenBank accession numbers for the human codon-optimized ORFs are as follows: EIAV Gag ( DQ421317 ) , HERV-K Gag ( DQ421322 ) , HTLV-1 Gag ( DQ421318 ) , MLV Gag ( DQ421319 ) , MPMV Gag ( DQ421321 ) , RSV Gag ( DQ421320 ) , SFV Gag ( DQ866825 ) , and TyA ( DQ421323 ) . | Exosomes are small , secreted organelles with the same topology as the cell and a similar size and composition as retrovirus particles . Based on these similarities , we proposed that retroviruses are , at their most fundamental level , exosomes . Little is known about the mechanisms of exosome biogenesis . We show here that higher-order oligomerization and plasma membrane binding are sufficient to target proteins into both exosomes and HIV virus-like particles . We also find that the HIV protein Gag , which possesses these general exosomal sorting elements , requires only these elements to bud from human cells . Others have proposed that the HIV p6 domain and the host class E vacuolar protein-sorting ( VPS ) machinery play direct , essential , and mechanistic roles in HIV budding . However , we show here that p6-deficient HIV can bud from cells at normal levels and that both p6-deficient HIV and exosomes can bud independently of class E VPS function . Thus , it appears that exosome biogenesis pathways mediate the budding of HIV from cells , whereas the HIV p6 domain and the class E VPS machinery promote budding indirectly . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"viruses",
"infectious",
"diseases",
"cell",
"biology",
"virology",
"microbiology"
] | 2007 | Higher-Order Oligomerization Targets Plasma Membrane Proteins and HIV Gag to Exosomes |
Behavioral output of neural networks depends on a delicate balance between excitatory and inhibitory synaptic connections . However , it is not known whether network formation and stability is constrained by the sign of synaptic connections between neurons within the network . Here we show that switching the sign of a synapse within a neural circuit can reverse the behavioral output . The inhibitory tyramine-gated chloride channel , LGC-55 , induces head relaxation and inhibits forward locomotion during the Caenorhabditis elegans escape response . We switched the ion selectivity of an inhibitory LGC-55 anion channel to an excitatory LGC-55 cation channel . The engineered cation channel is properly trafficked in the native neural circuit and results in behavioral responses that are opposite to those produced by activation of the LGC-55 anion channel . Our findings indicate that switches in ion selectivity of ligand-gated ion channels ( LGICs ) do not affect network connectivity or stability and may provide an evolutionary and a synthetic mechanism to change behavior .
Mapping the neural connections of nervous systems is often considered to be a fundamental step in understanding behavior [1 , 2] . However , a neural connectivity map carries no information about the activity of neurons and the nature of the connections that each neuron makes . Neurons are embedded in neural networks , which require a delicate balance between excitation and inhibition to maintain network stability [3 , 4] . Homeostatic processes , conserved from invertebrates to humans , can adjust synaptic and neuronal excitability to keep neural circuits functioning within their stable dynamic range [5–8] . In these circuits , ligand-gated ion channels ( LGICs ) are the principal signaling components that mediate fast inhibitory and excitatory neurotransmission . The Cys-loop LGIC receptors , which include the cation-selective nicotinic acetylcholine receptors ( nAChRs ) , serotonin type 3 receptors ( 5HT3Rs ) , and anion-selective GABAA and glycine receptors , form pentameric complexes in the plasma membrane [9 , 10] . Each individual subunit contains an extracellular N-terminal domain that harbors the ligand binding domain and four transmembrane spanning domains ( M1–M4 ) [11] . The charge selectivity of both anion and cation-selective channels is determined by residues in the M2 domain ( Fig 1A and 1B ) . In vitro studies have shown that LGIC channels can be switched from excitatory cation-selective to inhibitory anion-selective and vice versa through substitutions in the intracellular loop between M1 and M2 [12–15] . However , it is not known whether these channels with switched ion selectivity are functional in vivo . By switching the sign of a synapse , can the behavioral output of a neural circuit be reversed , or will a switch in the sign of a synapse cause defects in network development and stability ? Neuronal specification , receptor clustering , homeostatic processes , and behavioral feedback mechanisms may preclude such manipulations . The nematode Caenorhabditis elegans , the only animal with a completely defined neural wiring diagram [16 , 17] , is particularly suited to addressing these questions . The neural circuit that mediates the C . elegans escape response has been well characterized . The biogenic amine tyramine coordinates backward locomotion and the suppression of head movements during the C . elegans escape response elicited by touch to the anterior half of the body [18] . C . elegans has a single pair of tyraminergic neurons , the RIMs , which activate the homomeric tyramine-gated chloride channel , LGC-55 [19 , 20] . LGC-55 belongs to the Cys-loop LGIC family of receptors and is the only ionotropic tyramine receptor expressed in neurons and muscles that are directly postsynaptic to the tyraminergic neurons . Activation of LGC-55 induces the suppression of head movements and backward locomotion through the hyperpolarizaton of the neck muscles and premotor interneurons that drive forward locomotion . Here we changed the ion selectivity of LGC-55 , from an inhibitory tyramine-gated anion channel to an excitatory tyramine-gated cation channel , and reintroduced the excitatory channel in the native circuit . We show that switching the sign of the synapse within the escape circuit does not affect circuit development or stability and results in opposite behavioral responses .
The ion selective M2 domain of the tyramine-gated chloride channel LGC-55 is similar to the M2 domain of anionic Cys-loop receptors including the mammalian glycine receptors ( GlyRs ) , gamma-aminobutyric acid receptors ( GABAARs ) , and the C . elegans serotonin-gated chloride channel MOD-1 ( Fig 1B ) . To change the ion selectivity of LGC-55 , we replaced the residues of the M1–M2 loop with those that are conserved in structurally related cation channels . Using site directed mutagenesis , we generated cDNA clones encoding LGC-55 cation-I , containing the M1–M2 loop of the cationic 5HT3a channel ( RRSLPA to PDSGE ) , and LGC-55 cation-II , which includes an additional substitution at the 20ʹ position of the M2 segment ( R to D ) ( Fig 1B ) . The 20ʹ position of the M2 segment has been reported to increase the cation conductance [21 , 22] . To determine the ion selectivity of the engineered LGC-55 receptor , we recorded tyramine-elicited whole-cell currents in cultured muscle cells obtained from C . elegans strains that ectopically expressed either the wild type or engineered LGC-55 channel in body wall muscles . We analyzed current-voltage ( I-V ) relationships in varying ionic conditions: standard solution ( ES1 ) , low Na+ ( ES2 ) , and low Cl- ( ES3 ) . The reversal potential ( Erev ) of the wild-type LGC-55 anion channel in ES1 was -26 . 8 ± 3 . 1 mV ( n = 4 ) near the predicted Erev for a C . elegans anion-selective channel under our conditions . A reduction of extracellular chloride concentration lead to a rightward shift of the reversal potential ( Erev in ES3 = -1 . 9 ± 2 . 3 mV , n = 4 ) , while no significant differences in the Erev values for the LGC-55 anion receptor were observed when we reduced the Na+ concentration ( ES2 ) , consistent with our previous findings ( Fig 1C ) [19] . The reversal potential of the engineered LGC-55 cation-II channel in standard solution was 2 . 4 ± 1 . 2 mV ( n = 5 ) , near the Goldman–Hodgkin–Katz ( GHK ) -predicted value for a cation-selective channel in our conditions . Reduction of the extracellular Cl- concentrations did not lead to significant changes in this value ( Erev in ES3 = 1 . 7 ± 0 . 9 mV , n = 4 ) , whereas a shift to more negative potentials is observed when we decreased the extracellular Na+ concentration ( Erev in ES2 = -21 . 9 ± 2 . 6 mV , n = 5 ) . To analyze the relative Cl- and Na+ permeabilities ( PCl/PNa ) , we performed recordings using extracellular buffers containing different NaCl dilutions ( 1 , 0 . 5 , and 0 . 25 relative to the intracellular solution NaCl concentration; see Materials and Methods ) , and determined reversal potentials from current-voltage curves ( S1 Fig ) . The Erev values obtained for LGC-55 anion and engineered LGC-55 cation-II receptors were plotted against extracellular Cl- activity ( S1C Fig ) , and PCl/PNa values were obtained ( see Materials and Methods [14 , 23] ) . Wild-type LGC-55 exhibited a PCl/PNa of 18 . 8 , further confirming that these receptors are anion selective ( S1 Fig ) . In contrast , the PCl/PNa value of the engineered LGC-55 cation-II channel was 0 . 19 ( S1 Fig ) , indicating that the current passing through the chimeric LGC-55 cation-II channel is mainly carried by Na+ and that the Cl- dependent component is negligible ( Fig 1C ) . To further characterize the permeability properties of the LGC-55 cation-II channel , we analyzed Erev shifts after altering extracellular K+ and Ca2+ concentrations ( S2 Fig ) . An increase in the external K+ concentration significantly shifted the Erev towards more positive membrane potentials , whereas changes in the external Ca2+ had no significant effects on the Erev value ( S2 Fig ) , indicating that the engineered LGC-55 is mainly permeant to monovalent cations . Our observations are consistent with previous reports showing that similar mutations in the M1–M2 linker of GlyR dramatically increase the permeability to monovalent cations but not to calcium [14] . Does the engineered LGC-55 channel act as an excitatory receptor in vivo ? Transgenic animals that ectopically expressed the wild-type LGC-55 anion channel in body wall muscles quickly paralyzed on plates containing exogenous tyramine . Ligand-gated chloride channels hyperpolarize the C . elegans adult body wall muscle cells , which have a low intracellular Cl- concentration [24 , 25] . The activation of the LGC-55 anion channel caused muscle relaxation and overall body lengthening in animals overexpressing the anion channel in all body wall muscles ( Pmyo-3::LGC-55 anion ( zfEx31 ) : Δbody = 90 ± 13 μm , n = 53 ) . Wild-type animals displayed a slight , although not significant ( p = 0 . 07 ) , body lengthening in response to tyramine , which could be due to the endogenous LGC-55 anion channel expression in neck muscles ( wild type: Δbody = 19 ± 10 μm , n = 57 ) . In contrast , transgenic animals that expressed the LGC-55 cation-I or LGC-55 cation-II channel in all muscle cells became severely hypercontracted and displayed a shortened and contracted body posture in response to exogenous tyramine ( Pmyo-3::LGC-55 cation-I ( zfEx120 ) : Δbody = -180 ± 14 μm , n = 59; Pmyo-3::LGC-55 cation-II ( zfEx41 ) : Δbody = -220 ± 33 μm , n = 55 ) . Together , these data show that LGC-55 cation channels can function as excitatory receptors in vivo ( Fig 2A and 2B ) . Can the LGC-55 cation channel assemble into a functional synapse ? The tyraminergic RIM neuron make synaptic outputs onto the neck muscles and several head neurons that express LGC-55 . To visualize tyraminergic synapses , we expressed the synaptic vesicle marker , mCherry::RAB-3 in the RIM neurons . Expression of mCherry::RAB-3 in the RIM neurons localized to axonal puncta along the ventral process and in the nerve ring , consistent with presynaptic specializations with the AVB premotor interneurons , the neuromuscular junction ( NMJ ) and head motor neurons , respectively ( Fig 3A ) [16] . To examine the localization of the tyramine-gated chloride channel , we expressed a rescuing LGC-55 anion::GFP ( GFP , green fluorescent protein ) translational fusion under control of the lgc-55 promoter . LGC-55 anion::GFP receptors formed high-density clusters opposite presynaptic tyramine release sites in the nerve ring and the ventral process of the AVB premotor interneurons ( Fig 3B ) . In transgenic animals that expressed LGC-55 cation-II::GFP , we observed clustering to synaptic specializations opposite tyramine release sites in the nerve ring and along the ventral process similar to animals expressing the LGC-55 anion channel ( Fig 3B ) . To quantify the localization of the receptor to the post-synapse , we analyzed the pre- and post-synaptic densities of synapses from the RIM onto the AVB ( S3A Fig ) . Both LGC-55 anion::GFP and LGC-55 cation-II::GFP cluster in discrete regions of the ventral process of the AVB ventral process opposite the tyramine release sites ( S3B–S3D Fig ) . However , the RIM-AVB synaptic markers were slightly more diffuse in the LGC-55 cation-II transgenic animals ( S3B–S3D Fig ) . Synaptic markers also properly localized in tyramine-deficient , tdc-1 mutants and lgc-55 null mutants ( Fig 3B and S3 Fig ) . The postsynaptic densities were expanded in tyramine-deficient animals , whereas presynaptic densities were enlarged in the tyramine receptor mutants . The RIM-AVB synaptic markers were slightly more diffuse in tyramine signaling mutants and the LGC-55 cation-II transgenic animals . Our data indicate that tyramine signaling and the sign of the synapse may affect the morphology of the synapse but does not change the formation of proper pre- and postsynaptic specializations . To analyze the functional consequences of converting the ion selectivity of the LGC-55 channel , we compared the response of animals that expressed LGC-55 anion or LGC-55 cation under control of the native promoter to exogenous tyramine . LGC-55 is expressed in neck muscles , the RMD and SMD motor neurons that control foraging head movements , and the AVB premotor interneurons that drive forward locomotion ( Fig 4 ) . On plates containing exogenous tyramine , wild-type animals relax their neck and make long backward runs as a result of the activation of the LGC-55 anion receptor ( S1 Movie ) . The animals eventually become immobilized in part through the subsequent activation of a tyramine G-protein coupled receptor SER-2 [19 , 26] . We have previously shown that the relaxation is mediated through hyperpolarization of the neck muscles and the cholinergic RMD and SMD head motor neurons that express LGC-55 . Exogenous tyramine induced neck muscle relaxation and lengthening of the head in wild-type ( LGC-55 anion ) animals and lgc-55 null mutant animals that express a rescuing LGC-55 anion transgene ( wild type: Δhead = 10 ± 6 μm , n = 68; Plgc-55:LGC-55 ( zfEx2 ) : Δhead = 11 ± 4 μm , n = 75 ) ( Fig 5A and 5B ) . Head movements persisted in lgc-55 mutants [19] , with no significant change in head length ( lgc-55 ( tm2913 ) : Δhead = 3 ± 13 μm , n = 65 ) . In contrast , transgenic animals that expressed the engineered LGC-55 cation-I or LGC-55 cation-II channel under control of the native promoter had a hypercontracted and shortened head length in response to exogenous tyramine ( Plgc-55::LGC-55 cation-I ( zfEx8 ) : Δhead = -15 ± 2 μm , n = 49; Plgc-55::LGC-55 cation-II ( zfEx40 ) : Δhead = -28 ± 2 μm , n = 49 ) ( Fig 5A and 5B ) . In wild-type animals , exogenous tyramine also induced long backward runs preceding immobilization through the LGC-55 mediated inhibition of the AVB premotor interneurons that drive forward locomotion ( wild type: Δfwd-bwd = -8 . 46 ± 2 . 98 body bends , n = 40 ) ( Fig 5A and 5D , S1 Movie ) [19] . Backward locomotion was further increased in transgenic animals that expressed the LGC-55 anion under control of its endogenous promoter ( Plgc-55:LGC-55 ( zfEx2 ) : Δfwd-bwd = -28 . 5 ± 3 . 3 backward body bends , n = 29 ) . lgc-55 null mutants did not make long reversals when exposed to exogenous tyramine ( lgc-55 ( tm2913 ) : Δfwd-bwd = 2 . 58 ± 0 . 9 body bends , n = 34 ) . In sharp contrast , LGC-55 cation animals exhibit long forward runs ( Plgc-55::LGC-55 cation-I ( zfEx8 ) : Δfwd-bwd = 46 . 5 ± 8 . 4 body bends , n = 28; Plgc-55::LGC-55 cation-II ( zfEx40 ) : Δfwd-bwd = 80 . 6 ± 9 . 5 body bends , n = 39 ) , which continued for an extended period of time ( Fig 5A , 5C , and 5D; S2 Movie ) . The forward runs and head contractions were more pronounced in LGC-55 cation-II than in LGC-55 cation-I transgenic animals , supporting the notion that the R to D substitution at the extracellular ring of the M2 domain increases the cation conductance ( Fig 5C and 5D ) . However , we cannot exclude the possibility that the R to D substitution may also affect gating of the engineered LGC-55 cation-II channel . Animals that express the LGC-55 anion channel become immobilized more quickly than those expressing the LGC-55 cation channel . This suggests that the immobilization on exogenous tyramine is , in part , due to the inhibition of the forward premotor interneuron , AVB ( Fig 5C ) . The LGC-55 anion channel was shown to coordinate backward locomotion and the suppression of foraging head movements during the C . elegans escape response elicited by gentle anterior touch [19] . To test if the LGC-55 cation channel functions in response to endogenous tyramine release , we analyzed the escape response of transgenic LGC-55 cation animals ( Fig 5 ) . Laser ablation [27] , genetic analysis [18 , 19] , and calcium imaging experiments [28 , 29] support the following model for the circuit that controls the escape response ( Fig 4 ) : gentle anterior touch activates the mechanosensory ALM/AVM neurons that inhibit the PVC and AVB forward premotor interneurons and activate the AVD/AVA backward premotor neurons causing the animal to move backward . The tyraminergic motor neurons ( RIM ) are activated during the reversal through gap junctions with the AVA backward premotor interneurons [18] . Tyramine release promotes long backward runs and induces the suppression of head movements through activation of the LGC-55 anion channel in the AVB forward premotor interneurons and neck muscles , respectively ( Fig 4 ) [19] . In response to touch , wild-type animals suppressed head movements by relaxing their head ( Δhead = 5 ± 0 . 001 μm , n = 39 ) and reversed on average 3 . 14 +/- 0 . 18 backward body bends ( n = 100 ) ( Fig 6 , S4 Fig , S3 Movie ) . lgc-55 null mutant animals made shorter reversals than the wild type and fail to suppress the exploratory head movements during the reversal , with no significant change in head length ( 2 . 45 ± 0 . 15 backward body bends , n = 100 ) . Strikingly , transgenic animals that expressed the LGC-55 cation channel variants contracted their neck muscles in response to touch ( Plgc-55::LGC-55 cation-I ( zfEx8 ) : Δhead = -11 ± 1 . 6 μm , n = 32; Plgc-55::LGC-55 cation-II ( zfEx40 ) : Δhead = -14 ± 0 . 009 mm , n = 26 ) , and the average reversal length was markedly reduced ( Plgc-55::LGC-55 cation-I ( zfEx8 ) : 1 . 57 ± 0 . 1 body bends , n = 100; Plgc-55::LGC-55 cation-II ( zfEx40 ) : 1 . 22 ± 0 . 1 body bends , n = 100 ) ( Fig 6A and 6C ) . Furthermore , transgenic LGC-55 cation animals displayed ratchety backward locomotion , often pausing during their reversal ( S4 Movie ) . In contrast to animals expressing the LGC-55 anion , which made long spontaneous reversals , LGC-55 cation animals predominantly make short reversals , and the number of spontaneous reversals is increased ( S5 Fig ) . We previously proposed a model in which tyramine stimulates long reversals through the LGC-55 anion mediated hyperpolarization of the AVB forward premotor interneuron [19] . Our results strongly support this model in which the substitution of the LGC-55 anion with the LGC-55 cation induces depolarization of the AVB forward premotor interneuron during the reversal and the simultaneous activation of the forward and backward locomotion programs . We used optogenetics to determine if the contrasting behavioral responses in LGC-55 anion and LGC-55 cation animals is directly dependent on tyramine release from the RIM . Upon exposure to blue light , wild-type animals that expressed the light-activated cation channel , ChannelRhodopsin 2 ( ChR2 ) in the RIM , relaxed their neck muscles ( Ptdc-1::ChR2 ( zfIs9 ) : Δhead = 7 ± 1 . 4 μm , n = 28 ) ( Fig 6D , S5 Movie ) . In contrast , LGC-55 cation animals , which also expressed ChR2 in the RIM , hypercontracted neck muscles in response to blue-light exposure ( Plgc-55::LGC-55 cation-II ( zfEx213 ) :Δhead = -24 ± 1 . 4 μm , n = 20 ) ( Fig 5D , S6 Movie ) . The relaxation in the LGC-55 anion- and the contraction in the LGC-55 cation transgenic animals were abolished in tyramine-deficient tdc-1 mutants ( tdc-1 ( n3420 ) : Δhead = -2 ± 3 μm , n = 25; tdc-1 ( n3420 ) ; Plgc-55::LGC-55 cation-II ( zfEx275 ) : Δhead = -1 ± 3 μm , n = 16 ) ( Fig 6D ) . These data support the notion that tyramine , released from the RIM , directly activates the tyramine-gated chloride or cation channel , LGC-55 , in the postsynapstic muscle cells . Furthermore , these results indicate that the engineered LGC-55 cation channels are properly expressed and functional at the synapse within the neural circuit that modulates the C . elegans escape behavior .
We have shown that the replacement of the M1–M2 loop of the C . elegans inhibitory tyramine receptor , LGC-55 , with that of related cation channels changes the ion selectivity from anions to monovalent cations . We have demonstrated that these engineered receptors with switched ion selectivity properly localize to the synapse and are functional in vivo . Most strikingly , we show that behavioral outputs can be reversed by switching the sign of a synapse within a neural network . Mutations in the M1–M2 linker that change the ion selectivity can also lead to changes in the gating and desensitization kinetics of the channel [30] . While the LGC-55 cation channel may also exhibit kinetic differences compared to the native LGC-55 anion channel , the opposite phenotypes observed in animals that express the LGC-55 anion channel versus those that express the LGC-55-cation channel indicate that the difference in ionic selectivity is responsible for the reversal in behavioral outputs . Previous studies in both vertebrates and invertebrates have shown that neurotransmitter release is not required for the initial development of neural circuits [31 , 32] and does not affect clustering of postsynaptic LGICs [33] . This indicates that changing the sign of the synapse does not affect the proper wiring of neural circuits or functional synaptic transmission . While homeostatic mechanisms that maintain the balance of excitatory and inhibitory are important for network stability , these mechanisms may only occur upon perturbation within the dynamic range of the response but not when the perturbation changes the sign of the synapse . Our results indicate that neural connectivity and the sign of synaptic connections represent independent modules of the nervous system that provide a degree of freedom in generating behavioral outputs . For example , in the developing brain , GABA’s action switches from excitatory to inhibitory because of changes in the intracellular concentration of chloride [34] . Moreover , excitatory and inhibitory GABA signaling appears to coexist in the adult mammalian nervous system [35] . While in vertebrates acetylcholine ( ACh ) LGIC receptors are exclusively cation selective and GABA LGIC receptors are anion selective , this distinction is not as stringent in invertebrates . Molluscs have inhibitory anion-selective ACh receptors in addition to the typical excitatory cation-selective ACh LGICs [36] , and C . elegans has both anion- and cation-selective ACh and GABA-gated LGICs [37–39] . Phylogenetic analysis of ion channel domains of the LGICs indicates that the C . elegans GABA-gated cation channels are more similar to the anionic GABA channels , and molluscan anion ACh channels are more closely related to the cationic ACh channels ( Fig 7 and S6 Fig ) . This indicates that these nematode cationic GABA channels have evolved from their anionic ancestors through mutations in the ion selectivity domain , much like the engineered mutations causing the ionic switch in our engineered cation channel . The molluscan anionic ACh channels appear to have followed the opposite trajectory and changed the ion selectivity of their cationic ancestors [36] . Taken together with our results , this suggests that molecular changes in LGICs that result in a switch of the ions they flux provides an evolutionary mechanism to change behavior . Our synaptic engineering of chemical synapses , together with the recent introduction of synthetic electrical synapses [41] , indicates that the C . elegans connectome is remarkably stable . It will be interesting to see whether such manipulations are possible in neural circuits of other genetically tractable organisms . The engineering of ion selectivity of LGICs can be used as a general method to artificially change the sign of synapses in existing circuits . This synaptic engineering approach may have a broad range of applications in neuroscience , including reprogramming neurotransmitter outputs and the ability to test neural circuit models , and may present a new avenue to change behavior .
All C . elegans strains were grown at room temperature ( 22°C ) on nematode growth media ( NGM ) agar plates with OP50 Escherichia coli as a food source . The strains used in this study were Bristol N2 ( wild type ) , QW89 lgc-55 ( tm2913 ) , MT10661 tdc-1 ( n3420 ) , QW190 Pmyo-3::LGC-55 anion ( zfEx31 ) , QW925 Pmyo-3::LGC-55 cation-I ( zfEx120 ) , QW224 Pmyo-3::LGC-55 cation-II ( zfEx41 ) , QW606 lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55::LGC-55::GFP ( zfEx189 ) , QW900 lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55::LGC-55 cation-II::GFP ( zfEX349 ) , QW1124 tdc-1 ( n3420 ) ; lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55::LGC-55::GFP ( zfEx463 ) , QW827 lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3 ( zfIs61 ) , QW802 lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55short ( -120-0 ) ::LGC-55::GFP ( zfIs72 ) , QW875 lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55short ( -120-0 ) ::LGC-55 cation-II::GFP ( zfIs79 ) , QW876 tdc-1 ( n3420 ) ; lgc-55 ( tm2913 ) ; Pcex-1::mCherry::RAB-3; Plgc-55short ( -120-0 ) ::LGC-55::GFP ( zfIs72 ) , QW51 lgc-55 ( tm2913 ) ; Plgc-55::LGC-55 ( zfEx2 ) , QW74 lgc-55 ( tm2913 ) ; Plgc-55::LGC-55 cation-I ( zfEx8 ) , QW219 lgc-55 ( tm2913 ) ; Plgc-55::LGC-55 cation-II ( zfEx40 ) , CB151 unc-3 ( e151 ) , QW40 lgc-55 ( tm2913 ) ; unc-3 ( e151 ) , QW538 tdc-1 ( n3420 ) ; unc-3 ( e151 ) , QW637 lgc-55 ( tm2913 ) ; unc-3 ( e151 ) ; Plgc-55::LGC-55 cation-II ( zfEx207 ) , QW333 unc-3 ( e151 ) ; Ptdc-1::ChR2 ( zfIs9 ) , QW326 tdc-1 ( n3420 ) ; unc-3 ( e151 ) ; Ptdc-1::ChR2 ( zfIs9 ) , QW327 lgc-55 ( tm2913 ) ; unc-3 ( e151 ) ; Ptdc-1::ChR2 ( zfIs9 ) , QW747 tdc-1 ( n3420 ) ; lgc-55 ( tm2913 ) ; unc-3 ( e151 ) ; tdc-1::ChR2 ( zfIs9 ) ; Plgc-55::LGC-55 cation-II ( zfEx275 ) , and QW1283 lgc-55 ( tm2913 ) ; unc-3 ( e151 ) ; tdc-1::ChR2 ( zfIs9 ) ; Plgc-55::LGC-55 cation-II ( zfEx275 ) Standard molecular biology techniques were used . An lgc-55 rescue construct was made by cloning an lgc-55 genomic fragment corresponding to nucleotide ( nt ) -2663 to +3895 relative to the translation start site into the EcoRV site in yk1072c7 [19] . To make the chimeric LGC-55 cation-I receptor , we performed DpnI site-directed mutagenesis on the lgc-55 rescuing construct using a primer that corresponded to the genomic sequence of the M1–M2 loop of the 5HT3a channel with 20 nt on either side homologous to the same region in LGC-55 . The LGC-55 cation-II was made using DpnI site-directed mutagenesis with a primer that changed the codon at nts 1042–1044 relative to the translational start site , corresponding to a R to D substitution at the 20ʹ position of the M2 loop . LGC-55 anion::GFP and LGC-55 cation-II::GFP translational fusion constructs were made by cloning GFP into an engineered AscI restriction site in the respective Plgc-55::LGC-55 constructs in the sequence encoding the intracellular loop between TM3 and TM4 . For muscle-specific expression of LGC-55 and LGC-55 cation-II , the full-length lgc-55 or lgc-55 cation-II cDNA was cloned into pPD95 . 86 behind the myo-3 promoter . For the identification of synapses in the RIM , we cloned a 1 . 1 kb cex-1 promoter fragment , which drives expression in the only RIM , upstream of the mCherry::RAB-3 fusion protein from the Gateway vector , pGH8 , to produce the plasmid Pcex-1::mCherry::RAB-3 . Transgenic strains were obtained by microinjection of plasmid DNA into the germline . At least three independent transgenic lines were obtained , and data are from a single representative line . Transgenic animals were generated in an lgc-55 null background , unless otherwise noted . Transgenic animals were made by coinjecting Plgc-55::LGC-55 anion , Plgc-55::LGC-55 cation-I , Plgc-55::LGC-55 cation-II , Pmyo-3:LGC-55 anion , Pmyo-3::LGC-55 cation-I , Pmyo-3::LGC-55 cation-II , Plgc-55::LGC-55::GFP , or Plgc-55::LGC-55 cation-II::GFP at 20 ng/μl or Pcex-1::mCherry::RAB-3 at 5ng/μl along with the lin-15 rescuing plasmid pL15EK at 80 ng/μl into lgc-55 ( tm2913 ) ; lin-15 ( n765ts ) animals , unless otherwise noted . All strains were examined for colocalization of the presynaptic vesicle marker mCherry::RAB-3 in the RIM with the LGC-55 anion::GFP postsynaptic receptor using fluorescence confocal microscopy ( Zeiss and Pascal imaging software ) . Images shown are compressed z-stacks formatted using ImageJ software . Embryonic cells were isolated and cultured as described [42] . Briefly , adult animals expressing the Pmyo-3::LGC-55 anion or LGC-55 cation-II; Pmyo-3::GFP transgenes were exposed to an alkaline hypochlorite solution ( 0 . 5 M NaOH and 1% NaOCl ) . Eggs released were treated with 1 . 5 U/ml chitinase ( Sigma-Aldrich , St . Louis , Missouri ) for 30 to 40 min at room temperature . The embryonic cells were isolated by gently pipetting and filtered through a sterile 5 μm Durapore syringe filter ( Millipore Corporation , Billerica , Massachusetts ) to remove undissociated embryos and newly hatched larvae . Filtered cells were plated on glass coverslips coated with peanut lectin . Cultures were maintained at RT in a humidified incubator in L-15 medium ( Hyclone , Logan , Utah ) containing 10% fetal bovine serum . Complete differentiation to muscle cells was observed within 24 h . Electrophysiology experiments were performed 2 to 8 d after cell isolation . Muscle cells from transgenic animals were identified by GFP expression . Whole-cell patch clamp recordings were performed using a HEKA EPC-9 patch clamp amplifier . Recording pipettes with a resistance of 3–7 MΩ were used . The intracellular solution ( I1 ) contained 115 mM K-gluconate , 25 mM KCl , 0 . 5 mM CaCl2 , 50 mM HEPES , 5 mM Mg-ATP , 0 . 5 mM Na-GTP , 0 . 5 mM cGMP , 0 . 5 mM cAMP , and 1 mM BAPTA ( PH 7 . 4 ) . For ionic selectivity experiments , extracellular solutions with different concentrations of Na+ and Cl- were used: ES1 ( standard solution , 150 mM NaCl , 5mM KCl , 1mM CaCl2 , 4 mM MgCl2 , 15 mM HEPES , 10 mM glucose , and pH 7 . 2 with NaOH ) , ES2 ( low Na+ , as ES1 except 15 mM NaCl , 135 mM NMDG-Cl ) ES3 ( low Cl- , as ES1 , except 30 mM NaCl , 120 mM Na-gluconate ) . For K+ and Ca2+ permeability studies , the solutions used were ES4 ( as ES2 except 140 mM KCl and 0 mM NMDG-Cl ) and ES5 ( as ES2 except 25 mM CaCl2 , 85 mM NMDG-Cl ) . Current-voltage relationships were determined by measuring the current peak after 250 ms perfusion of extracellular solution containing 0 . 5 mM tyramine at holding potentials ranging from -60 to +60 mV in 20 mV steps . For the dilution-potential experiments , the intracellular and extracellular buffer composition were similar to those previously reported [43] . The intracellular solution for these experiments was ( I2 ) 145 mM NaCl , 1mM CaCl2 , 1 mM MgCl2 , 1mM EGTA and 10 mM HEPES , 10 mM glucose , pH 7 . 2 . Control extracellular solution ( 1NaCl , symmetrical condition ) contained 145 mM NaCl , 1mM CaCl2 , 1 mM MgCl2 , and 10 mM HEPES ( pH 7 . 2 ) . The NaCl concentration were reduced to 72 . 5 and 36 . 25 mM in the extracellular buffers used in the dilution experiments ( 0 . 5 and 0 . 25 NaCl , respectively ) . Osmolarity was maintained by adding sucrose . PCl/PNa permeability ratios were obtained by fitting shifts in the Erev to the GHK equation: Erev = ( RT/F ) ln {[PNa ( aNa ) o + ( aCl ) i PCl] / [PNa ( aNa ) i + ( aCl ) o PCl]} , where Erev is the potential where the current is zero , R is the gas constant , T is the temperature , F is the Faraday’s constant , Pion is the permeability of the ion , and ( aion ) is the activity of the ion in the extracellular ( subscript o ) or intracellular ( subscript i ) solutions . Data analyses were performed using Igor Pro software ( Wavemetrics Inc , Lake Oswego , Oregon ) . Mean currents were fitted by a single exponential function: I ( t ) = Io exp ( -t/τd ) + I∞ , where Io is the current at the peak , I∞ is the current at the end of the recording , and τd the current decay time constant . Data were normalized to Imax , and the mean peak value in each condition was obtained after averaging three different traces ( obtained not consecutively but in different voltage protocols in the same experiment ) . If the difference in current peak values was more than 80% for a given condition , the whole experiment was discarded . Reversal potential values are shown as mean ± standard error of 4–5 independent experiments for each extracellular solution . Curve fitting and statistical analysis were performed using Sigma Plot 11 . 0 ( Systat Software . ) . All behavioral analysis was performed with young adult animals ( 18–24 h post L4 ) at room temperature ( 22 ºC ) ; different genotypes were scored in parallel , with the researcher blinded to the genotype . Quantification of tyramine resistance and tyramine-induced reversals was performed as described [19] . To quantify body length on exogenous tyramine , animals were placed on agar plates supplemented with 30 mM tyramine . Still frames were taken at 5 min after exposure to tyramine , and animals were measured using ImageJ software . To quantify head length on exogenous tyramine , animals were placed on 30 mM tyramine plates . Still frames were taken at 5 min after exposure to exogenous tyramine . The neck was defined as the length from the anteriormost point of the buccal cavity to the posterior of the pharyngeal bulb ( as illustrated in Fig 3B ) . Head contraction assays in response to touch and optogenetic activation of the RIM were performed in an unc-3 ( e151 ) mutant background . unc-3 ( e151 ) animals have normal head and neck movements but have defects in the specification ventral cord neurons that affect locomotion [43] . The unc-3 ( e151 ) genetic background was used in these assays to prevent backward locomotion in response to touch and to maintain the animal in the field of view at high magnification that would allow for accurate neck measurement . Head lengths were measured using ImageJ software . To quantify head lengths in response to touch , animals were filmed using a Sony SX910 camera and AstroII DC software for 10 s before and after a touch posterior to the pharyngeal bulb with an eyelash . Still frames were taken from the video just prior and just after the touch . Head lengths were measured from these still frames using ImageJ software . For optogentic experiments , L4 animals were transferred to assay plates that were seeded with either OP50 E . coli that was supplemented with or without all-trans retinal to a final concentration of 660 μM . Animals were raised overnight on plates with or without all-trans retinal . To quantify head lengths in response to optogenetic activation of the RIM , animals were filmed for 10 s before and after a 2-s blue light pulse . Still frames were taken from the video just prior to and during the blue light exposure . | Fast neurotransmission in the nervous system is mediated by ligand-gated ion channels . Within the nervous system , the sign of synaptic connections , i . e . , whether they are excitatory or inhibitory , is determined by the charge of the ions that flow through these channels . In general , channels that conduct positive ions are excitatory , whereas channels that conduct negative ions are inhibitory . Here , we investigate if it is possible to flip the behavioral output of a neural circuit by changing the sign of a synapse within that circuit . The neural circuit we study controls the escape response of the nematode Caenorhabditis elegans . Activation of the inhibitory receptor , the tyramine-gated chloride channel LGC-55 , coordinates the suppression of head movements and backward locomotion during the C . elegans escape response . We selectively mutated LGC-55 to transform it from an inhibitory into an excitatory ion channel and generated transgenic worms in which the inhibitory channel is replaced by the excitatory version of LGC-55 . We show that the excitatory version of LGC-55 is properly localized to the postsynaptic compartments , is activated by the natural ligand , and does not affect network connectivity or stability . However , its expression leads to behavioral responses that are opposite to those produced by activation of the native inhibitory channel . We propose that switching the sign of a synapse not only provides a synthetic mechanism to flip behavioral output but could also be an evolutionary mechanism to change behavior . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | A Change in the Ion Selectivity of Ligand-Gated Ion Channels Provides a Mechanism to Switch Behavior |
The traditional cardiac model-building paradigm involves constructing a composite model using data collected from many cells . Equations are derived for each relevant cellular component ( e . g . , ion channel , exchanger ) independently . After the equations for all components are combined to form the composite model , a subset of parameters is tuned , often arbitrarily and by hand , until the model output matches a target objective , such as an action potential . Unfortunately , such models often fail to accurately simulate behavior that is dynamically dissimilar ( e . g . , arrhythmia ) to the simple target objective to which the model was fit . In this study , we develop a new approach in which data are collected via a series of complex electrophysiology protocols from single cardiac myocytes and then used to tune model parameters via a parallel fitting method known as a genetic algorithm ( GA ) . The dynamical complexity of the electrophysiological data , which can only be fit by an automated method such as a GA , leads to more accurately parameterized models that can simulate rich cardiac dynamics . The feasibility of the method is first validated computationally , after which it is used to develop models of isolated guinea pig ventricular myocytes that simulate the electrophysiological dynamics significantly better than does a standard guinea pig model . In addition to improving model fidelity generally , this approach can be used to generate a cell-specific model . By so doing , the approach may be useful in applications ranging from studying the implications of cell-to-cell variability to the prediction of intersubject differences in response to pharmacological treatment .
Mathematical models of cardiac electrophysiology trace their roots to Hodgkin and Huxley’s seminal work from 1952 [1] . Since then many models have been developed describing cardiac electrophysiology in a number of species and cell types helping to make modeling an integral part of cardiac research [2–5] . The typical method for model development and parameterization is a bottom-up approach . Individual ionic membrane currents are characterized using voltage-clamp experiments from which mathematical equations are derived [6 , 7] . Although it has led to many advances , this traditional approach to model development has several limitations , including: Several studies have looked into how to improve model parameterization . Approaches in cardiac myocyte modeling have included the parameterization of individual channel dynamics , typically when making more complex Markov models of ionic currents [6 , 15–17] . Whole-cell optimization approaches have focused on generating models that can match action potentials from different types of cardiomyocytes , using both simple models consisting of a few generic currents [13 , 18–20] and more physiologically detailed ionic models [21–25] . A resulting synthesis is that optimization results are improved when models are fit to data beyond a single action potential , e . g . , action potentials from multiple pacing rates [13 , 19 , 22] or voltage waveforms during varying current injection [20 , 21 , 24] . In particular , using a global search heuristic applied to an ionic model , Syed et al . demonstrated that it is feasible to estimate conductance parameters for experimental data and showed that the fits improved when using data recorded during multiple periodic pacing frequencies [22] . Sarkar and Sobie presented a much simpler , but more approximate , linear regression strategy to estimate model conductances based on biomarkers from simulated model output and have used it to investigate how specific conductances relate to particular model outputs [26] . In neuroscience , considerably more research has been carried out on parameter estimation problems ( e . g . , [27–30] ) and a few studies have developed protocols that allow parameterization of cell-specific models [31 , 32] . However , these protocols are not directly applicable to cardiac myocytes , due to intrinsic differences in electrophysiological behavior between neurons and cardiomyocytes . Here , we present a novel strategy to develop cardiac models by pairing dynamically rich electrophysiology protocols with powerful computational parameter fitting methods . We first developed novel electrophysiology protocols that probe the dynamics of a subset of ionic currents in an intact cardiac myocyte without ion channel inhibitors , agonists , or unphysiological ion concentrations . The protocol consists of ( 1 ) stochastic current-clamp stimulation and ( 2 ) multi-step voltage clamping . As will be discussed , stochastic stimulation represents a quick method to sample the rate-dependent cardiac dynamics , while the multi-step voltage-clamp protocol is designed to highlight individual currents using a tailored sequence of holding potentials . Based on the assumption that ion-channel kinetics are preserved among ( healthy ) subjects while maximal conductances vary as a result of differences in expression levels , the resulting data are used to estimate maximal conductance values of several ionic currents and maximal flux of calcium ion transporters in the model . Because of the complexity of the data , hand parameter tuning is not feasible; thus , we utilized a genetic algorithm ( GA ) , which is an efficient method for such complex optimization applications [33] . The approach was first developed and validated computationally . It was then used to develop cell-specific models of isolated guinea pig ventricular myocytes that simulate the electrophysiological dynamics significantly better than does a standard guinea pig model .
We first developed our parameter estimation strategy using a guinea pig ventricular myocyte model ( Faber and Rudy [34] , the “FR” model ) and tested the ability of the optimization procedure to return the original parameter values . Traditionally , one of the main criteria for cardiac electrophysiology model quality is the ability of a model to describe the cardiac action potential . Therefore , we first ran the parameter estimation using a single FR model action potential as the target objective . Nine model parameters , describing maximal conductances of ionic currents [the sodium current ( INa ) , the L-type calcium current ( ICaL ) , the T-type calcium current ( ICaT ) , the inwardly rectifying potassium ( IK1 ) , the rapid and slow delayed rectifier potassium currents ( IKr and IKs ) , the plateau potassium current ( IKp ) , and the sarcolemmal calcium pump current ( IpCa ) ] and the maximal flux of the sarcoplasmic reticulum Ca2+-ATPase ( JSERCA ) were estimated using a GA technique . A GA optimization is initialized by a population of models with different parameter values . We used a population size of 500 model instantiations generated by randomly drawing values for the nine parameters from a range of 0 . 01–299% of the published value . The GA methodology uses ideas from evolution [23 , 33] . In GA terminology , the initial state is referred to as generation 0 and the 500 models as individuals . Fig 1 A–1 C shows three different individuals in generation 0 . Most of these generate action potentials that are very different from the target action potential ( Fig 1A and 1C ) . However , by chance , a few of the 500 individuals provide reasonably good fits to the action potential , even though the parameter values are very different from those of the baseline model ( parameter scaling of 1 , Fig 1B ) . The optimization proceeds in generations ( steps ) , for which the GA applies crossover ( parameter swapping ) , mutation ( parameter variation ) , and selection ( discarding poorly performing models ) to increase the fitness of the population ( reduce the error between model output and target objective ) . We ran the GA for 100 generations , as this was sufficient for the error of both the population average and the best individual to reach a minimum ( Fig 1 F and 1G ) . Although the optimized model action potential matches the optimization objective to a very high degree , the estimated parameter set does not match that of the FR model ( Fig 1E ) . This is consistent with previous results showing that if only a single action potential is used for parameterization , cardiac models may be overdetermined and more than a single set of parameter values can describe that action potential [20 , 21 , 26] . The duration of the action potential , and to a lesser extent its morphology , varies with stimulation interval and history . Models tuned to single action potentials during periodic pacing ( as in Fig 1 ) would not be expected to accurately reproduce such dynamics . A more complete method to probe cellular behavior is the restitution portrait [35] , which is a systematic , but prohibitively long , mapping of this rate dependence . An alternative approach that would significantly reduce the protocol duration , while maintaining some dynamic information , is random sampling of the rate dependence . To accomplish this , we utilized a stochastic stimulation protocol containing 11 randomly timed stimuli delivered over 5 s . When applied to the FR model , the stochastic stimulation results in considerable action potential variability ( Fig 2 A and 2B ) . We used this stochastic stimulation protocol and resulting voltage response as an optimization sequence to test the extent to which dynamic stimulus timing would improve the parameter estimation . Because the GA parameter estimation is a stochastic method , it was run 10 times with 10 different initial populations . For each run , we selected the best individual , i . e . , the model instantiation with the best fit to the target voltage trace . All 10 best individuals matched the voltage trace very closely ( Fig 2 A and 2B show the best one ) as was the case for the single action potential fitting . Compared to the single action potential , the stochastic stimulation leads to a modest overall improvement of the parameter estimation , but it did notably better in determining the maximal conductances of IKr , ICaL , and IKs ( Fig 2C ) . However , as shown in Fig 2C , a few parameters remain incorrectly estimated ( maximal conductances of ICaL and IKs ) , and some are estimated with a large spread ( maximal conductance values of IKp , IpCa , ICaT , and IKr ) . Sensitivity and correlation analyses illuminate why these parameters are less well determined—they have low sensitivity ( in which case they have minimal effect on the fitting objective , making them difficult to probe ) and/or they have inter-correlations ( in which case two or more parameters lack independent contributions to the fitting objective and are therefore difficult to discriminate ) ( S1–S4 Figs and S1 Text ) . To measure the predictive capabilities of the optimized models , we presented the 10 best individuals from the GA runs with a novel stochastic stimulation sequence ( S6 Fig ) . When subjected to this new prediction sequence , both the individuals trained with the stochastic stimulation sequence and those trained to the single action potential matched the baseline FR model response well ( S6 Fig ) , but the stochastic stimulation protocol lead to the better match , as shown by the smaller prediction error ( error between optimized model and target during the prediction sequence ) in Fig 2D . Thus , although the parameter recovery seem only modestly improved for the stochastic stimulation compared to the single action potential target , the stochastic stimulation outperforms the single action potential in that it results in models that are significantly better at predicting the response to a novel set of stimuli . To improve parameter estimation accuracy , more improvement is typically gained from adding measurements of a different state variable than adding additional measurements of the same state variable [36] . This suggests that a longer stochastic stimulation protocol is unlikely to yield much improvement . This idea is in line with our finding above ( Figs 2D and S6 ) that models optimized to the 5s stochastic stimulation protocol matched well when subjected to a novel stochastic stimulation sequence . Therefore , to improve the parameter estimation accuracy , we added a multi-step voltage clamp protocol to the objective function . Traditionally , voltage clamp is applied to a cell as a set of holding potentials varied systematically either in its timing or its potential to characterize one particular current . We developed a voltage clamp protocol consisting of a sequence of holding potentials , with each step designed to emphasize specific currents relative to the others ( Fig 3 ) . The rationale is that if a particular conductance contributes most of the total current for a particular holding potential , then only models fit with a correct value of that conductance will reproduce the current target for that phase of the protocol and therefore have a low fitting error . Our 6s long voltage clamp protocol effectively isolates IK1 , ICaL , and IKs as shown by the disproportionally large contributions of these currents in step -120 mV , +20 mV , and +40 and -30 mV , respectively ( Fig 3 B–3F ) . Therefore , we hypothesize that this protocol will directly improve the conductance estimation for these currents . Indeed , using the voltage clamp protocol as the objective during an optimization recovers the conductances for IK1 and IKs very accurately ( Fig 4B ) . The estimation of the ICaL conductance is very close to 1 , but is slightly overestimated in all runs . Less predictively , IpCa and IKp were also estimated more precisely than during stochastic pacing alone ( Fig 4B ) . However , a few conductances were estimated poorly ( in particular JSERCA and ICaT ) and models optimized based on voltage clamp data alone were , not surprisingly , inferior at predicting complex action potential dynamics during stochastic pacing ( Fig 4C ) . The extension of the target objective to include the multi-step voltage clamp protocol results in a joined unitless error function ( Eq 3 in the Methods ) . As both the stochastic pacing recording and the voltage clamp data is fit increasingly well during optimization , the error contribution from each sequence decreases ( Fig 4A , left ) . Although the main contribution to the total error comes from the stochastic pacing segment , the error from voltage clamp segment drops more during the optimization process , suggesting that both protocols help constrain the parameters . Running the optimization with the combined objective does indeed lead to improved accuracy of the parameter estimation , with all nine current parameters being recovered to within one standard deviation ( orange symbols , Fig 4B ) . For six of the nine current parameters , the combined protocol results in parameters whose mean estimates are closer to 1 and/or have less variational spread than either of the individual protocols alone ( Fig 4B ) . Only currents that were estimated very accurately by the voltage clamp protocol alone ( IKs , IK1 , and IpCa ) did not show improvement with the combined protocol . For some of the currents , one protocol segment is clearly better than the other in terms of parameter recovery ( e . g . , stochastic pacing for INa and voltage clamp for IKs ) . However , for other currents , in which both individual protocol segments result in off-target outcomes , the combined protocol produces estimates spanning 1 ( e . g . , IKp ) . Such improvement is consistent with the combined protocol restraining parameter space and avoiding local minima . Again , we tested the ability of the 10 best individuals to predict the response to a stochastic stimulation sequence to which they were not fit . The prediction error of the 10 individuals from the combined objective function runs was significantly lower than the error for the individuals that were estimated using only the stochastic stimulation protocol ( Fig 4C ) . Hence , the combined stochastic pacing and voltage clamp protocol improves both parameter recovery and prediction performance . To further improve the estimation results , the results of the first 10 GA runs were used as the new parameter bounds for a second set of runs ( see Methods ) , e . g . 0 . 01% to 299% changed to 92 . 8–114 . 9% for INa . Note that this method only works when the fits of the first 10 runs span the correct solution as is the case with the combined protocol . During this second , local , iteration , better fits are generated causing the error for both the voltage clamp and the current segments , as well as the total error for the best individual , to drop ( Fig 4A , right ) . Thus , using this iterative approach , the error bounds around the estimated parameter values decreased ( magenta symbols , Fig 4B ) and the prediction error reduced markedly ( Fig 4C ) . In summary , the combined protocol , consisting of stochastic stimulation and multi-step voltage clamp , allows accurate parallel estimation of eight maximal conductance values and maximal pump rate of SERCA for the FR guinea pig ventricular model . Such validation simulations laid the groundwork for using the method to fit computational models to real cardiac cell data . The parameter estimation method was next applied to four guinea pig left basal ventricular myocytes from four different animals . Each cell was subjected to the stochastic stimulation protocol in current clamp mode , followed by the multi-step voltage clamp protocol , using the perforated patch clamp technique . All four myocytes exhibited action potentials and membrane current responses that were very different from the baseline FR model ( Fig 5 shows output from one cell , S7–S9 Figs presents the results from the remaining three cells ) . In particular , their action potentials were substantially longer than those of the FR model and their current response to prolonged depolarization was substantially smaller . For each cell , the GA estimate from the experimental data fit much better than did the FR model ( Fig 5 and S7–S9 Figs ) . In particular , the optimization leads to very accurate voltage dynamics , which is important for arrhythmogenesis prediction . The total current is fit less well , potentially due to mismatch in ion channel kinetics ( see Discussion ) . Overall , the optimization results in more accurate predictions , with the prediction error being an order of magnitude lower for the fitted models than for the FR model ( Fig 5C and 5D ) . The dissimilarities between the original FR model and the experimental data led to considerable changes in the estimated values for the model parameters for all four cells ( Fig 6 ) . Interestingly , these changes were qualitatively similar between all four myocytes for most of the parameters , indicating conserved differences between our experimental data and the FR model . In particular , IKs and IKp are scaled down significantly and JSERCA is slightly reduced . In contrast , for all four cells , maximal conductance of IKr and IK1 are increased around 2-fold compared to the FR model , while ICaL is slightly increased . The results for INa show variation among cells , with a significant increase for three out of four cells and a small decrease for one cell . In summary , the optimized models show a much closer match to the experimental data as reflected in the individual voltage and current traces as well as in the prediction error . In addition , the optimization identified similar trends in the underlying channel conductance values for different cells from a particular region in the heart . Considered together with the demonstration that the approach accurately identifies model parameters ( Figs 2–4 ) , these findings suggest that the approach significantly improves the fidelity of the model for cellular data , relative to the published generic model .
In cardiac modeling , a single action potential or biomarkers derived from it such as amplitude and duration , is often used as a minimal objective for model parameterization . Ionic models can be optimized to fit single action potentials using , e . g . , global search heuristics [22 , 23] , but because the optimization problem is overdetermined , fits may be improved when adding more data , such as data recorded at multiple pacing rates [19 , 22] . In fact , relative to a single action potential , more complex driving protocols have the potential to dramatically improve parameterization by creating target objectives that are richer in information . On the other hand , to be experimentally feasible , protocols have to be relatively short in duration due to the inevitable current rundown that occurs in patched myocytes , even when using perforated patch . As a compromise , we utilized a stochastic stimulation protocol because it rapidly samples the rate-dependence of the action potential . In addition , irregular excitation patterns are present in many cardiac arrhythmia; thus models tuned to aperiodic excitation patterns are inherently better suited for modeling irregular arrhythmia . In our simulations , we found that the estimates for IKr and IKs were improved the most by the stochastic stimulation objective . During a single action potential , IKr and IKs have similar and compensatory effects ( S4 Fig ) , which impedes estimation of their conductances . In contrast , stochastic stimulation more thoroughly explores their kinetics , thereby revealing small differences throughout the protocol , resulting in a more accurate estimation . Although the stochastic stimulation protocol led to at most a modest improvement in the parameter estimation for the remaining parameters , the prediction error was reduced by an order of magnitude , compared to using a single action potential ( Fig 2 ) . Thus , significant model improvement is obtained through the use of a dynamically rich objective , as this helps the optimization avoid the false alternatives that can appear to fit well when dynamically sparse data are used for fitting . In addition to such current-clamp experiments , currents recorded during voltage clamping add additional data to improve fitting and optimization [21 , 31 , 32] . While our multi-step voltage clamp protocol alone is very useful for estimating many of the parameter values ( Fig 4B ) , it tends to generate models that fail to predict novel stochastic pacing data well ( Fig 4C ) , which is unsurprising given that it does not train the models according to membrane potential . In our simulations , the addition of the multi-step voltage clamp objective to the stochastic current-clamp stimulation objective enhanced the quality of the parameter estimation compared to using only stochastic stimulation ( Fig 4B ) . This improvement was the result of: ( i ) some parameters being estimated accurately by the voltage-clamp protocol and ( ii ) information on two , rather than a single , state variable putting more constraints on the parameter values [36] . In particular , estimates for all nine parameters became centered on their baseline values and the prediction error dropped by another order of magnitude relative to that of stochastic pacing alone . Finally , the iterative optimization approach [31] refined our in silico parameter estimation by decreasing the spread of the returned parameter sets , which caused the prediction error to again decrease by an order of magnitude . Generic models have the advantage that direct comparisons can be made among different simulation studies . However , when comparing a generic model such as the out-of-the-box FR model to our experimental data , there are substantial differences , which likely would cause inaccurate predictions if simulating , e . g . , effects of pharmacological agents or genetic variations . For one , there are clear distinctions in action potential morphology , e . g . , in the plateau phase ( Fig 5 ) . This difference in plateau phase most likely explains the method’s downscaling of the IKp conductance . Our recorded action potentials are also of considerably longer duration , which is consistent with the finding of a much reduced IKs in the voltage-clamp experiments . The step to -120 mV in the voltage-clamp protocol induced a much larger current in the experiments than in the FR model and IK1 conductance was increased accordingly in all four cells . These consistent changes in voltage traces and currents between our cells and the FR model may be due to lab-to-lab variability and to the fact that the FR model is not region-specific . Despite such consistent changes , the parameterization also points to important cell-to-cell variability , in particular for the INa conductance , which is increased in three cells and decreased in one . In neuronal modeling , it has become clear that different combinations of conductance parameter sets can give rise to the same activity pattern and that using average values of the conductances may fail to generate that pattern [10 , 11 , 37] . The differences in cell-to-cell variation in current densities have been linked to mRNA expression differences or post-translational modifications [38 , 39] . The extent to which such variation occurs in healthy cardiomyocytes remains to be seen , but some examples of functional coupling among ionic currents in perturbed systems have been described [40–42] . This failure-of-averaging concept may also extend to cardiac tissue: although intrinsic cellular heterogeneity tends to be smoothed out when myocytes are electrically coupled , coupled cells do not necessarily behave like their average . For example , a myocyte with intrinsically shorter action potential duration may promote repolarization in a cell pair [43] . Also , a range of synchronization patterns have been described in coupled pacemaker cells [44] . Thus , there may be important utility to developing cell-specific models . Indeed , cardiac cell-specific models have a range of potential application areas . First , the models can obviously be used to study cell-to-cell variability [45] . Second , in the clinic , inter-subject variability can lead to response differences among patients to pharmaceutical treatment . A dramatic example of this variation is the response to IKr-block , which can vary from minor changes in the electrocardiogram to ventricular tachyarrhythmias [46] . Understanding and predicting this variability is an important step towards patient-specific treatment . In turn , model optimizations such as those developed here represent an advancement towards patient-specific prediction . Finally , multiple models could be grouped into a heterogeneous population and used to generate more realistic responses than those of a randomly-generated population [47 , 48] . The developed protocols allow accurate estimation of nine conductance/flux parameters . To characterize a single cell more thoroughly , additional flux parameters could be included ( e . g . , those describing the sodium/calcium exchanger and the sodium/potassium pump ) , but as inclusion of more parameters makes the optimization problem harder , this may necessitate tweaking of the methods described here . As detailed below , possible strategies for improvement of the parameter estimations are: 1 ) improving the stochastic stimulation and voltage-clamp protocols; 2 ) adding measurements of different state variables during the same protocols ( e . g . , intracellular calcium or membrane resistance ) ; 3 ) incorporating altered solutions and/or ion channel blockers to improve isolation of individual currents [49]; 4 ) including ion channel kinetic parameters in the optimization; or 5 ) including relative weights for the current and the voltage contributions to the summed error . Our multi-step voltage-clamp protocol effectively isolates IKs , ICaL , and IK1 . An improved voltage-clamp sequence that isolates the remaining currents could improve estimation of their conductance/flux parameters . We designed the voltage clamp steps based on a priori knowledge of the current-voltage ( IV ) relations in guinea pig ventricular myocytes . As a way to design better protocols , an automated optimization approach may be feasible , i . e . , an optimization of the optimization protocol . Further , differences in structure , channel kinetics and IV-relationships between model and experiment are likely to result in less accurate parameter estimations [31] and may underlie the deviations between fit and experimental data during voltage clamp ( Fig 5 and S7–S9 Figs ) . Adding parameters describing ion channel kinetics to the optimization process would likely improve the fits and predictability , but would almost certainly necessitate longer voltage clamp protocols [6 , 16] . As channel kinetics are not expected to vary substantially among cells of the same type , a possible strategy is to first parameterize average channel kinetics in a cell population , then apply our method to derive cell specific models . Additionally , improvement could likely be gained by simultaneously recording calcium fluorescence and adding that to the objective function [36] , a strategy with merits illustrated by Fig 1 of Ref . [26] . As expected , local sensitivity analysis on simulated calcium traces demonstrates that they are most sensitive to changes in ICaL , IpCa , and JSERCA ( S5 Fig ) , which leads to the speculation that the estimation of these parameters could improve . Inclusion of calcium data may also allow determination of INaCa , which depends on and influences both intracellular calcium and transmembrane potential . Incorporation of membrane resistance in the objective function would also be expected to improve the fitting , as shown in recent work by Kaur et al . [25] . A potential caveat in such multi-objective optimization is that simultaneous good fits are not always achievable , necessitating trade-offs between the different objectives . In that case , balancing which objective ( s ) to prioritize would be application dependent . Although we allow a generous range for the conductance parameters ( 0 . 01–299% of baseline ) , some parameters did reach the bounds when fitting the experimental data ( Fig 6 ) . Increasing the range will likely require running the GA optimization with a larger population size or for more generations , as will including additional parameters . The main computational cost of the GA is that of simulating the individual models . As this process is inherently parallel , it is straightforward to take advantage of parallel computing . Future implementations could decrease run time by utilizing a GPU , on which optimization for neuronal data has been shown to be feasible [32] . Finally , although the four cells tested in this study provide a strong proof-of-concept for the approach , to further develop the method , it could be applied to a larger number of cells . In the novel approach developed here , cell-specific cardiac models are developed by coupling complex electrophysiology protocols with genetic algorithm parameter fitting . Neither the electrophysiological data ( which are too complex to fit by hand ) , nor the fitting algorithm , would offer much advantage alone . However , merging the two enables markedly improved models that can more accurately simulate dynamically rich cardiac dynamics than can models developed using traditional approaches . Given the widespread use of ionic cardiomyocyte models in investigating arrhythmogenesis , there is utility in models that are better at reproducing such rich electrophysiological dynamics , which are more representative of the complex dynamics that are often inherent to arrhythmias . In addition to improving model fidelity generally , because this approach can be used to generate a model from a single cardiac myocyte , it may be useful in applications ranging from studying the implications of cell-to-cell variability to the prediction of intersubject differences in response to pharmacological treatment .
All animal care and handling for this study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of Weill Cornell Medical College ( protocol number: 0701-571A ) . The cardiac guinea pig model developed by Faber and Rudy ( "FR model" ) [34] was used . Model intracellular and extracellular ionic concentrations were set to the values used in our in vitro experiments ( see below ) , after which the model was simulated to a steady state in current clamp mode for 1800 beats at a pacing cycle length of 500 ms . Stimuli were square pulses of 1 ms duration and -40 A/F amplitude . Stochastic stimulation sequences were 5 s in duration . Stimulation times were randomly drawn from a uniform distribution with a range of 100–700 ms . Stimulation times for the optimization sequence were: 216 . 48 , 623 . 36 , 764 . 64 , 1101 . 12 , 1790 . 16 , 2073 . 10 , 2642 . 28 , 3183 . 10 , 3786 . 07 , 3959 . 02 , and 4579 . 72 ms ( Fig 2A ) . Prediction sequence stimulation times were: 247 . 40 , 705 . 30 , 1020 . 60 , 1347 . 90 , 1462 . 60 , 1705 . 60 , 2038 . 30 , 2546 . 70 , 3150 . 30 , 3706 . 70 , 3939 . 70 , 4077 . 80 , and 4645 . 50 ms ( S6 Fig ) . After each parameter change during the GA optimization ( details below ) , current clamp was simulated for 9 beats at a static pacing interval of 500 ms to dampen transients . The voltage clamp protocol was designed on general , a priori , IV relations for the individual channels ( e . g . , IK1 is the predominately active current at a holding potential of -120 mV ) . The 6000 ms protocol was composed of the following steps: 50 ms at -80 mV , 50 ms at -120 mV , 500 ms at -57 mV , 25 ms at -40 mV , 75 ms at +20 mV , 25 ms at -80 mV , 250 ms at +40 mV , 1900 ms at -30 mV , 750 ms at +40 mV , 1725 ms at -30 mV and 650 ms at -80 mV ( Fig 3 ) . The contribution of individual membrane currents to the voltage clamp protocol was evaluated using the FR model and the following equation: Contribution=100%⋅∑t=t1t2|Ix ( t ) |∑j=1N∑t=t1t2|Ij ( t ) | ( 1 ) Eq 1 calculates the percentage contribution of the absolute individual current ( Ix ) relative to the absolute sum of all N currents for all time points during one of the holding potentials ( t1 to t2 ) . This calculation was done for all model currents at all holding potentials . During GA optimization , the simulated multi-step voltage clamp is preceded by 5 s holding at -80 mV to allow the model to settle after parameter changes . Multiple global search heuristics have been applied to electrophysiology models , including gradient-based descent [13 , 19–21] , simulated annealing [50] , particle swarms [18 , 24] , and genetic algorithms [22 , 23 , 25] . We chose a genetic algorithm as it is effective for a range of the number of parameters [50] , is computationally simple and readily parallelizable , and has been shown to be successful at optimizing sophisticated ionic models to experimental data [22 , 23 , 25] . The GA used in this study was originally developed by Sastry [33] and was used with the settings for selection , crossover , mutation , and elitism strategy as in Bot et al . [23] . Compared to the study of Bot et al . , we increased the parameter search range to 0 . 01–299% of the baseline model values . This larger range , in combination with the increased number of parameters and the diminished requirements for computation speed relative to the Bot et al . study , caused us to enlarge the population size to 500 and raise the number of generations to 100 , based on test runs showing consistent convergence when using these values . Because the GA is inherently stochastic , it was run 10 times per optimization problem . In addition , an iterative approach was implemented , based on the study of Hobbs and Hooper [31] . For each parameter , the iterative approach uses the span of the 10 best individuals from the first 10 GA runs as the search boundaries for a second set of GA runs . With these new search ranges , the GA was again run 10 times with a population of 500 individuals for 100 generations . We used mean squared differences for the objective functions ( errors ) that the GA works to minimize: E1=∑t=tIC , starttIC , end ( Vtarget ( t ) −Vindividual ( t ) ) 2 ( 2 ) E2=∑t=tIC , starttIC , end ( Vtarget ( t ) −Vindividual ( t ) ) 2+∑t=tVC , starttVC , end ( Itarget ( t ) −Iindividual ( t ) ) 2 ( 3 ) where E1 is the objective function when only current clamp data ( i . e . , stochastic stimulation or a single action potential ) was fit , and E2 is the objective function for the combined stochastic stimulation and multi-step voltage clamp protocol . In both equations , Vtarget is the membrane potential during current clamp of the target ( i . e . , either the simulated nominal model or the experimental data ) , and Vindividual is the membrane potential of a simulated individual . Itarget and Iindividual are the current responses during voltage clamp of the target and a simulated individual , respectively . Errors are summed over the entire duration of the protocols . Although of different units , the voltage clamp and the current clamp components to E2 were simply summed into a single objective ( Eq 3 ) as we expect them to be minimal for the same range of parameters , rather than being competitive as in typical multi-objective optimization . E2 is therefore unitless . The estimated model parameters are the maximal conductances of INa , ICaL , ICaT , IK1 , IKr , IKs , IKp , and IpCa , and the maximal flux of JSERCA . Optimizations were run on a 3 . 2Ghz Intel Xeon W3670 6-core , 6GB memory , machine and took approximately 8 hours per run for the iterative approach and then combined stochastic pacing and voltage clamp protocol . Two sample t-tests were performed with a significance level of 0 . 05 . Numbers and error bars indicate average ± standard deviation . Guinea pigs ( n = 4 ) were anesthetized using an intraperitoneal injection with Euthasol ( Virbac Corporation , Fort Worth , TX ) , 550 mg/kg . Excised hearts were then Langendorff retrograde perfused , and myocytes were isolated from the base ( top 1/3 ) of the left ventricle through enzymatic digestion . Myocytes were stored in Dulbecco’s Modified Eagle Medium ( DMEM ) with 5% fetal bovine serum ( FBS ) . Amphotericin-B ( Sigma-Aldrich Corp . , St . Louis , MO; 480 μg per 1 ml pipette solution ) perforated patch clamp technique was used to record cellular action potentials . Bath solution contained ( in mmol/l ) 139 . 4 NaCl , 5 . 4 KCl , 1 . 0 MgSO4 , 10 . 0 Hepes , 10 . 0 dextrose , 2 . 0 CaCl2 , pH 7 . 35 with NaOH , osmolality 310 ± 3 mmol/kg . Intracellular solution contained ( in mmol/l unless otherwise noted ) 125 KCl , 10 NaCl , 5 . 5 dextrose , 0 . 5 MgCl2 , 11 KOH , 10 Hepes , 10 μmol/l CaCl2 , pH 7 . 1 with HCl , osmolality 295 ± 3 mmol/kg . Recordings were performed at 35°C . Patch-clamp measurements were recorded using an Axopatch 200A amplifier ( Molecular Devices , Sunnyvale , CA ) . The Real-Time eXperiment Interface [RTXI; rtxi . org; [51 , 52]] software platform was used to control the amplifier and record data . Cells were initially paced in current clamp mode at a BCL of 500 ms to steady state ( 500–1000 ) beats using suprathreshold square pulses of 1 ms duration . Next , the optimization and prediction stochastic stimulation sequences were applied . Amplifier mode was then switched to voltage clamp and series resistance measured ( 4–8 MΩ ) and compensated for ( 70–90% ) . The multi-step voltage clamp protocol was then applied in triplet . Holding potentials were corrected for a liquid junction potential of -3 mV . The magnitude of the Donnan equilibrium was estimated to 0 mV using the IV-curve of INa and therefore not corrected for . To remove stimulus artifacts from current-clamp traces , data from a 1 . 3 ms window following the start of each stimulus were excluded from the optimization . In addition , voltage-clamp data from a 1 . 2 ms window following each potential change were excluded from the GA optimization because of the capacitance transient . From the set of three voltage-clamp trials , the current response trace with the shortest time to peak INa ( step to -40 mV at 600 ms ) was selected for each cell . | Mathematical models of cardiac cell electrophysiology are widely used as predictive and illuminatory tools , but have been developed for decades using a suboptimal process . The models are typically constructed by manual adjustment of parameters to fit simple data and therefore often underperform when used to predict complex behavior such as arrhythmias . We present a novel method of model parameterization using automated optimization and dynamically rich fitting data and then demonstrate that this approach is better at finding the “real” model of a cell . Application of the method to cardiac myocytes leads to cell-specific models , which may enable well-controlled studies of both cellular- and subject-level population heterogeneity in disease propensity and response to therapies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Cell-Specific Cardiac Electrophysiology Models |
Candida albicans is a normal resident of the gastrointestinal tract and also the most prevalent fungal pathogen of humans . It last shared a common ancestor with the model yeast Saccharomyces cerevisiae over 300 million years ago . We describe a collection of 143 genetically matched strains of C . albicans , each of which has been deleted for a specific transcriptional regulator . This collection represents a large fraction of the non-essential transcription circuitry . A phenotypic profile for each mutant was developed using a screen of 55 growth conditions . The results identify the biological roles of many individual transcriptional regulators; for many , this work represents the first description of their functions . For example , a quarter of the strains showed altered colony formation , a phenotype reflecting transitions among yeast , pseudohyphal , and hyphal cell forms . These transitions , which have been closely linked to pathogenesis , have been extensively studied , yet our work nearly doubles the number of transcriptional regulators known to influence them . As a second example , nearly a quarter of the knockout strains affected sensitivity to commonly used antifungal drugs; although a few transcriptional regulators have previously been implicated in susceptibility to these drugs , our work indicates many additional mechanisms of sensitivity and resistance . Finally , our results inform how transcriptional networks evolve . Comparison with the existing S . cerevisiae data ( supplemented by additional S . cerevisiae experiments reported here ) allows the first systematic analysis of phenotypic conservation by orthologous transcriptional regulators over a large evolutionary distance . We find that , despite the many specific wiring changes documented between these species , the general phenotypes of orthologous transcriptional regulator knockouts are largely conserved . These observations support the idea that many wiring changes affect the detailed architecture of the circuit , but not its overall output .
The transcriptional networks that orchestrate gene expression are complex . Even in single-celled organisms , these networks must specify different cell types , must coordinate responses to different external cues , and must maintain homeostasis in a constantly changing environment . The evolution of such networks occurs by numerous mechanisms , including gains , losses , and modifications of transcriptional regulators and the DNA sequences they recognize ( cis-regulatory sequences ) . With over forty genomes sequenced , the ascomycete fungi are highly amenable to detailed study of regulatory network evolution . The wealth of data for the model organism Saccharomyces cerevisiae serves as a particularly strong basis of comparison . In this paper , we broadly explore transcription networks in Candida albicans , the most prevalent fungal pathogen of humans , and compare them to those in S . cerevisiae . These two organisms last shared a common ancestor some 300–900 million years ago [1] , [2] , and , in terms of coding sequences , their genomes are approximately as divergent as those of fish and humans [3] . Previous studies have documented similarities in transcriptional wiring between S . cerevisiae and C . albicans , but have also revealed major differences . In some cases it has been possible to trace plausible evolutionary pathways for these changes [4]–[6] . In these latter studies , a specific transcriptional regulator or set of target genes was typically studied in detail in both organisms . Here we analyze a much larger number of transcriptional regulators ( TRs ) in C . albicans , 143 in all . We monitored a broad spectrum of phenotypes produced when each TR was knocked out by homologous recombination , and we were thus able to interpret the phenotypes of each mutant in the context of the phenotypes observed across the whole mutant collection . Since C . albicans is a diploid organism , two rounds of gene disruption were required to produce each deletion mutant . Because unlinked mutations can occur during the knockout procedure , two or more independent knockout strains were constructed for each TR . Overall , 317 strains were created , representing 143 TRs . Each strain was carefully vetted to ensure that both copies of the appropriate gene had been eliminated . The strain collection was then assayed using 55 different growth conditions to provide an expansive set of phenotypic data . Although a smaller library of TRKOs has been constructed in C . albicans [7] , extensive phenotyping has not been reported , as the library was designed primarily for preliminary genetic screens . We believe the extensive strain collection described here , along with the phenotypic data , will be an important resource to the C . albicans community . Specific functions can now be assigned to many transcriptional regulators that were previously uncharacterized . Moreover , investigators studying various aspects of C . albicans biology , especially those that relate to issues associated with the human host ( e . g . drug resistance , morphological variation , iron acquisition ) , can immediately identify the TRs that control the process of interest and acquire the relevant set of knockout strains . Finally , as demonstrated by Nobile and Mitchell [7] , a set of TR deletion strains is a useful reagent for de novo genetic screens . Because transcriptional regulators typically control expression of many genes , this approach provides a wide “net” to capture genes involved in any process for which an assay can be devised . Additional strategies ( e . g . full-genome chromatin immunoprecipitation ) can then be used to link the transcriptional regulator to its target genes . The high level of quality control and the representation of each regulator by at least two independent knockout strains make our deletion set especially well-suited for such genetic screens . Our large set of phenotypic data , when compared with the comprehensive sets of data generated for S . cerevisiae [8]–[14] , allows for the first systematic cross-species comparison of the phenotypic conservation of TR function . To ensure the most meaningful comparison between orthologous TRs , we repeated the phenotypic studies on a subset of S . cerevisiae TR deletion strains using the same conditions that were applied to the C . albicans strains . We provide a systematic analysis of phenotypes associated with orthologous S . cerevisiae and C . albicans TRs , and , despite the numerous examples of transcriptional rewiring documented to have occurred since the species diverged , we find a high degree of phenotypic conservation .
The list of candidate genes for inclusion in our transcriptional regulator knockout ( TRKO ) library was compiled from multiple sources ( Dataset S1; [7] , [15]–[22] ) . We defined transcriptional regulators as any protein that binds DNA in and around a gene and influences its transcription rate . We placed an emphasis on proteins with sequence-specific DNA binding domains , and did not include proteins that influence the transcription of most genes in the cell ( e . g . histones , subunits of mediator , and the general transcription factors ) . To create the TRKO strains , we utilized a fusion-PCR based approach [23] that employs long stretches of flanking homology to maximize recombination ( Figure 1A; see Materials and Methods ) . C . albicans is diploid , and the construction of each knockout strain thus required two rounds of gene disruption . Although not used in this study , signature tags were incorporated into each knockout strain to enable strains to be screened in groups . Because it is not straightforward to perform back-crosses with C . albicans [24] ( and thereby ensure that a given phenotype segregates with a gene disruption ) , we created two fully independent knockout strains for each TR . This strategy greatly increases the likelihood that a phenotype observed in both strain isolates resulted from the gene knockout rather than an unrelated mutation that arose during the gene disruption procedure . This is an important consideration , as we estimate that as many as 10% of gene knockouts have additional mutations that produce at least moderate phenotypes ( see Text S2 ) . Our approach , coupled with the phenotypic screen described below , yielded high-quality TRKOs for 143 of the 184 TRs in our original list . To be classified as high-quality , two independently derived knockout strains must exhibit the same set of phenotypes . ( In some cases , additional isolates were created to resolve inter-isolate inconsistencies . ) An additional 23 TRKOs are included in our collection , but are classified as lower confidence . In some cases ( 11 TRKOs ) only a single deletion strain was obtained , and in others ( 12 TRKOs ) , the independent isolates produced overlapping but distinct phenotypes . The combined collection containing the 143 high-confidence KOs and the 23 lower-confidence KOs contains 166 KOs , represented by 365 total strains . In the following discussions , we focus on the high-confidence TRKOs . Phenotypic profiles for each TRKO were established by a large primary screen of 55 conditions augmented by a series of case-by-case supplemental screens ( Figure 1B; Materials and Methods ) . Phenotyping media were selected to probe a broad spectrum of regulatory networks . We used nutritional cues , temperature , signals that induce morphological changes , antifungal drugs , and a variety of stress conditions . When possible , drug/toxin/nutrient concentrations were calibrated such that both impairment and enhancement of growth relative to wild-type could be observed . A summary of the media utilized in this study , including commentary on their known properties ( e . g . modes of action of drugs ) , is provided as supporting information ( Text S1 ) . In the primary screen , independent isolates of each TRKO were plated as 1× and 5× dilutions on a wide range of solid media using a bolt-replicator and then photographed several times over the course of growth . These images were processed and archived using custom Java software ( Figure 1C ) and scored for growth and morphological phenotypes by comparison to a wild-type control strain included on the same plate . This approach generated over 100 , 000 individual growth and morphology scores , which were then merged – across time-points and across the knockout isolates of each TR – into single growth and morphology scores for each TRKO on each growth medium ( Dataset S2 ) . The scoring system classified the strength of the reduction or enhancement of both growth and morphology relative to wild-type ( see Text S2 and Figure 2A legend ) . Because we observed growth of all strains at two different dilutions and repeatedly over several days , we could easily score subtle phenotypes that might not have been apparent from a single concentration and time-point . We paid particular attention to colony morphology as a phenotype . On most solid media , colonies of C . albicans are composed of three types of cells: budding yeast ( round cells ) , pseudohyphae ( strings of ellipsoidal cells that remain attached to one another following cell division ) , and hyphae ( highly elongated cylindrical cells that remain attached following cell division ) . All three forms are also found in infected tissues , and the transition between these forms is key for normal pathogenesis ( reviewed by Biswas et al . [25] and Whiteway and Bachewich [26] ) . Colony morphology serves as a sensitive assay for differences in the way cells regulate the transition between the three morphological forms . By observing the collection of TRKO mutants on a variety of media over time courses of several days , we were able to identify a broad spectrum of differences in colony morphology . All images generated in this study will be made available via a Java application hosted by the Candida Genome Database ( CGD ) [16] . We have also hand-annotated a phenotypic overview of each TRKO ( Dataset S2 ) . The primary phenotypic screen identified at least one moderate phenotype for over 50% of the tested TRKOs and at least one strong phenotype for over 40% of the tested TRKOs . Many of these transcriptional regulators were completely uncharacterized , and this study presents the first direct experimental data relevant to their function . The phenotypic profiles generated by the primary screen are provided in Figure 2 . The assay conditions have been separated into the broad categories of nutrition ( Figure 2A ) , stress ( Figure 2B ) , and morphology ( Figure 2C ) , with the understanding that these categories do not have precisely defined boundaries . The color scale represents a range of phenotype strength from strong enhancement of growth or morphology ( blue circles ) to strong reduction of growth or morphology ( red circles ) . In order to highlight phenotypes that are more likely to reflect a direct role of a given TR , we scaled the diameter of the circles to reflect a “specificity score” . A high specificity score ( large diameter ) indicates that the TR deletion shows a strong phenotype under the indicated condition and an overall low level of pleiotropy ( i . e . few phenotypes overall ) relative to the other TRKOs that exhibited a phenotype under the given growth condition . This approach thus deemphasizes a highly pleiotropic TRKO ( small diameter ) if other less pleiotropic TRKOs share the phenotype in question . The calculation of the specificity score ( Text S2 ) was conducted independently for enhancement and reduction phenotypes , and only strong phenotypes were considered . A high specificity score ( large diameter circle ) serves as a visual marker for those TRs that are likely to control relatively small and discrete circuits . In other words , the TR is likely to regulate a small set of genes whose misregulation in the deletion mutant causes a restricted set of phenotypes . In contrast , a low specificity score could indicate that ( 1 ) the TR directly controls many genes involved in many different biological processes , ( 2 ) the regulator regulates one or more TRs with higher specificity scores , or ( 3 ) that the regulator directly controls a relatively small circuit but that disruption of the circuit causes many indirect phenotypes . The TRKOs with high specificity scores and strong phenotypes form the basis of much of our analysis . In the following three sections , we discuss , through specific examples , general ways in which the phenotypic profiles can be applied to problems in C . albicans biology . Although we cite specific examples as support , we do so primarily to illustrate the generality of these approaches . These three sections are followed by a more focused discussion of the regulation of C . albicans morphology . We conclude with a discussion of the evolution of transcriptional circuits based on a comparison of biological roles of orthologous regulators in C . albicans and S . cerevisiae . S . cerevisiae is a particularly well-studied eukaryotic organism , and observations made in this species have often been used as the starting point for studies in C . albicans . This approach has had mixed success; the failures often result from homologous proteins playing markedly different biological roles in the two species . Our results can help reveal the extent to which a transcriptional circuit worked out in detail in S . cerevisiae is directly applicable to the understudied species C . albicans . As an example , we discuss the collection of C . albicans mutants that affected the TOR pathway , a critical regulator of cell growth ( reviewed in [27] ) . When nutrients are abundant , the TOR pathway promotes cell growth and represses genes involved in the utilization of non-preferred nutrient sources . Conversely , when nutrients are limiting , the TOR pathway slows cell growth and redirects cellular resources to scavenge for nutrients . Although components of the TOR pathway are conserved across the eukaryotic lineage , the extent of TOR pathway conservation between C . albicans and S . cerevisiae was not known with certainty; nor was it known how additional features of C . albicans might be connected to the TOR signaling pathway . The drugs rapamycin and caffeine both inhibit function of the Tor1 kinase [28]–[31] , resulting in an artificial signal of cell starvation . In our primary screen , we assayed the deletion collection for sensitivity and resistance to caffeine to identify TRs connected to TOR function . We subsequently tested caffeine-sensitive and -resistant mutants with rapamycin and found a near-perfect correspondence ( Dataset S2 , and see below ) , providing additional support for a growing consensus that the primary mechanism of action of caffeine is interference with TOR function rather than disruption of cAMP signaling [29] . The screens identified 22 TRKOs with moderate or strong caffeine ( Figure 2B ) and rapamycin ( Dataset S2 ) phenotypes . A detailed analysis of these genes is provided in the supporting materials ( Text S3 ) , and here we emphasize four points that emerged from the C . albicans-S . cerevisiae comparison . First , the core regulatory network governing TOR function is highly conserved between the two species . Specifically , orthologs of five of the six TRs known to interact with Tor1 in S . cerevisiae [32] have strong caffeine and rapamycin phenotypes in C . albicans ( the sixth has no clear ortholog in C . albicans; see Text S3 for details ) . Second , the caffeine screen identified eight additional regulators in C . albicans that are homologous to regulators of nutritional pathways in S . cerevisiae . These results support the prevailing model that Tor1 signaling is governed by a core regulatory network with additional regulators governing specific nutritional inputs and outputs; these additional regulators appear to be largely the same in C . albicans and S . cerevisiae . Third , five of the C . albicans TRKOs with altered caffeine sensitivity also showed profound alterations in colony morphology , indicating an intimate connection between the TOR pathway and the large cell morphology network ( see below ) . It has been previously reported that rapamycin can both inhibit hyphal formation on solid medium [33] , [34] and promote flocculation and aggregation in liquid medium [33] in C . albicans [34] . Our results support this connection and further identify the transcriptional regulators likely to mediate it . Finally , although an excellent correspondence between caffeine and rapamycin phenotypes was observed , we did identify one mutant ( ΔΔorf19 . 4166 ) where this correspondence was lost: the mutant exhibits heightened sensitivity to caffeine , but not rapamycin ( Dataset S2 ) . It is possible that this TR regulates genes influencing the import , export , or degradation of caffeine but not rapamycin . Alternatively , this TR may regulate a caffeine-specific cellular target . In summary , comparing the S . cerevisiae TOR regulatory network with homologous regulators in C . albicans reveals a strong conserved core pathway that is closely connected to transcriptional circuits governing colony morphology . In general , this approach provides a rapid means of identifying core regulators in C . albicans , and in this case it indicates that most of the work on the TOR pathway in S . cerevisiae can be directly superimposed onto C . albicans . Although the TOR pathway regulators exhibit a high degree of functional conservation between S . cerevisiae and C . albicans , there are multiple examples of C . albicans homologs ( and even orthologs ) of S . cerevisiae transcriptional regulators that have different biological roles in the two species [35] , [36] . These case studies illustrate the danger in assigning biological roles to C . albicans TRs based solely on homology arguments . A comparison of the phenotypic data presented here with the extensive sets of data available for S . cerevisiae can experimentally validate homology assignments ( as for the TOR pathway discussed above ) ; it can also reveal examples of rewiring of a regulatory circuit . As an example of the latter , we consider the regulatory networks governing iron acquisition . The sources and abundance of available iron vary greatly with microenvironment , and iron-acquisition and homeostasis is a special challenge for microorganisms such as C . albicans that compete for iron in a mammalian host ( reviewed by Sutak et al . [37] ) . In Figure 3A , we have integrated the data from our phenotypic screen with data from previous studies of iron acquisition in both S . cerevisiae ( reviewed by [38] , [39] and also [9] ) and C . albicans [40]–[42] to highlight differences in the regulation of iron acquisition and homeostasis between these two species . The data from the screen is based on three growth phenotypes associated with perturbation of iron homeostasis ( Figure 3B ) . The first and most direct phenotype , sensitivity to the iron chelator bathophenanthroline disulfonate ( BPS ) , likely reflects a defect in the iron acquisition circuitry . The second phenotype , sensitivity to elevated copper levels , is linked to iron homeostasis by virtue of the strong inter-connection between copper and iron homeostasis networks: copper is a critical cofactor for high affinity iron uptake [39] . The final phenotype is sensitivity to alkaline pH . Studies in S . cerevisiae have established that copper and iron become limiting nutrients in an alkaline growth environment [43] . The phenotypic analysis provides strong support for the idea that the iron acquisition circuit has undergone a major change in regulation since S . cerevisiae and C . albicans last shared a common ancestor . As shown in Figure 3A , the circuit is positively regulated by Aft1 in S . cerevisiae and negatively regulated by Sfu1 in C . albicans ( see Text S4 ) . This is most easily seen by comparing the effects of a Sfu1 deletion in C . albicans ( Figure 3B ) with that of an Aft1 deletion in S . cerevisiae ( Dataset S3 ) . Incidentally , our results also add a new regulatory branch to the iron-acquisition model , one controlled by the transcriptional regulator SEF1 ( Figure 3 ) . Sef1 was identified in our C . albicans screen as a positive regulator of iron acquisition and , although it had not been previously reported , we found a similar role for the Sef1 from S . cerevisiae ( Dataset S3 ) . As assays for specific aspects of C . albicans pathogenesis are developed and refined , new genetic screens can be carried out using our set of deletion strains . This strategy can provide an entry point into studying a particular problem . As an example , we consider the action of two antifungal drugs . The primary screen included resistance and sensitivity to two antifungal agents , fluconazole and fenpropimorph , which block different steps of the ergosterol biosynthetic pathway [44] , [45] . We identified 34 TRKO strains with enhanced or reduced sensitivities to these drugs , only five of which ( Upc2 [46] , Ndt80 [47] , Crz1 [48] , Tac1 [49] , and Rim101 [50] ) had been previously described ( Figure 2B ) . We note an unexpected discordance between the fluconazole and fenpropimorph phenotypes in some TRKOs . In many cases resistance or sensitivity was only observed with one of the two drugs , and in a few cases resistance to one drug was accompanied by sensitivity to the other . Of the 34 TRKOs with decreased or increased drug sensitivity , eight had high specificity scores ( Figure 2B ) . Of these , only UPC2 exhibited a strong defect in growth under anaerobic growth conditions ( Dataset S2 ) , a phenotype consistent with a strong defect in ergosterol biosynthesis . We predict that the other seven TRs influence resistance/sensitivity through mechanisms other than activation of ergosterol biosynthetic pathways . Four of these seven TR knockouts – ΔΔaaf1 , ΔΔmnl1 , ΔΔorf19 . 6182 , and ΔΔorf19 . 5133 – acquire resistance to fluconazole or fenpropimorph , a phenotype that – to our knowledge – has not been previously described in either C . albicans or S . cerevisiae . Although the mechanism of this resistance is not known , several additional observations in the literature link these TRs to drug resistance . AAF1 is upregulated in response to the antifungal drug caspofungin [51] , suggesting that this TR may serve a general role in antifungal response . MNL1 has been shown to activate stress response genes [52] . ORF19 . 6182 is similar to S . cerevisiae PDR1 , a known master regulator of drug resistance [53] . For ORF19 . 5133 , the observed high-specificity fenpropimorph resistance is the first description of this regulator . Given that over 20% of the TRKOs screened affected resistance to either fluconazole or fenpropimorph , it seems clear that a large number of genomic targets , only a few of which have been previously described , can contribute to acquisition of resistance to these compounds . Although these antifungal agents have specific and focused mechanisms of action , we conclude that susceptibility to them can be influenced by perturbations of a surprisingly large number of transcriptional circuits . We regard these observations as a starting point for more exhaustive studies of these regulators . A central feature of C . albicans is its ability to grow in three distinctive morphological forms: budding yeast , pseudohyphae , and hyphae . All three forms are found at sites of infection , and the transition appears to be closely linked to pathogenesis . On solid media , C . albicans exhibits a variety of colony morphologies which reflect the transitions among these three cell forms [54] . A number of transcriptional regulators of colony morphology have been identified in C . albicans , and a subset has been extensively studied ( reviewed by Whiteway and Bachewich [26] ) . In screening the knockout library , we noticed that a significant fraction of the TRKOs ( over 25% ) , including many that had not been previously characterized , exhibited distinctive colony morphology phenotypes . Because of the importance of cell morphology to C . albicans interaction with its human host , we paid particular attention to this phenotype and its analysis . As colonies grow , different microenvironments are formed and the different cells of the colony respond accordingly , giving a progression of colony phenotypes over time . C . albicans colonies are complex structures that can be described in terms of both invasiveness and colony structure . Invasive growth – penetration into the agar surface by pseudohyphae and hyphae – was scored by examination of the colony perimeter and by observing cell retention after washing the colony from the agar surface . As colonies developed , the wild-type strain exhibited invasive growth on a variety of media . The wild-type strain also exhibited a range of colony structures , depending on the time-point and media composition . The two extremes in colony structure were “wrinkled” and “smooth” . The “wrinkled” structure was characterized by heavily ridged colonies consisting of yeast , pseudohyphal and hyphal cell types . These colonies had the consistency of rubber , likely due to extensive extracellular matrix deposition , as has been described for both C . albicans [55] and S . cerevisiae [56] , [57] . As these colonies grew , the invasion into the agar described above took place . The “smooth” colony structure was characterized by dome-shaped colonies consisting primarily of yeast cells and having a paste-like consistency , likely reflecting the absence of an extensive extracellular matrix . The primary phenotypic screen captured the progression of colony morphology across multiple days of growth ( Figure 2C ) , and was supplemented by a more detailed screen of colonies derived from single cells instead of patches ( Figure 4 ) . 28% of the TRKOs in our collection exhibited altered colony morphology in at least one growth condition . Although the data are extensive , several generalizations can be made . About half of the TRKOs with altered colony morphology showed a reduction in a morphological characteristic such as wrinkling or invasion , and the remaining half showed an enhancement of these features . The likely explanation , supported by a number of studies in the literature ( see reviews [25] , [26] ) , is that these morphological transitions are under both negative and positive transcriptional control . Indeed , TRs previously known to control morphology ( e . g . the negative regulators of filamentous growth NRG1 and TUP1 and the positive regulator TEC1 ) exhibited high specificity scores . Our screen identified 20 additional transcriptional regulators that had not previously been implicated in this network . Our results also indicate that the parameters of colony morphology can be controlled independently . For example , we observed colonies with enhanced invasion ( e . g . ΔΔorf19 . 6874; see the colony periphery in Figure 4 at 30°C on day 7 ) , colonies with wild-type levels of invasion but minimal wrinkling ( e . g . ΔΔcsr1 ) , colonies with enhanced wrinkling but no peripheral invasion ( e . g . ΔΔfgr15 ) , and colonies exhibiting neither invasion nor wrinkling ( e . g . ΔΔgat2 ) . Our results also indicate that some TRs can be assigned to specific features of colony development , while others act more broadly . For example , Gat2 and Orf19 . 4988 appear to act more generally . Deletion of either of these regulators resulted in smooth colonies with almost no invasion under all conditions tested ( Figure 2C ) . GAT2 has been previously identified as a positive regulator of colony morphology [58] , and ORF19 . 4998 is a previously uncharacterized zinc finger TR . The broad phenotypic effects of these two TRs suggest that they regulate ( perhaps together ) a core pathway governing the formation of colony wrinkling , extracellular matrix production , and invasion . In contrast , many other transcriptional regulators have more specific effects and are likely involved in the transmission of specific environmental signals . For example , ΔΔorf19 . 1685 showed a colony morphology defect only on Spider medium , and ΔΔorf19 . 2748 showed a defect only on Lee's medium ( Figure 2A ) . The former deletion strain , ΔΔorf19 . 1685 , is also deficient in the utilization of mannitol as a carbon source ( Figure 2A ) , and mannitol is the primary carbon source of Spider medium . Similarly , the latter deletion strain ( ΔΔorf19 . 2748 ) is unable to utilize proline as a nitrogen source ( Figure 2A ) , and proline is highly abundant in Lee's medium . Thus these two regulators appear to link specific cues in the environment to colony phenotype . Many other examples of TRKOs that affected colony morphology are given in Figure 2C . These results contribute to the goal of a complete description of the very large transcription circuit that controls morphological development in C . albicans . The results support a model in which a core pathway regulates the formation of a multi-cellular colony – consisting of different types of cells held together by an extracellular matrix – and is impinged upon by environmental cues to determine the overall output of the circuit . A next step in the analysis would be to determine , by full genome chromatin IP , the target genes for each of the core regulators . This analysis would reveal not only the transcriptional connections between the regulators themselves , but also the structural and enzymatic proteins that execute the program . The phenotypic analysis of the C . albicans TRKO collection provided an opportunity to systematically examine the conservation of transcriptional regulator function between C . albicans and S . cerevisiae . A few specific examples were discussed above , and in this section we examine the question more systematically . Specifically , we determined whether orthologous regulators in the two species controlled similar or different phenotypes . We use the term orthologous in its conventional sense , to indicate genes in the two species that derived from a single gene in the last common ancestor . Before proceeding , we discuss several difficulties inherent to these inter-species comparisons , and how we addressed them . First , for the comparison to be valid , the phenotypic assays compared between species must employ similar conditions and methodologies . Although several high-throughput phenotypic analyses of S . cerevisiae have been conducted ( e . g . [8] , [9] , [12] , [14] ) , the extent of concordance between these studies is sufficiently low that these data are not suitable for inter-species comparison . To enable a more meaningful comparison , we conducted a limited phenotyping of S . cerevisiae TRKO mutants ( Figure 1B , Dataset S3 ) using the same basic conditions that we employed for the C . albicans phenotyping . A second complication in phenotypic comparison is that baseline sensitivities to environmental cues ( e . g . nutrient deprivation or drug exposure ) may vary between species . Although these differences may have interesting explanations , they can result in false negatives , where the absence of phenotype in one species may simply reflect insufficient concentrations of the agent . To address this issue , our phenotypic assays of both yeasts tested a range of concentrations of agents such as caffeine , rapamycin , fluconazole , and fenpropimorph . As described in the supplemental materials ( Text S2 ) , this approach was used to select appropriate concentrations of agents for the screens . A third issue concerns confidence in the assignment of true orthologs given the gene duplications and losses that have occurred in the ascomycete lineage . In order to identify high-confidence orthologs ( as opposed to mere homologs ) in C . albicans and S . cerevisiae , we employed a combination of two different algorithms , SYNERGY [59] and INPARANOID [60] supplemented by case-by-case orthology assignments ( Dataset S1; described in Text S2 ) . For our comparison , we considered only ortholog pairs that: ( 1 ) produced a strong knockout phenotype in at least one of the two species , and ( 2 ) had been reliably assayed on the medium of interest in both species . These criteria produced a set of 24 1-to-1 orthologs for further analyses ( Figure 5A ) . The results show that most TRs with clear orthology between C . albicans and S . cerevisiae exhibit the same basic phenotype upon deletion . The conserved phenotypes ranged from the specific , such as impaired utilization of a nitrogen source or sensitivity to EDTA , to less defined phenotypes such as strongly impaired growth on rich medium ( see Text S5 for details ) . Of the 24 pairs included in the analysis , we identified 11 cases of clear phenotypic conservation and an additional 8 cases where primary phenotype ( s ) were present in both species but where one or more additional phenotypes were exhibited by one species but not the other . Despite the trend toward similar phenotypes produced by orthologous TRKOs , we did find exceptions , which likely reveal instances of major network rewiring . In particular , we found five cases in which a TRKO phenotype was evident in only one of the two species . One of these TRs , GAL4 , has been previously described as a case of network rewiring [35] . S . cerevisiae mutants deleted for GAL4 are unable to use galactose as a carbon source , but deletion of the C . albicans GAL4 ortholog does not produce this phenotype ( [35] and Dataset S2 ) . A second example is seen with the regulator RTG1 . Deletion of this regulator in S . cerevisiae results in glutamate and aspartate auxotrophies [61] , yet deletion of the C . albicans ortholog does not . Although the 1-to-1 orthology between these genes is not entirely certain ( a 2-to-1 relationship may exist , with EDS1 included as a second ortholog in S . cerevisiae ) , this regulator appears to have undergone either an acquisition or loss of metabolic regulatory function since C . albicans and S . cerevisiae shared a common ancestor . It is of course possible that C . albicans has a redundant regulator that masks the true role of RTG1; however , this would still indicate that a rewiring event had occurred . Three additional differences , each suggestive of network rewiring , are listed in Figure 5A . The data can also be used to address phenotypic conservation for orthology relationships more complex than a simple 1-to-1 . Because the gene pairs that arose from the whole genome duplication ( WGD ) in the S . cerevisiae branch of the ascomycetes have been carefully curated [62] , it is also possible to identify with high confidence the 1-to-2 ( C . albicans to S . cerevisiae ) orthologous relationships that arose from this event . Such a comparison allows us to ask whether zero , one , or both of the two S . cerevisiae duplicates have the same overall role as the single gene in C . albicans . We analyzed eight high-confidence 1-to-2 orthologous relationships , and found several patterns of conservation ( Figure 5B ) . First , we observed cases ( exemplified by C . albicans SKN7 and the two S . cerevisiae orthologs SKN7 and HMS2 ) where the likely ancestral function was preserved in one S . cerevisiae gene but apparently lost in the other . Deletion of C . albicans SKN7 and S . cerevisiae SKN7 both result in sensitivity to oxidative stress , whereas deletion of S . cerevisiae HMS2 does not . Thus , HMS2 appears to have diverged ( at least in the phenotypes its deletion produces ) from the ancestral gene . A second type of relationship is seen with the S . cerevisiae MET31 and MET32 genes relative to the single C . albicans ortholog , ORF19 . 1757 . Deletion of either MET31 or MET32 from S . cerevisiae reveals no major phenotypes , whereas the double deletion produces a methionine auxotrophy . In our screen , deletion of C . albicans ORF19 . 1757 does not produce a methionine auxotrophy or any other tested phenotype . Thus , it is likely that either the C . albicans gene or the S . cerevisiae genes retain the ancestral function and the function has changed in the other species . A third scenario is exemplified by comparison of the C . albicans gene ORF19 . 5026 with the two S . cerevisiae orthologs YML081w and RSF2 . Of these three genes , only YML081w exhibited a phenotype ( impaired growth in rich medium ) under the range of conditions tested . While far from complete , our data represent a first step towards a systematic approach to the analysis of the phenotypic output of regulatory networks in divergent species . We found an overall conservation of phenotypic output in the majority of clear 1-to-1 orthologs , but also noted several differences . Even with the small number of 1-to-2 orthologs , we observed several different phenotypic relationships , indicating that there is likely no stereotypical pattern; instead , each case must be individually explored by experiment . We are aware that the set of high-confidence orthologs is biased against orthologs that have diverged to the extent that their assignment becomes ambiguous . Nonetheless , even our high-confidence orthologs exhibit considerable divergence , and yet the phenotypic outputs are largely conserved . Given the significant rewiring of transcriptional networks documented in the fungal lineages [4] , [5] , [35] , [63] , [64] , the high degree of phenotypic conservation we observed between S . cerevisiae and C . albicans orthologs may seem unexpected . However , we know that transcriptional rewiring can take place without losing an ancestral connection between a transcriptional regulator and a process . For example , the mating circuitry between C . albicans and S . cerevisiae has undergone extensive evolutionary rewiring [5] , but the same ( orthologous ) regulators still govern mating in both species . Likewise , the transcriptional regulator STE12 controls the pheromone response in S . cerevisiae and likely also in C . albicans , yet the direct target genes of STE12 , as well as the pheromone response itself , differs significantly between the two yeasts [64]–[66] . Our results indicate that despite the rewiring that has taken place , the overall function of transcriptional regulators ( defined broadly by the phenotypes caused by their deletion ) often remains preserved from the common ancestor .
The C . albicans deletion library will be made available through the Fungal Genetics Stock Center ( http://www . fgsc . net/ ) . All C . albicans deletion strains were constructed in strain SN152 using auxotrophic marker cassettes targeted with long-flanking homology , as previously described [23] . All deletions were verified by diagnostic PCR of the flanks surrounding the introduced markers . The absence of the gene targeted for deletion was further verified by attempting to amplify a small internal fragment of the ORF . For a successful deletion , this intra-ORF PCR yielded no product while a wild-type control yielded a strong product . The strain background of the deletion strains was arg4Δ/arg4Δ , leu2Δ/leu2Δ , his1Δ/his1Δ , URA3/ura3Δ , IRO1/iro1Δ , with HIS1 and LEU2 function restored by the auxotrophic marker introduced at the targeted transcriptional regulator . A ‘wild-type’ control strain was created by reintroduction of a single allele of HIS1 and LEU2 ( amplified from the C . albicans strain SC5314 ) into the parent strain . Composition of the media used for phenotyping is described in Text S1 . All S . cerevisiae deletion strains were obtained from the Saccharomyces Genome Deletion Project collection [11] . All S . cerevisiae deletion strains were from the homozygous deletion collection ( MATa/α his3Δ1/his3Δ1 , leu2Δ0/leu2Δ0 , lys2Δ0/LYS2 , MET15/met15Δ0 , ura3Δ0/ura3Δ0 ) , with the exception of the Δsko1 strain , which was haploid ( MATa his3Δ1 , leu2Δ0 , met15Δ0 , ura3Δ0 ) . In all cases where the S . cerevisiae strain exhibited a phenotype that appeared divergent from the C . albicans ortholog ( s ) , the S . cerevisiae strain deletion was validated using the primers suggested by the Saccharomyces Genome Deletion Project protocols . The primary phenotypic screen assayed the growth and colony morphology of two or more independent isolates of each deletion strain on a variety of media . Strains were streaked from frozen glycerol stocks to YEPD medium and incubated overnight at 30°C . In the morning , the strains were thinly re-streaked to new YEPD plates and incubated an additional 4–6h at 30°C . This second growth period was included to ensure that the majority of cells were actively growing . Cells from each strain were then diluted in water to an OD600 of 0 . 080 ( ‘1×’ dilution ) and transferred to a 96-well plate where an additional ‘5×’ dilution was made in the neighboring column of the plate . This format allowed the plating of 24 ‘1×’ and ‘5×’ strain dilutions with a 48-pin bolt replicator ( V&P Scientific; VP 404A ) to each of the assay plates . At least one wild-type control strain was included with each batch of 24 strains . The plates were incubated and photographed over the course of a week using a Nikon CoolPix 4300 camera . The days photographed varied among the media types , depending upon the growth rate and the emergence of colony morphologies . All images were imported into custom Java viewing software for subsequent scoring and analysis . A detailed explanation of the criteria used for scoring phenotypes and the algorithm used for compiling the data is provided in Text S2 . The phenotyping methodology used in the primary screen was modified in some of the follow-up phenotyping screens ( see Text S2 for details ) . | A key goal in the understanding of the biology of an organism is the description of the regulatory networks that control the expression of its genes . Changes in gene expression result in new cellular phenotypes that can be acted upon by evolutionary forces to influence the configuration of these networks . We have developed a phenotypic description of the transcriptional regulatory networks of the major fungal pathogen of humans , Candida albicans , by individually deleting genes that encode transcriptional regulators and observing the resulting phenotypes in a variety of environmental conditions . This approach provides insight into the biological roles of many previously uncharacterized regulators , and allows us to assign groups of regulators to specific biological roles , many of which are relevant to pathogenesis . For example , we identified groups of regulators that influenced sensitivity to antifungal drugs , the ability to acquire iron ( a challenge for organisms in a human host ) , and the ability to form complex multi-cellular colonies . Our results also allow us to analyze how the phenotypes associated with transcriptional regulators change as organisms diverge . A comparison of C . albicans data with that from the well-characterized yeast S . cerevisiae revealed strong phenotypic conservation between related transcriptional regulators , despite the more than 300 million years which separate the species . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/gene",
"expression",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"infectious",
"diseases/fungal",
"infections",
"genetics",
"and",
"genomics/gene",
"function",
"evolutionary",
"biology/morphogenesis",
"and",
"cell",
"biology"
] | 2009 | A Phenotypic Profile of the Candida albicans Regulatory Network |
To better understand genome regulation , it is important to uncover the role of transcription factors in the process of chromatin structure establishment and maintenance . Here we present a data-driven approach to systematically characterise transcription factors that are relevant for this process . Our method uses a linear mixed modelling approach to combine datasets of transcription factor binding motif enrichments in open chromatin and gene expression across the same set of cell lines . Applying this approach to the ENCODE dataset , we confirm already known and imply numerous novel transcription factors that play a role in the establishment or maintenance of open chromatin . In particular , our approach rediscovers many factors that have been annotated as pioneer factors .
In higher eukaryotes , certain sequence-specific transcription factors ( TFs ) , which we will call chromatin accessibility regulators ( CARs ) , are responsible for establishing and maintaining open chromatin configurations [1 , 2] . CARs therefore play a fundamental role in transcriptional regulation , because open chromatin configurations are necessary for additional TFs to bind and transcriptionally activate target genes . CARs that can bind closed chromatin and open up chromatin are called pioneer TFs [3] . The comprehensive identification of pioneer TFs with high confidence still needs further research . While some pioneer TFs are well studied , others have only preliminary evidence , or are only computationally predicted . Some well studied examples include FOXA1 , whose winged helix domains disrupt DNA–histone contacts , and POU5F1 , SOX2 and KLF4 , which are used in production of induced pluripotent stem cells ( iPSC ) [4 , 5] . Further pioneer TFs such as ASCL1 , SPI1 and the GATA factors are used in transdifferentiation , and PAX7 plays a role in pituitary melanotrope development [5–7] . However , not all pioneer TFs are involved in development and cell type conversions: the CLOCK-BMAL1 heterodimer is part of the circadian clock and the tumour suppressor TP53 is involved in the cell cycle , while its close homolog TP63 is involved in skin development [8–10] . Recent studies suggest that maintaining open chromatin is a dynamic process with pioneer and other TFs binding and unbinding rapidly and continually recruiting additional chromatin remodelling factors that are not sequence specific [2 , 11 , 12] . TFs vary in their ability to recruit particular remodelling factors , for example the TFs STAT5A/B and MYOG motifs enrich in binding sites of the SWI/SNF remodelling complex but not in ISWI remodelling complex binding sites , whereas YY1 motifs were found exclusively in ISWI complex binding sites [2] . A natural question then is which TFs are relevant to maintain open chromatin and can therefore be called CARs . One approach to test whether a given TF is a CAR is to perform a knock-down of this TF followed by an open chromatin assay to see whether chromatin regions containing the respective motif preferentially change from open to closed [13] . However , this approach is very time consuming because it requires a separate knock-down experiment for each TF . To define pioneer TFs specifically , one can check if the TF has the ability to bind nucleosomal DNA in vitro and validate the results in vivo [14] . Recently , a computational method called Protein interaction Quantification ( PIQ ) has been published that aims to recover pioneer TFs by estimating both TF binding and ensuing chromatin changes from the same Dnase1 hypersensitivity ( DHS ) experiments [15] . However , PIQ did not predict some well known pioneer TFs such as FOXA1 , SOX2 and POU5F1 showing that further improvements are possible [3] . Here we introduce a data driven approach to predict CARs . Our approach relies on the joint analysis of a large collection of DHS and coordinated gene expression data to estimate TF activity independently of DHS data . We first define the motif accessibility score for a given TF for each cell line based on the enrichment of its binding motif in regions with open chromatin . We then associate these scores with gene expression values across all available cell lines . This should allow us to predict which factors have a role either in establishment or maintenance of open chromatin , although it will not reveal which mode predominates ( to determine this , further experiments will be necessary ) . We used our approach on data generated as part of the ENCODE project [16 , 17] . This uncovered numerous TFs whose motif accessibility is robustly associated with mRNA expression across 109 cell lines suggesting either a role in the establishment or maintenance of open chromatin . Also , we see that our uncovered TFs are strongly enriched for known pioneer TFs . This suggests that the TFs we identified are good candidates for CARs .
Our approach rests on the assumption that the activity of a CAR is correlated with the amount of open chromatin in the vicinity of its potential binding sites . Both quantities can be estimated from genomic data: For the CAR activity we use its gene expression level as a proxy for the active protein concentration . The effect of this activity is approximated by the open chromatin fraction of the genome around its binding motif instances ( Fig 1 ) . Specifically , we count the number of instances of the binding motif of a given TF in the open chromatin fraction of the genome to define a motif accessibility score . A naive approach would be to use standard linear regression between the motif accessibility score and the expression level of a given TF to identify CAR candidates . Yet , this method has an elevated type I error rate , as it does not account for confounding due to cell line relatedness or batch effects . To overcome this limitation , we use here a linear mixed model ( LMM ) framework , where a random effect accounts for such confounding factors ( which has been shown to work well in genetic association studies [18–20] ) . For a given motif , we use the linear mixed model framework to find the association p-value between its accessibility score and the measured expression of the TF gene . We then compare this p-value to the p-values calculated using the measured expression of each of the other genes as regressors . If confounding is controlled for , most association p-values should follow a uniform [0 , 1] distribution . Furthermore , if the TF is a CAR , its p-value should be low compared to other genes . We thus define the CAR rank of a TF as the rank of its association p-value among all genes ( see example in Fig 1 ) . Low CAR ranks indicate strong association between motif accessibility and TF expression , suggesting that the TF is a CAR . Specifically , we used DHS data as well as mRNA expression data across 109 cell lines . To calculate motif accessibility scores we used 325 TF binding motifs from the HOCOMOCO database [21] . As expected , we observed severe confounding when using standard linear regression , which was controlled using linear mixed effect model regression ( Fig 2 ) . Our method relies on TF motif accessibility and expression data to predict CARs . However , evolutionarily related TFs have similar binding motifs [23] . Motif accessibility may therefore associate not only with the expression of the annotated TF , but also with the expression of a homologous TF with a similar motif . Therefore , we mapped TFs into subfamilies using the homology-based clustering TFClass [24] . The 1 , 557 TFs were grouped into 397 subfamilies . Using a collection of 329 ChIP-seq profiles from ENCODE , we saw strong enrichment of TF motifs in ChIP-seq peaks of the TF as well as its subfamily members ( Fig 3 ) . We therefore consider any strong association between a motif and a member of the subfamily of its TF as a signal for a CAR . Next , we used the linear mixed model strategy to predict CARs among TFs . We used 325 motifs from HOCOMOCO ( after filtering motifs showing low overlap with DHS signal , see Methods ) . For each motif , we used a linear mixed effect model to compute its association with mRNA expression for 1 , 188 known TFs . Due to the redundancy of motifs within the same TF subfamily ( see preceding section ) , we also computed CAR ranks at the level of TF subfamilies . To this end , we retained the most significant association p-value within each subfamily corrected for subfamily size ( see Methods and S2 Fig ) . Under the null model ( when TFs are not CARs ) , CAR ranks should be uniformly distributed across all subfamilies , so that deviation from uniformity indicates presence of CARs . We found strong enrichment of low CAR ranks at the subfamily level ( Fig 4 , S1 Table ) . The enrichment was stronger when using mixed modelling instead of standard linear regression , underlining again the importance of proper control for confounding factors . When looking at the threshold that leads to 10-fold enrichment of low CAR ranks compared to uniformity ( i . e . , 10% false discovery rate ) , we found that 25% of all subfamilies have a CAR rank that falls below that threshold . These results show that many TFs do have an impact on the open chromatin fraction and can be defined as CARs . To validate our results based on the ENCODE dataset , we applied our CAR calling strategy to data from another large scale effort , the ROADMAP Epigenomics consortium [26] . Coordinated open chromatin and expression data have been released for 56 samples . For 29 of these samples , open chromatin was assayed directly . For the other samples , open chromatin information was imputed from other available epigenetic measurements . The ROADMAP collection is derived mainly from human tissue samples and primary cell lines ( whereas ENCODE is biased towards immortalized cell lines ) . Further differences are that expression was measured using RNA sequencing . We applied our method to these datasets and compared results to the results derived in ENCODE . Most subfamilies predicted to be CARs in ROADMAP were recovered in ENCODE ( see S3 Fig ) . Furthermore , while subfamilies predicted to be CARs in ENCODE showed enrichment for low CAR ranks in ROADMAP , subfamilies not predicted to be CARs in ENCODE did not show enrichment for low CAR ranks in ROADMAP ( see S4 Fig ) . These results are concordant with both datasets , pointing toward the same factors being CARs and the higher power of the ENCODE data to detect CARs , potentially due to higher sample size , reliance on direct measurements of DHS and lower fraction of complex tissue samples . To evaluate the impact of the motif search strategy , we investigated the robustness of the pipeline with respect to the motif search . Results were stable and power was only affected by varying motif cutoffs ( S5 Fig , S6 Fig ) . Additionally , we investigated whether choosing the cutoff based on ChIP-seq data changed results . For each TF with available ChIP-seq data , we used an individual cutoff such that all called binding sites have fixed true positive rate ( using the ChIP-seq data as the ground truth ) . Again , results were stable no matter how the cutoff was assigned ( S7 Fig and S8 Fig ) . As mentioned above , one well-defined class of CARs are pioneer TFs that can bind and open closed chromatin . Therefore , subfamilies annotated to known pioneer TFs should have low CAR ranks . To test enrichment formally , we used a recently published list of established pioneer TF subfamilies ( Methods ) [3] . We asked whether these subfamilies were predicted as CARs using our methodology . For eight subfamilies in the list for which we had the motif , six showed at least ten-fold enrichment ( i . e . having a CAR rank at the subfamily level below ten ) ( Fig 5 ) . To assess significance , we used the Wilcoxon ranksum test leading to a p-value of 0 . 0087 . When using the hypergeometric test with 10-fold enrichment cutoff ( Fig 4 ) , the p-value was even lower ( P = 0 . 0016 ) . Because our approach to uncover CARs is biased towards TFs with large mRNA expression variability ( S9 Fig ) , we sought to control for potential confounding introduced by the fact that the tested pioneer factors might also have large expression variability . Controlling for expression variability only slightly increased the p-values from 0 . 0087 to 0 . 024 and from 0 . 0016 to 0 . 0027 , respectively . It is known that the activity of some TFs is mainly regulated by the level of their cofactors rather than their own protein concentration [27] . These TFs are often present in their inactive form in the cell , which can then be quickly activated upon binding of the cofactor . This allows the cell to rapidly respond to environmental cues . An example of this phenomenon are steroid receptor TFs , which initiate transcriptional changes upon steroid hormone binding [28] . In such cases , one would not expect a strong association between the mRNA expression level of a receptor TF and its motif accessibility because mRNA expression would rather be correlated to the amounts of inactive TF protein in the cell , while TF activity should depend on the strength of the environmental stimulus . However , if the TF strongly activates mRNA expression of other genes , it might be possible to predict whether the TF is a chromatin accessibility regulator by looking at associations between the motif accessibility of the TF and the expression of its downstream genes . To explore this strategy , we looked at associations across all genes and motifs that were below the overall Bonferroni threshold ( 9 . 6 x 10−9 ) . For five out of 13 such motifs , members of the corresponding subfamily had top scores . In four further cases , a gene from a TF subfamily was ranked close to the top that was highly related ( i . e . part of the same family [24] ) to the motifs’ corresponding subfamily but not identical with it . This suggests that the TF subfamily clustering was too fine-grained in these cases . Surprisingly , for one motif , the significant association had a negative effect size ( the negative association was observed between NUDT11 and the motif for RARG ) , which might reflect an indirect effect . The remaining three motifs were all annotated to the GR-like receptors , which encompass four TFs ( AR , NR3C1 , NR3C2 , PGR ) . The accessibilities of these three motifs all associated strongly with the expression of three genes ( FKBP5 , ZBTB1 , TSC22D3 ) . When using the STRING database to check for functional links between these genes , all genes had links to a GR-like receptor ( Fig 6 ) [29] . In fact , all three genes are known to be glucocorticoid response genes . These results suggest that some GR-like receptors might act as a CAR . For strongly activating factors , the power of the analysis can therefore be strengthened by incorporating results from downstream genes .
It is well known that TF binding correlates with open chromatin [17] . However , for many TFs , it is not clear whether their binding is the cause or the consequence of open chromatin . Here , we used datasets provided by ENCODE to predict chromatin accessibility regulator candidates , i . e . , TFs that are able to establish or maintain open chromatin configurations . We devised an approach using linear mixed models to deal with the extensive confounding that one encounters in genome-wide data from heterogeneous sources . Our method uncovers a set of TFs whose expression is associated with their motif accessibility , suggesting a role in maintenance of an open chromatin configuration . Potentially our methodology could be extended to histone modification data instead of DHS data . We applied our method to H3K4me3 data for cell-lines but did not see strong enrichment ( S10 Fig ) . Because pioneer TFs are by definition CARs , our predictions should be enriched for known pioneer TFs . We tested this formally for a list of pioneer TF subfamilies recently published by Iwafuchi-Doi et al . [3] . Six out of eight pioneer subfamilies were indeed predicted by our method to be CARs: FOXA1 , GATA6 , KLF4 , SOX2 , SPI1 and TP63 were the pioneer TFs driving these signals . The two subfamilies not predicted to be CARs were POU5 and CLOCK . SOX2 was the gene most strongly associated with POU5F1 motif accessibility with a low p-value of 5 x 10−6 ( S11 Fig ) . POU5F1 acts together with SOX2 to maintain undifferentiated states [30] . The two TFs also physically interact and a recent study proposed a model where SOX2 guides POU5F1 to target sites [31] . The CLOCK subfamily members have a role in the cell cycle , acting as TFs for the circadian pacemakers [32] . It is possible that average mRNA expression of these TFs in unsynchronized cell lines is not a meaningful measure for their activity . In addition to the eight aforementioned factors we found further factors discussed in the pioneer TF literature such as TFAP2C , EBF1 , CEBPD/B , OTX2 , NFKB and STAT5 ( Table 1 ) [22 , 33–37] . In addition , when combining our predictions with those from the PIQ method[15] , we observed substantial performance improvement compared to either method alone ( S12 Fig ) . One limitation of our approach is that it cannot discern between open chromatin establishing TFs and open chromatin maintaining TFs . A way to discern the relative roles could be to perform overexpression and knock-down experiments followed by an open chromatin assay for the TFs found by our approach . While this is out of the scope for the current study , we hope that our method can help in prioritizing such experimental efforts . Further , by its very nature , our methodology cannot with certainty resolve between TFs that belong to the same sub-family . It shares this weakness with almost any method relying on TF motifs . The procedure associates the expression values of each TF separately to the motif accessibilities and one strong association is enough to lead to low CAR ranks for the subfamily . The TF in the subfamily whose expression is the most strongly associated to one of the subfamily motif is naturally also the strongest candidate for CAR activity . ( This information is given in Table 1 as well as in S1 Table ) . However , if the expression values of the subfamily members are also strongly correlated , we cannot be sure which ones are driving the association . It is also clear that multiple conditions have to be met for the approach to work . First and foremost , mRNA expression has to be correlated sufficiently with protein concentration of the CAR . Typically , only a fraction of the variation in protein concentration can be explained by variation in mRNA abundances [39] . Nevertheless , better power of our approach can always be achieved by increasing sample size , as long as there is at least some correlation . Further , it is reasonable to assume that our approach will perform better on TFs with a large dynamic range across cell types . This seems indeed to be the case , since most TFs predicted to be CARs tend to have large mRNA expression variance ( S9 Fig ) . Sampling more and diverse cell lines could address this issue , because it should increase the dynamic range . This restriction would also suggest that our approach is biased against cell type specific TFs . However , when looking at tissue expression patterns ( www . gtexportal . org [40] ) of the predicted CARs , we found both: TFs that showed expression in a large proportion of cell lines such as EBF1 and STAT5B as well as quite specific TFs . Examples of specific CARs are SPI1 , which only showed expression in whole blood , and OTX2 , which only showed expression in some brain regions . It is possible that the use of immortalized cell lines leads to larger gene expression variability in the sample facilitating the detection of such tissue-specific CARs . For some TFs , activity mainly depends on cofactors . For example , for steroid hormone receptors , hormone molecules activate a pool of inactive TF already present in the cell . In such cases measuring TF activity with gene expression measures can be misleading and one would not expect an association between the expression of a TF and the accessibility of its motif . For example , for the accessibility score of NR3C1 , we saw much stronger associations with the expression levels of a small set of glucocorticoid response genes ( ZBTB16 , FKBP5 , TSC22D3 ) than that of NRC1 itself [41–43] . This difference in signal strength is in line with the activity of NR3C1 being mainly regulated by glucocorticoid binding and not NR3C1 gene expression levels . Of note , NR3C1 was reported to have pioneer activity [1] . In summary , we exploited the rich data source of ENCODE to find TFs whose mRNA expression levels are directly linked to the open chromatin fraction of the genome . Although our approach in its current form is able to find TFs with strong associations , it is also clear that increasing power by adding more cell lines would find more TFs with an association . From the current data , we would estimate that at least 25% of TF subfamilies show a low CAR rank at the subfamily level , suggesting that the regulation of chromatin accessibility is a pervasive phenomenon amongst TFs .
Annotated open chromatin ( FDR <0 . 01 ) peaks were downloaded from the EBI website ( see URL section ) and trimmed to the top 90 , 000 peaks for each cell line . 426 motifs were downloaded from the HOCOMOCO website and aligned to the reference genome with FIMO [21 , 44] . Motif occurrences with a p-value below 10−5 were kept for processing . For each motif , we counted the number of DHS peaks overlapping a motif instance in a given cell line using bedops [45] . Results were filtered to motifs that were present in at least 150 DHS peaks on average , leaving 344 motifs . For a given motif , we quantile-normalized the values to follow a normal distribution yielding the raw motif-activity matrix with rows corresponding to motifs and columns corresponding to cell lines . The resulting matrix was iteratively scaled to zero mean and unit standard deviation , first row-wise ( across cell lines ) then column-wise , until convergence [46 , 47] . Next , we saw that the cell-line wise covariance matrix had a very large first eigenvalue , with a corresponding eigenvector that did not track well the different tissue origins of the various cell lines . Assuming that this leading principal component largely captured batch effects , we chose to regress out the first eigenvector from each row of the matrix , leading to better agreement between expression and motif accessibility correlation matrices ( S13 Fig ) . After this step , we quantile-normalized the data per motif to follow a normal distribution to ensure that the assumptions of the applied statistical model were met . To map motifs to TFs and TF subfamilies , we used the TfClass hierarchy [24] . Of the 344 tested motifs , we mapped 330 to a TF and its subfamily . Of these , 325 had expression data available for a subfamily member . We downloaded raw expression microarray data from the GEO repository ( GSE1909 and GSE15805 ) . ( ENCODE micro-array data was used instead of RNA-seq because to-date more cell lines with DHS information have also RNA expression measured by micro-array than RNA-seq ) . We background corrected and normalized using the RMA-algorithm implemented in the oligo package to process all arrays for which DHS data was also available [48 , 49] . Only the core set data was used . The data were summarized to gene level [50] . Only results that had a one-to-one mapping between genes and gene probesets were kept . 15 , 119 genes could be annotated in this fashion . Because for many cell lines more than one experiment was conducted , we summarized multiple plates by averaging gene results across experiments . The resulting matrix was iteratively scaled to zero mean and unit standard deviation , first row-wise ( across cell lines ) then column-wise , until convergence [46 , 47] . The model proposed is y=xiβi+δi+εi . Where y is a vector of motif accessibility scores across n cell lines , xi is the expression vector of gene , i , βi is the effect size of gene i: εi∼Nn ( 0 , σr2In ) and δi∼Nn ( 0 , σe2Ce ) . Ce is the covariance matrix of the n x p expression matrix: Ce=1p∑i=1pxixiT . For each gene i , βi , σr , and σe are estimated via maximum likelihood and the null hypothesis βi = 0 is tested via a likelihood ratio test [18 , 20] . More details on this procedure are given in S1 Appendix . For each motif in HOCOMOCO , we used the mixed model association results across all 1 , 188 known TFs for which we had mRNA expression data [21] . This yielded a matrix of association p-values for all pairs of 325 motifs ( belonging to 147 TFClass subfamilies ) and 1188 TFs ( belonging to 368 TFClass subfamilies ) . Due to the fact that homologous TFs have similar binding motifs , we sought to aggregate results into CAR ranks at the subfamily level ( S1 Fig ) . To achieve this , we reduced the 325 x 1188 motif-TF association matrix to a 147 x 368 matrix of associations between motif subfamilies and TF subfamilies . In practice , for each motif subfamily-TF subfamily pair we collected the most significant p-value among all motif-TF pairs in these subfamilies and multiplied it with the total number of such motif-TF pairs to correct for subfamily size . Finally , for each motif subfamily , we ranked the adjusted p-values across all TF subfamilies and defined its CAR rank as the rank of its corresponding TF subfamily . To get an external annotation of pioneer factors , we used a recently published list of established and predicted pioneer factors ( Table 1 in Iwafuchi-Doi et al . [3] ) . We used a hypergeometric test at the 10-fold enrichment cut-off ( Fig 4 ) , as well as a ranksum enrichment test . To derive a ranksum statistic , we summed the CAR ranks of the eight subfamilies annotated as pioneers . To assess significance of this statistic , we used permutation tests: For each of the 50 , 000 permutation samples , we picked eight CAR ranks from the set of subfamilies not annotated as pioneers and summed them to derive 50 , 000 permutation sample statistics . The p-value was approximated as the fraction of permutation sample statistics of greater or equal size as the statistic derived for the annotated pioneers . To control pioneer enrichment for mRNA expression variation , we first calculated the expression variance of each TF across all cell lines . The distribution of variance values was transformed to follow a standard normal distribution . We then used the maximal expression variance observed for any TF in each subfamily . To assess significance , we used permutation tests: we sampled eight non-pioneer subfamily level CAR ranks 50 , 000 times . However , subfamilies were not sampled uniformly: We sampled four non-pioneer subfamilies with maximal expression variance between the 0th and the 50th quantile of the eight pioneer subfamilies , and four non-pioneer subfamilies with maximal expression variance between the 50th and the 100th quantile of the eight pioneer subfamilies . RNA-seq data were downloaded from the ROADMAP website ( see section ‘URLS’ ) for 56 cell lines . We used only genes with average read count above 50 , which removed 12% of genes . The number of reads plus a pseudo-count of one to were log-transformed . Samples were then quantile normalized to the average mean distribution [51] . The resulting matrix was iteratively scaled to zero mean and unit standard deviation , first row-wise ( across cell lines ) then column-wise , until convergence [46 , 47] . To derive motif accessibility scores , imputed DHS data were downloaded for 56 cell lines from the ROADMAP website ( see section ‘URLs’ ) . From these datasets motif accessibility scores were derived in the same fashion as for the ENCODE DHS data . To derive CAR ranks , the same strategy was employed as for the ENCODE dataset . To compare fixed motif cutoffs to a variable motif cutoff guided by ChIP-seq , the following procedure was used . ChIP-seq data from the Myers and Snyder lab in the ENCODE collection for which dnase1 and expression data were available were downloaded and each ChiP-seq experiment was mapped to a dnase1 experiment based on cell line and to the motif of the TF , yielding mappings to 75 motifs ( belonging to 50 subfamilies ) . For a given motif and cell line pair for which ChIP-seq data ( as well as DHS data ) was available , each DHS region was annotated with the p-value of its most significant motif instance ( given that they contained a motif with p-value below 5x10-5 ) as well as whether it overlapped with a ChIP-seq peak . The motif p-value cutoff was defined such that a fixed fraction of peaks with motifs below that cutoff would validate in the ChIP-seq experiment . Three true positive rates were chosen for this comparison 0 . 3 , 0 . 5 and 0 . 7 ( see S7 Fig , S8 Fig ) . Only experiments were used for which it was possible to choose a motif cutoff such that the highest validation rate ( i . e . 0 . 7 ) could be reached . If multiple ChIP-seq experiments were available per motif , the median p-value cutoff was chosen for each validation rate . We compared these strategies using a fixed cutoff for all motifs of 10−5 , which was used throughout the rest of the paper . Results obtained are similar when using ChIP-seq guided cutoffs or fixed cutoffs . Code for reproduction ( including scripts for data download ) is available at: https://github . com/dlampart/csrproject ENCODE DHS peaks were downloaded from: http://ftp . ebi . ac . uk/pub/databases/ensembl/encode/integration_data_jan2011/byDataType/openchrom/jan2011/fdrPeaks/ ROADMAP expression data were downloaded from: http://egg2 . wustl . edu/roadmap/data/byDataType/rna/expression/57epigenomes . N . pc . gz ROADMAP imputed DHS peaks were downloaded from: http://egg2 . wustl . edu/roadmap/data/byFileType/peaks/consolidatedImputed/narrowPeak/ ENCODE histone files were downloaded from: ftp://hgdownload . cse . ucsc . edu/goldenPath/hg19/encodeDCC/wgEncodeUwHistone/ | Transcription factor binding occurs mainly in regions of open chromatin . For many transcription factors , it is unclear whether binding is the cause or the consequence of open chromatin . Here , we used datasets on open chromatin and gene expression provided by the ENCODE project to predict which transcription factors drive transitions between open and closed states . A signature of such a factor is that its expression values are correlated to chromatin accessibility at its motif across the same set of cell lines . Our method assesses this correlation while accounting for the fact that some tested cell lines are more related than others . We find many transcription factors showing evidence of driving transitions and a high proportion of these transcription factors are known pioneer factors , i . e . , they play a role in opening up closed chromatin . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"random",
"variables",
"covariance",
"regression",
"analysis",
"regulator",
"genes",
"mathematics",
"statistics",
"(mathematics)",
"transcription",
"factors",
"gene",
"types",
"epigenetics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"chromosome",
"biology",
"proteins",
"mathematical",
"and",
"statistical",
"techniques",
"gene",
"expression",
"histones",
"probability",
"theory",
"biochemistry",
"cell",
"biology",
"linear",
"regression",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"statistical",
"methods"
] | 2017 | Genome-Wide Association between Transcription Factor Expression and Chromatin Accessibility Reveals Regulators of Chromatin Accessibility |
Plant growth depends on stem cell niches in meristems . In the root apical meristem , the quiescent center ( QC ) cells form a niche together with the surrounding stem cells . Stem cells produce daughter cells that are displaced into a transit-amplifying ( TA ) domain of the root meristem . TA cells divide several times to provide cells for growth . SHORTROOT ( SHR ) and SCARECROW ( SCR ) are key regulators of the stem cell niche . Cytokinin controls TA cell activities in a dose-dependent manner . Although the regulatory programs in each compartment of the root meristem have been identified , it is still unclear how they coordinate one another . Here , we investigate how PHABULOSA ( PHB ) , under the posttranscriptional control of SHR and SCR , regulates TA cell activities . The root meristem and growth defects in shr or scr mutants were significantly recovered in the shr phb or scr phb double mutant , respectively . This rescue in root growth occurs in the absence of a QC . Conversely , when the modified PHB , which is highly resistant to microRNA , was expressed throughout the stele of the wild-type root meristem , root growth became very similar to that observed in the shr; however , the identity of the QC was unaffected . Interestingly , a moderate increase in PHB resulted in a root meristem phenotype similar to that observed following the application of high levels of cytokinin . Our protoplast assay and transgenic approach using ARR10 suggest that the depletion of TA cells by high PHB in the stele occurs via the repression of B-ARR activities . This regulatory mechanism seems to help to maintain the cytokinin homeostasis in the meristem . Taken together , our study suggests that PHB can dynamically regulate TA cell activities in a QC-independent manner , and that the SHR-PHB pathway enables a robust root growth system by coordinating the stem cell niche and TA domain .
Plants , unlike animals , grow continuously and dynamically adjust their architecture . Meristems at the apices of shoots and roots harbor stem cells and serve as the center of cell division and growth . In the root apical meristem ( RAM ) , the quiescent center ( QC ) maintains a stem cell population to form a stem cell niche [1] , [2] . There are two main pools of stem cells , the proximal and distal stem cells , which are named based on their position relative to the QC ( reviewed by [3] ) . Distal stem cells produce the root cap . Post-embryonic root growth in an apical direction , hereafter referred to as root growth , occurs via iterative divisions of the transit-amplifying cells ( TA cells ) , derived from the proximal stem cells , and their subsequent cell elongation . Two main regulatory programs are crucial for the root stem cell niche: one directed by PLETHORAs ( PLTs ) and the other by SHR and SCR . The PLTs play an essential role in establishing the QC and the RAM during embryogenesis [4] . SHR and SCR , the GRAS family transcription factors , act together to maintain the stem cell niche and root growth [5] , [6] , [7] . When SHR or SCR is knocked out , QC cells are not maintained and root growth terminates prematurely . The essential role of SHR and SCR in QC specification was further evident in a regeneration study of QC cells . In contrast to wild-type plants , QC cells failed to recover in shr and scr mutants after laser ablation [8] . The role of the QC within the stem cell niche has been extensively studied in Arabidopsis . Laser ablation of QC cells first indicated that the QC maintains the adjacent stem cell population [2] . In this study , the role of the QC in distal stem cell maintenance is evident; however , the extent to which the QC contributes to the maintenance of the proximal stem cells and the TA cell population is unclear [3] . In wox5 mutants , which affects the identity and morphology of the QC cells , the size of the TA cell population is unaltered and root growth is relatively normal even though the distal stem cells are not maintained [9] . Furthermore , root regeneration experiments have suggested that the QC is not required for the regeneration competency of the meristem cells [10] . In addition to the stem cell niche , cell division activities of the TA cell population significantly affect root growth . Cytokinin signaling plays a key role in this process . In the absence of cytokinin , TA cell proliferation is strongly inhibited . As a result , severe defects in root growth are observed in the triple mutant of ARR1 , 10 , and 12 , the type B Arabidopsis response regulator ( B-ARR ) genes , and the triple mutant of AHK2 , 3 and 4 , cytokinin receptor genes [11] , [12] . Under high cytokinin levels , the transition of TA cell status from dividing to differentiating is promoted , reducing the size of the meristem and slowing down root growth [13] . This process is mediated by ARR1 and 12 , which up-regulate the expression of SHY2 , an inhibitor of auxin signaling [14] . A recent study indicated that SCR in the QC regulates this process in a non-cell autonomous manner [15] . The HOMEODOMAIN-LEUCINE ZIPPER Class III ( HD-ZIP III ) genes are key regulators of shoot and root meristems ( reviewed by [16] , [17] ) . In Arabidopsis , the HD-ZIP III family consists of five members: PHABULOSA ( PHB ) , PHAVOLUTA ( PHV ) , CORONA ( CNA ) , REVOLUTA ( REV ) , and ARABIDOPSIS THALIANA HOMEOBOX8 ( ATHB8 ) [18] . These genes are post-transcriptionally regulated by microRNA ( miRNA ) 165/6 [19] , [20] . Studies have shown that during root meristem establishment , HD-ZIP III proteins are excluded from the basal embryo pole [21] , [22] . SERRATE ( SE ) , a zinc-finger ( ZnF ) domain-containing protein , is a component of dicing bodies that process pre-miRNAs together with DICER-LIKE1 ( DCL1 ) [23] . In se mutants , the production of mature miRNAs is reduced , thus PHB and PHV expand to the basal embryo pole , and formation of the root meristem fails [21] . SHR and SCR also suppress HD-ZIP III in the root by directly regulating the transcription of miR165A and 166B [24] . However , unlike SE , SHR and SCR mainly affect post-embryonic root development . In summary , significant progress has been made to identify key transcription factors and signaling molecules in the function and maintenance of the root stem cell niche and TA cells . However , it remains unclear how these factors and molecules are integrated to control root growth . Here , we show that PHB , the concentration of which is controlled by SHR , governs TA cell activities by repressing B-ARRs . Such a repressive effect of PHB on B-ARR activities is enhanced by high cytokinin . Therefore , restricting the PHB level via the SHR-microRNA pathway is a critical step for maintaining root growth .
Previously , we reported that SHR-dependent expression of miR165/6 in the ground tissue layer quantitatively and spatially restricts HD-ZIP III expression within the stele , which is necessary for normal xylem tissue patterning [24] . We also observed that the short root phenotype of shr mutants could be partially recovered when miR165/6 was miss-expressed in the ground tissue or stele , suggesting that ectopic HD-ZIP III expression reduced root growth in this mutant . However , at the time , it was not known which of the HD-ZIP III genes was responsible for root growth . We tested this by analyzing double mutants of shr and each of the five HD-ZIP III genes . In the double mutant of shr and phb-6 ( hereafter phb ) , a loss-of-function mutant allele of PHB with a Ds insertion , the roots were significantly longer than the shr mutant roots and more or less comparable to the wild type 7 days after germination ( DAG ) ( Fig . 1A; S1A Fig . ) . Under the growth conditions used in this experiment , phb single mutant roots were slightly shorter than the wild type ( S1A Fig . ) , which excluded the possibility that the rescue of root length in shr phb mutants is an additive effect . The negative influence of PHB on root length in the shr mutant was further supported in our suppressor screen of EMS-mutagenized shr plants . In this screen , we identified two new loss-of-function phb alleles , which are phb-15 and phb-16 ( S1A Fig . ) . Mutations in CORONA ( CNA ) or ATHB8 , two other HD-ZIP III genes , however , failed to rescue root length in the shr mutant , whereas PHAVOLUTA ( PHV ) and REVOLUTA ( REV ) elicited only a weak recovery ( Fig . 1B ) . Introducing a knockout mutant of PHV , the HD-ZIP III family member most closely related to PHB [19] , to the shr phb did not yield a further increase in shr phb root length ( Fig . 1B ) . To understand how root length is rescued in shr phb mutants , we analyzed root growth and meristem activities in comparison with the wild type and shr mutant ( Fig . 1C-1E ) . The shr mutant roots barely grew beyond 5 DAG , whereas shr phb roots displayed growth similar to , or faster than , wild-type roots until 7 DAG . After 7 DAG , however , the growth of these plants decelerated and had almost ceased by 15 DAG ( Fig . 1D ) . Consistent with the recovery of root growth , the meristem in the shr phb roots was significantly larger than the shr meristem ( S1B Fig . ) . A time-course analysis of root meristem [25] showed that both wild-type and shr phb roots displayed an increase in meristem size between 3 and 5 DAG , whereas shr roots showed a steady decrease ( Fig . 1C ) . At 5 DAG , the meristem of shr phb roots was larger than that of the wild type ( Student’s t-test; P < 0 . 001 , α = 0 . 05 ) . We also examined meristem cell division using a G2-M cell cycle marker , pCycB1 . 2:GUS [26] . GUS expression at 5 DAG indicated full recovery of cell division in the proximal meristem of the shr phb roots ( Fig . 1E ) . Interestingly , pCycB1 . 2:GUS expression was comparable in the wild-type and shr phb roots at 15 DAG , despite the deceleration of root growth in the shr phb plants ( S1C and 1D Fig . ) . These data suggest that the slowing of growth is not caused by a complete loss of cell cycle activity but rather it may be due to some other defects that are not directly related to meristematic function . For example , the shr phb mutant has defective development of the phloem sieve elements , which could affect the transport of nutrients required for root growth ( S2A Fig . ) . SCR regulates miR165/6 expression together with SHR [24] . Thus , we examined whether PHB also affects root growth activity controlled by SCR , and indeed , scr-4 phb roots displayed a significant recovery in the size of the proximal meristem ( S1B Fig . ) and root length compared with scr-4 roots ( Fig . 1F ) . Consistent with these findings , shr scr-4 phb triple mutant plants also recovered root length ( Fig . 1F ) . Finally , we investigated whether phb could suppress the short root phenotype of plt1 plt2 mutants . However , plt1-4 plt2-2 phb triple mutants did not exhibit any noticeable recovery in root length ( Fig . 1F ) . Taken together , these data suggest that PHB is a key downstream factor of SHR and SCR in the regulation of the meristem and root growth . The short root phenotype of the shr and scr mutants has been ascribed to the failure of stem cell maintenance in the absence of a QC . The lack of stem cell maintenance in shr mutants is supported by a steady decrease in root meristem size upon germination ( Fig . 1C ) . Because shr phb mutants displayed both an active increase in root meristem size and a rescue in root length , we examined whether the QC is rescued in the shr phb roots . To our surprise , none of the QC markers analyzed in actively growing shr phb roots indicated QC recovery ( Fig . 2; S2B Fig . ) . The expression of pWOX5:erGFP ( endoplasmic reticulum-localized GFP ) and pQC25:GUS , which is strong and specific to the QC cells in wild-type roots , was below the level of detection in the QC regions of both shr phb and shr roots ( Fig . 2A and 2C ) [9] , [27] . In the wild-type roots , expression of pPLT2:CFP , pPLT1:CFP , pSCR:erGFP , pAGL42:erGFP , and pQC6:erGFP [6] , [28] , [29] , [30] , are at their highest in the QC and are lower in the neighboring cell types . Analysis of these markers also indicated the lack of a QC in the shr phb roots ( Fig . 2B; S2B Fig . ) . We also analyzed pWOX5:erGFP in the wild-type , shr and shr phb embryos to determine whether the different WOX5 expression patterns in the shr and shr phb embryos could explain the recovery of root growth . Up to the mid-torpedo stage , we detected GFP expression in both wild-type and shr embryos ( S3A Fig . ) . Thereafter , contrary to that in the wild-type embryos , pWOX5:erGFP expression declined dramatically in the shr mutants and was no longer detectable ( S3B Fig . ) . The dynamics of pWOX5:erGFP expression in the shr phb mutant were indistinguishable from those in the shr mutant . Without the QC , the distal stem cells cannot be maintained . Lugol staining indicated that distal stem cells differentiate prematurely in shr phb roots , similar to that in shr roots visualized by the presence of starch granules that marks differentiated columella cells ( S3C Fig . ) . This is consistent with the lack of QC marker expression in the shr phb mutant , which supports the observation that the QC is absent in these plants . Recently , it was suggested that the QC promotes root growth by regulating the transition between cell proliferation and differentiation in a non-cell autonomous manner [15] . Therefore , we asked whether the presence of the QC in the shr phb mutant could fully restore root growth to wild-type levels . To investigate this , SHR fused to nuclear-localized GFP was expressed in shr phb roots under the WOX5 promoter ( pWOX5:SHR:nlsGFP; shr phb ) . Unlike the shr phb roots in which WOX5 expression diminished in the late stage of embryogenesis , three independent shr phb transgenic lines maintained SHR-nlsGFP expression in the QC region of post-embryonic roots , suggesting at least a partial recovery of QC driven by WOX5 ( Fig . 2D ) . Furthermore , the meristem of the shr phb roots expressing WOX5 appeared to be more organized than that of the shr phb plants , and the columella stem cells were restored as indicated by Lugol staining ( Fig . 2D ) . Expression of WOX5 in the QC position increased the growth potential of the shr phb roots; however , their lengths were significantly shorter than those of the wild-type roots ( Fig . 2E ) . Together these data suggest that proliferation of TA cells is controlled by the level of PHB , and that the PHB level plays a significant role in root growth , in combination with the QC maintenance and other unknown factors that affect root growth . SHR posttranscriptionally restricts HD-ZIP III mRNA to the center of the stele [24] . In the root meristem of shr mutants , PHB is observed throughout the stele ( S4B Fig . ) , which results in the short root phenotype that can be rescued by the misexpression of miR165/6 in the ground tissue . Hence , high PHB levels in the stele of the root meristem probably influence root growth . To confirm this , we increased the dosage of PHB in a specific manner throughout the stele in the wild-type root meristem . We modified PHB so that it was not efficiently targeted for degradation by miR165/6 , and then expressed it under the WOODEN LEG ( WOL ) promoter , which drives gene expression in stele cells in the root meristem [31] ( S5 Fig . ) . Two types of modified PHB were used: one with a single silent mutation that partially interferes with miR165/6 binding ( PHB-m ) [24] , and the other with four silent mutations that strongly block miR165/6 binding ( PHB-em ) ( S4A Fig . ) . Driving PHB-m under the PHB promoter results in the broadening of PHB domain throughout the stele as we previously reported ( S4B Fig . ) . Expression analysis using confocal microscopy showed that PHB-GFP is present at a significantly higher level in the stele cells of roots expressing pWOL:PHB-em:GFPNLS than in those expressing pWOL:PHB-m:GFPNLS ( Fig . 3B and 3C ) . Consistent with high PHB-GFP levels inhibiting normal root growth , roots expressing pWOL:PHB-em:GFPNLS were much shorter than those expressing pWOL:PHB-m:GFPNLS , and were either similar to or shorter than the shr mutant roots ( Fig . 3A ) . Accordingly , the meristem size of the pWOL:PHB-em:GFPNLS roots was also dramatically reduced ( S4C Fig . , upper panel ) . Furthermore , these root meristems had significantly fewer stele cells compared to wild-type root meristems ( S4C Fig . , lower panel ) . TA cell proliferation , as detected by pCycB1 . 2:GUS expression , decreased dramatically in these roots compared to the wild-type ones ( Fig . 3D ) . Collectively , these data suggest that high levels of PHB in the stele cells of the root meristem suppress TA cell proliferation and root growth , and therefore signaling from the stele is a rate-limiting factor in root growth . Next , we examined the QC status in the pWOL:PHB-em:GFPNLS lines that displayed root meristem and growth phenotypes similar to that of the shr mutant . The expression patterns of pQC25:GUS and pWOX5:YFP in pWOL:PHB-em:GFPNLS lines were similar to that in the wild type ( Fig . 3E and 3F ) ; their expression persisted even in older roots ( 15 DAG; S4D and S4E Fig . ) . Lugol staining of roots at 5 and 7 DAG also supported the presence of a QC ( Fig . 3G ) . Nevertheless , in the pWOL:PHB-em:GFPNLS roots , the cells in the QC position were often enlarged and showed aberrant cell divisions ( S4F Fig . ) . This phenotype was more frequent in older roots ( 7 DAG or older ) . Furthermore , in situ analysis indicated that there is a decline in WOX5 mRNA levels in these transgenic roots ( S4G Fig . ) . These data further confirm that PHB regulation of the TA cell proliferation and root growth is independent of the QC status . To determine how high levels of PHB in the stele suppress root growth in shr mutant plants , we generated stele-specific genome-wide gene expression data from wild-type , shr-2 , and shr-2 phb-6 roots . In this experiment , we used a protoplast/cell-sorting technique to isolate and collect the stele cells from roots expressing pWOL:erGFP in three genotypes , and mRNAs in the stele were profiled using the Arabidopsis Tiling 1 . 0R arrays from Affymetrix [32] ( S5 and S6A Fig . ; S1 Table ) . To identify genes or pathways that are regulated by PHB , we identified genes that are differentially expressed between the shr mutant and the wild type ( Corrected p-value , FDR <0 . 05; fold difference >2 ) . Under the given criteria , 1933 genes were revealed as differentially expressed between the two ( S2 Table ) . We then investigated the expression patterns of these genes in the shr , shr phb , and wild-type plants using principal component analysis ( PCA ) and QT clustering [33] ( S6B and S6C Fig . ; S2 Table ) . These analyses indicated that most changes in gene expression in the shr mutant ( 92% represented by 1781 genes in clusters 1–3 ) were due to high levels of PHB . In total , 1170 genes were activated ( cluster 1 ) and 611 genes were repressed ( clusters 2 and 3 ) by PHB in shr mutant plants ( S6C Fig . ) . Next , we analyzed the spatial expression of genes that are activated or repressed by high levels of PHB in the shr mutant using the expression data generated following micro-dissection of a root along the longitudinal axis and the cell type-specific root expression data ( Fig . 4A; S7 Fig . ) [24] , [29] , [34] , [35] , [36] . Hierarchical clustering of genes present in the root micro-dissection data indicates that ∼40% of the genes repressed by PHB are highly expressed in the TA domain , whereas those activated by PHB are enriched in the maturation zone . GO term enrichment analysis suggested that the pathways repressed by high levels of PHB are related to cell proliferation processes [37] ( S3 Table; Fig . 4B ) . This suggests that in the stele , PHB actively represses genes involved in TA cell activities . In contrast , pathways activated by PHB are biotic and abiotic stress responses , and those involved in hormone production and signaling . This includes genes in auxin , ABA , and cytokinin biosynthesis and response ( S3 Table; Fig . 4B ) . Cell-type specific expression of nearly 40% of genes repressed by PHB is enriched in the xylem precursor cells ( S7 Fig . ) [38] . This is consistent with the recovery of xylem patterning in shr phb mutant plants [24] . In contrast to the genes repressed by PHB , those activated by PHB showed enriched expression in multiple cell types . The overrepresentation of genes involved in multiple signaling processes might explain the lack of enriched patterns of gene activation in response to PHB in particular cell types . A balance between auxin and cytokinin signaling is important for root growth and meristem activity [39] . Our stele-specific mRNA analysis suggests that the SHR-PHB pathway might affect the balance of these two hormones . We further examined this by analyzing the expression pattern of pDR5:YFPvenus and pPIN1:PIN1:GFP , in shr , shr phb , and wild-type roots ( S8A–S8C Fig . ) . As reported recently [40] , pPIN1:PIN1:GFP expression declined in both 5- and 10-day-old shr roots . Its expression was moderately restored in the roots of shr phb plants , which is consistent with our data from stele profiling . A similar expression pattern was observed for pDR5:YFPvenus . These results suggest that high PHB levels in shr mutant plants reduce auxin signaling ( transport/biosynthesis ) in the root meristem , but its effect appears to be relatively minor . Previous studies have suggested the existence of a homeostatic feedback loop mechanism between auxin and cytokinin in the root , in which cytokinin positively regulates a subset of genes involved in auxin biosynthesis [41] , [42] . Our tissue-specific stele expression profiling revealed that high PHB levels in shr mutants up-regulate genes involved in auxin biosynthesis/metabolism . In addition , the level of PIN1 protein , which is degraded by high levels of cytokinin , was reduced in the shr mutant and was then moderately restored in the shr phb mutant ( S8B and S8C Fig . ) [43] . Moreover , cytokinin levels are increased in shr roots [44] . These results led us to examine whether high PHB in shr mutants affects the cytokinin status in the roots . ISOPENTENYL TRANSFERASE ( IPT ) 1 , 3 , 5 , and 7 promote cytokinin biosynthesis in Arabidopsis roots [45] . Our stele profiling data indicated that IPT5 and 7 are upregulated by PHB in the shr stele . Previous expression data in the shr and wild-type root tips revealed the up-regulation of IPT3 and 7 in shr mutants [35] . We corroborated these findings by quantifying the expression levels of IPT3 , 5 , and 7 in the wild-type , shr , and shr phb roots using qRT-PCR . Consistently , mRNA levels of IPT3 and 7 were found to increase in shr roots compared with wild-type roots , and their levels were partially restored in shr phb roots ( Fig . 5A ) . These results suggest that high levels of PHB enhance the expression of IPT3 and 7 . We further confirmed this by measuring the mRNA levels of IPT3 and 7 in the root tips of transgenic plants expressing pWOL:PHB-em:GFPNLS ( Fig . 5A ) . We then asked whether PHB regulation of IPT3 and 7 influences root meristem/growth activity by measuring the size of the root meristem and the length of the root in shr ipt3-2 ipt7-1 triple mutant plants . The shr ipt3-2 ipt7-1 plants showed a recovery in the root length and meristem size compared to shr plants , but to a much lesser degree than the shr phb roots ( Fig . 5B and 5C; P < 0 . 001 , α = 0 . 05 ) . Under our growth conditions , the ipt3-2 ipt7-1 double mutant roots were slightly shorter than those of wild-type plants . Collectively , these data indicated that PHB influences root meristem and growth activity in the shr plants , at least partly via the regulation of cytokinin biosynthesis . To determine the significance of cytokinin biosynthesis regulation by PHB in the root meristem , we measured cytokinin levels in wild-type , shr , and shr phb roots . Consistent with the previous study , an increase in cytokinin level was observed in the shr roots ( Fig . 6A; S9A Fig . ) . Various forms of trans-zeatin increased by approximately four-fold in shr plants in comparison to the wild type . However , we failed to detect a decline in global cytokinin levels in shr phb roots , as compared with those in shr roots . This suggests that there is another layer of regulation mediated by PHB in the cytokinin pathway , which is probably involved in restoring the shr phb root growth . In our stele profiling data , the expression of multiple A-ARRs , such as ARR5 , 6 , 7 , 8 , 9 , and 15 , was slightly induced in shr mutants but was further upregulated in shr phb mutants ( S9B Fig . ) . We validated this finding by qRT-PCR analysis ( S9C Fig . ) . Considering that global cytokinin levels were comparable in both shr and shr phb roots , this result implied that high PHB levels in the shr mutant might be actively suppressing the expression of A-ARRs . All A-ARR genes have cis-regulatory elements that are directly bound by B-ARRs [46] . pTCS:erGFP is a synthetic cytokinin reporter construct with B-ARR binding elements , and its expression therefore reflects the transcriptional activity of B-ARRs . When the in vivo cytokinin level increases in the wild-type root meristem , more B-ARRs are activated and thereby enhance pTCS:erGFP expression [47] . Thus , we investigated the involvement of PHB in B-ARR function by analyzing the expression of pTCS:erGFP in the wild-type , shr , and shr phb roots [47] . Despite a high cytokinin level , shr roots displayed a dramatic reduction in pTCS:erGFP expression ( Fig . 6B ) . In contrast , TCS expression was restored in shr phb mutants to levels higher than those in wild-type plants . Thus , high levels of PHB appear to interfere with B-ARRs that induce TCS transcription in the root stele . Consistent with this theory , TCS expression was found to be absent or very weak in the root stele of phb-7d mutants carrying a semi-dominant point mutation in the microRNA target site of PHB [24] ( S9E Fig . ) . A high level of PHB could interfere with B-ARR activity directly or indirectly . To determine how PHB affects B-ARR activity , we developed a transgenic line expressing PHB-em fused to the glucocorticoid receptor ( GR ) under the control of the PHB promoter , and crossed a line expressing this construct with pTCS:erGFP ( S10A and S10B Fig . ) . The resulting F1 seedlings were treated with dexamethasone for 3 hours and pTCS:erGFP expression in roots was analyzed in comparison with that of the untreated control plants . GFP expression in roots harboring both pTCS:erGFP and pPHB:PHB-em:GR was much weaker than that of roots carrying pTCS:erGFP alone , suggesting that the PHB-em:GR is leaky . Nevertheless , with dexamethasone treatment , we found a significant reduction in GFP expression in roots with both pTCS:erGFP and pPHB:PHB-em:GR , not in roots with pTCS:erGFP alone . Since the response of pTCS:erGFP occurs only after 3 hours , we investigated the possibility that PHB expressed at high levels can interact with the promoters of A-ARR genes . ARR7 , an A-ARR that is highly responsive to PHB in the root meristem , was selected and used to perform a chromatin immunoprecipitation ( ChIP ) assay ( Fig . 6C; S9B and S9C Fig . ) . The ChIP assay , combined with quantitative real-time PCR , indicated enrichment of PHB binding to the ARR7 promoter region where B-ARR binding elements are present . We then tested whether an enhanced ARR7 expression contributes to root length recovery in shr phb mutants by analyzing root length in shr phb arr7 triple mutant plants ( S9D Fig . ) . The shr phb arr7 roots showed a partial reduction in root length compared with that of shr phb , suggesting that A-ARRs may influence the restoration of root growth in shr phb mutant plants ( Student’s t-test; P < 0 . 001 , α = 0 . 05 ) . The suppression of negative regulators of cytokinin signaling was recently shown to enhance cytokinin sensitivity in Arabidopsis roots [48] . Consistent with this , our expression analysis of two cytokinin-sensitive markers , AHP6 [49] and S32 [30] , [44] , indicated the enhanced cytokinin sensitivity in the shr . The expression of pAHP6:erGFP , which is inhibited by high levels of cytokinin , was absent in shr mutants and was partially restored in shr phb roots ( Fig . 6D ) . The expression of pS32:erGFP , which is induced by high levels of cytokinin , increased and expanded in the shr roots but was consistent with the wild-type pattern in shr phb roots ( S9F Fig . ) . Our stele-specific microarray data also supported the dynamics of these markers . Thus , it appears that the increased cytokinin sensitivity in shr mutants is due to the suppression of A-ARR expression by PHB . Our data so far indicate that high levels of PHB inhibit B-ARR activity . Since ChIP data suggest that PHB acts on B-ARR binding targets , we hypothesized that PHB might block B-ARRs in a quantifiable manner . To test this hypothesis , we employed a protoplast transient assay system ( Fig . 7A; S11A Fig . ) . We measured luciferase activity driven by the TCS promoter ( pTCS:LUC ) in protoplasts transfected with 1 ) ARR10 alone , 2 ) PHB-em alone , 3 ) ARR10 and PHB-em , 4 ) PHB alone , and 5 ) ARR10 and PHB which were all driven by the 35S promoter . In two independent experiments , PHB mRNA levels expressed from p35S:PHB-em were found higher than PHB from p35S:PHB ( Fig . 7B; S11B Fig . ) . In the absence of exogenous cytokinin , we did not detect a dramatic difference in luciferase activity between the protoplasts transfected with different constructs . However , in the presence of 6-benzylaminopurine ( BAP ) , PHB protein expressed by p35S:PHB-em strongly suppressed ARR10 activity . This repressive effect was also observed in protoplasts transfected with PHB-em alone . However , the effect of PHB expressed by p35S:PHB on ARR10 activities was inconsistent . This seems to be due to the influence of miRNA 165/6 that targets PHB mRNA but not PHB-em ( Fig . 7B; S11B Fig . ) . Collectively , these data confirm that high levels of PHB repress B-ARR activities , more effectively under high cytokinin . Our protoplast assay further indicated that the root growth defect in shr-2 , phb-7d and pWOL:PHB-em:GFPNLS plants might be due to depleted , as opposed to enhanced , cytokinin signaling . If this mechanism were important in regulating root growth , supplying more B-ARR to the shr or phb-7d mutant would overcome the inhibitory effect of PHB on root growth . To test this hypothesis , we introduced pWOL:ARR10 into the phb-7d mutant . Analysis of four T2 plants showed a noticeable restoration in root growth , supporting our hypothesis ( Fig . 7C ) .
Controlling the spatial domains/levels of HD-ZIP III genes is critical for multiple developmental processes: ad/abaxial organization of lateral organs [50] , specification of apical/basal axis in embryos [22] , and vascular patterning [24] , [51] . In this study , we demonstrate that this regulation is also essential for proper organization of the root meristem and root growth . Among the five HD-ZIP III genes , PHB seems to be the major player , because deletion of the other HD-ZIP III genes did not restore root growth in the same way as phb did in the shr mutant . Consistent with the results of a previous study showing that SCR acts in conjunction with SHR to posttranscriptionally suppress PHB and other HD-ZIP IIIs , the scr phb mutant is also able to restore root length . Despite recovery of root growth , we did not detect any visible sign ( based on cell morphology and QC marker expression ) of an accompanying recovery of QC cells in the shr phb mutant . TA cell proliferation becomes very active in shr phb , as indicated by the expression of pCycB1 . 2:GUS , and the size of the meristem over time , displayed very similar dynamics between wild-type and shr phb roots . These findings indicate that the proliferation of TA cells is prolonged in the absence of QC cells in shr phb mutants . Conversely , pWOL:PHB-em:GFPNLS plants expressing a high level of PHB in the root stele cannot maintain TA cell proliferation and root growth while sustaining a QC . Therefore , PHB in the stele cells regulates TA cell activities required for root growth , independently of the QC in a non-cell autonomous manner . Interestingly , once the TA cells stop functioning in the pWOL:PHB-em:GFPNLS roots , the QC starts to lose its identity as a secondary effect . In animal systems , the maintenance of stem cell niches requires a communication between stem cells and their differentiated progenies , or TA cells [52] , [53] , [54] . When TA cells in the intestine are damaged , mitotically inactive cells in the stem cell niche become active and show stem cell characteristics [55] , [56] . These features suggest that the mechanisms of stem cell niche functioning in animals and plants might be similar , even though the components involved are distinct . Root growth is a dynamic process . Growth accelerates to a certain time point and then decelerates to terminate ( reviewed by [57] ) . During this process , the meristem structure also changes . Toward the end of the root growth period , the QC degenerates , which is defined by its cell division . This results in the transition of the closed meristem to an open meristem in the Arabidopsis root [58] , [59] . Therefore , the presence of a QC appears to be important for maintaining root growth over time . In this context , the RAM might operate in a similar way to the shoot apical meristem , where the organizing center is important for prolonging the meristem activity [60] . shr phb roots have open meristems and grow in a determinate manner . This determinate root growth does not appear to be caused by the exhaustion of TA cells because the root meristem size of shr phb plants does not change from that of the wild type until 15 DAG . Therefore , the early termination of root growth in shr phb mutants might be due to the lack of QC cells . Consistent with this theory , when the QC is recovered in shr phb roots by expressing pWOX5::SHR:nlsGFP , the growth becomes more active and less determinate than the shr phb . However , the overall root growth in the presence of QC in these plants remained below the wild-type level . Together , these data suggest that there is another aspect of root growth regulation , or that the QC is not fully restored . It has been reported that root growth in shr mutants can be partially rescued via the expression of cell-autonomous SHR expression in the stele [24] , [61] . Our further investigation suggests that this restores the defect in phloem development and accompanies a partial recovery in root growth ( manuscript in preparation ) . Cytokinin affects root meristem and growth in a concentration-dependent manner [62] . Our study suggests that PHB regulates the activity of the root meristem and growth by two mechanisms: one involving cytokinin synthesis and the other involving changes in B-ARR activities . Our gene expression analysis indicated an up-regulation of IPT3 and IPT7 expression in accordance with elevated levels of PHB , and vice versa . Direct regulation of IPT7 expression by PHB was recently demonstrated [63] . However , activation of this pathway is not sufficient to produce a change in global cytokinin levels in the root , as we could not detect a change in cytokinin levels in the shr phb roots compared to that in shr plants . Nevertheless , when we prevented the up-regulation of IPT3 and IPT7 by PHB in shr mutants by generating shr ipt3 ipt7 triple mutants , root length and meristem size were partially restored . This is consistent with the results from our protoplast assay , which showed that the alteration of B-ARR activity by PHB is enhanced under high cytokinin levels . Parallel to local cytokinin production , PHB regulates cytokinin signaling , and this process appears to be the main factor affecting root meristem and growth activity in the shr mutant . Despite the fact that cytokinin levels were similarly high in the shr and shr phb mutants , the expression of cytokinin-responsive markers was not consistent in shr and shr phb roots . Expression patterns of pAHP6:GFP and pS32:GFP indicated a higher cytokinin status in the shr root stele , while the shr phb mutants were similar to wild-type plants . On the other hand , pTCS:GFP , which reflects the transcriptional activities of B-ARRs is strongly suppressed in the shr and phb-7d mutants , but is recovered in the shr phb plants , demonstrating the effect of elevated PHB on cytokinin signaling pathway . Furthermore , consistent with the pTCS:GFP results , our stele-specific gene expression data obtained from the shr , shr phb , and wild-type roots indicated that a high level of PHB in the shr stele actively suppresses the expression of at least 50% of A-ARRs in the genome . On the basis of these data , together with the ChIP-PCR result suggesting that PHB can bind to the ARR7 promoter , we propose that increased PHB expression suppresses the cytokinin-mediated transcriptional activation by B-ARRs . This hindrance in B-ARR activity on one hand seems to increase cytokinin sensitivity in the shr roots , but on the other hand it affects TA cell proliferation leading to root growth arrest in a manner comparable to a cytokinin depleted mutant such as the arr1 10 12 triple mutant . Data from shr phb arr7 and pWOL:ARR10 provide further support for the role of PHB in B-ARR activities . The behavior of PHB in the shr mutant was different to that reported previously for PHB in the cytokinin-signaling pathway . Our analysis of PHB in the protoplast assay suggests that PHB can suppress B-ARR activities . High cytokinin levels somehow augment the action of PHB on B-ARRs . The effect of cytokinin in this context might be related to changes in the phosphorylation status of B-ARRs , which somehow enhances repressive activities of PHB via physical interaction , competition for the same promoter element , or alteration of protein stability . On the basis of the results presented in this study , we propose a model explaining how the stem cell niche and TA cell activities are coordinated via the SHR , SCR , PHB , and cytokinin pathways ( Fig . 8 ) . In this model , SHR and SCR regulate root meristem and growth via two pathways . One pathway , which maintains the QC , is required for stem cell maintenance and the subsequent extension of root growth , and the other pathway , mediated by posttranscriptional suppression of PHB , is critical for TA cell proliferation . In the latter , PHB regulates the root meristem and growth activities by modulating local cytokinin production and B-ARR activities in a context dependent manner . Such a dynamic regulatory system might enable the root meristem to respond appropriately to the ever-changing growth environment . Future investigations into the detailed mechanisms underlying how PHB alters B-ARRs , could help to determine how these pathways orchestrate root meristem activity and growth .
The Arabidopsis thaliana mutant/marker lines used are listed in S4 Table . Ecotype Columbia , or lines introgressed into Columbia , were used in this study unless stated otherwise . Seeds were surface-sterilized and germinated on MS agar plates supplemented with 1% sucrose at 22–23°C under a photoperiod of 16-h light/8-h dark . The various mutant combinations used were generated via genetic crossing , and in each case , the genotypes were verified by polymerase chain reaction ( PCR ) . The oligonucleotide sequences used are provided in S5 Table . For the suppressor screen , shr homozygous seeds were treated with EMS as previously described [64] . The phb-15 and phb-16 alleles were identified in reciprocal crosses with shr-2 phb-6 mutants and subsequently confirmed by PCR sequencing . Both the phb-15 and 16 alleles harbor a stop codon ( C>>T ) at amino acid positions 149 ( exon 3 ) and 384 ( exon 8 ) , respectively . Primary roots of 5 to 21-day-old vertically grown seedlings were used for various analyses . Root lengths ( from the base of the hypocotyl to the tip of the primary root ) were measured using Image J software ( www . rsbweb . nih . gov ) . The results presented here are the average data of 15–50 seedlings . GUS staining , Lugol staining , stereo ( Leica DM5500B ) and confocal microscopy ( Leica TCS SP5 Laser Scanning Confocal Microscope ) were conducted , with excitation ( Ex ) and emission ( Em ) wavelengths ( Ex/Em ) as follows: 405/461–504 nm for CFP , 488 nm/505–530 nm for GFP , 514 nm/525–555 nm for YFP , and 488–514 nm/613–679 nm for propidium iodide . Observation of callose on sieve plates in the phloem was performed as previously described [65] . In addition , qRT-PCR ( using 2-mm-long segments of root tips , ABI 7900HT , Applied Biosystems ) , embryo analysis and whole-mount RNA in situ hybridization were performed as previously described [24] , [66] , [67] , [68] , [69] , [70] , . ChIP experiments ( three biological replicates per experiment ) were performed as previously described [7] . We used 5-day-old roots from seedlings expressing pWOL:PHB-em:GFPNLS . The oligonucleotide sequences used are listed in S5 Table . Wild-type , shr and shr phb plants harboring pWOL:erGFP were germinated and grown for 6 days on solid MS medium under the same conditions as described above . Apical half portions of roots were cut and harvested . Protoplasts were prepared from root tissues as previously described [32] , and pWOL:erGFP expressing protoplasts were collected with an Aria high-speed flow cell sorter ( BD Biosciences ) and were subsequently used for microarray analysis . Total RNA was isolated from collected protoplasts using an RNeasy Plant Mini Kit ( Qiagen ) . RNA integrity was determined using a bioanalyzer ( Agilent BioAnalyzer 2100 ) . Poly-adenylated RNA was isolated from total RNA using a RiboMinus Plant Kit for RNA-Seq ( Invitrogen ) . Probe preparation was conducted according to the manufacturer’s instructions ( GeneChip Whole Transcript Double-Stranded Target Assay Manual from Affymetrix Inc . ) , and then biotinylated double-stranded DNA probes were hybridized to the Arabidopsis Tiling 1 . 0R arrays ( Affymetrix Inc . ) at Cornell Life Sciences Core Laboratories Center . Two ( shr and shr phb mutants ) to three ( wild-type plants ) biological replicate datasets with high correlation coefficients were generated for analysis from each genetic line . Microarray data were normalized using the RMA algorithm [71] in BIOCONDUCTOR , with Tiling 1 . 0R array CDF that contains gene-specific single-copy exonic probe sets [72] . Genes that were differentially expressed between the wild-type plant and the shr mutant were detected using the LIMMA package [73] . Principal component analysis ( PCA ) and clustering of differentially expressed genes were performed and visualized using MultiExperimental Viewer [74] . PHB-em , generated via site-directed mutagenesis following the manufacturer’s protocol ( QuikChange II Site-Directed Mutagenesis Kit , Stratagene ) , was cloned into pDONR221 ( Invitrogen ) . The 4 . 5-kb upstream sequence of WOX5 was PCR amplified from Arabidopsis genomic DNA and cloned into pDONRP4_P1R . pWOX5:SHR:nlsGFP was generated by fusing SHR cDNA to nlsGFP and the WOX5 promoter . Transgenes were placed into dpGreen vectors using the Invitrogen MultiSite Gateway system and were mobilized into plants using the floral dip method , as previously described [75] , [30] . For protoplast assays , we used a pTCS:fLUC ( Firefly Luciferase ) construct as a reporter system and a pUBQ10:rLUC ( Renilla Luciferase ) construct as an internal transfection control . For the effector construct , a region containing the coding sequences of target genes ( ARR10 , PHB , PHB-em ) was inserted downstream of the 35S promoter using a modified pHBT-sGFP ( S65T ) -NOS vector . As empty vector , pSPYCE , a binary vector with cauliflower mosaic virus 35S promoter and the C-terminal ( 156–239 aa ) domain of YFP , was used . Cytokinin quantification was performed as previously described [76] , with some modifications . Briefly , ∼140 mg of the fresh half-bottom part of roots was collected from 6-day-old seedlings for each experimental replicate . Root tissues that were frozen in liquid nitrogen were processed in a vibration mill with tungsten carbide beads and 1-mL Bieleski buffer [77] . Deuterium-labeled cytokinin internal standards ( Olchemim Ltd . , Czech Republic ) were added at the extraction stage , at 3 pmol per sample , to evaluate the recovery during purification and validate the quantification . The crude extract was then centrifuged ( 15 000 × g at 4°C ) , and the pellets were re-extracted in the same way . The combined supernatants were purified using a 1-g SCX column , evaporated to water phase in a vacuum followed by immunoaffinity chromatography using an immobilized wide-range anti-cytokinin monoclonal antibody as previously described [78] . The fraction containing cytokinin 9-glucosides , bases , and ribosides was evaporated to dryness , re-dissolved in the mobile phase and analyzed by UPLC-MS/MS [79] . The data were analyzed using Masslynx 4 . 1 software ( Waters , Milford , MA , USA ) and quantified by the standard isotope-dilution method . Arabidopsis mesophyll protoplasts were isolated and transfected as previously described [80] , and DNA was prepared using a large-scale preparation of high-quality plasmid DNA [81] . Protoplasts ( 6 × 104 cells ) were transfected with 40 μg of plasmid DNA with different combinations of reporter ( pTCS:fLUC ) , effector ( p35S:ARR10 , p35S:PHB , p35S:PHBem ) , and internal control ( pUBQ:rLUC ) plasmids ( S6 Table ) . Protoplasts were incubated in WI [0 . 5 M mannitol , 4 mM MES ( pH 5 . 7 ) , 20 mM KCl] at room temperature for 1 h in the light . After 1-h incubation , protoplasts were treated with 100 nM of 6-benzylaminopurine ( BAP ) and incubated for 5 h under the same conditions . Luciferase reporter activity was determined by the Dual Luciferase Assay System ( Promega ) and detected by Microlumat 96V Luminometer ( Berthold , Bad Wildbad , Germany ) . To analyze PHB expression , quantitative RT-PCR analyses were carried out using total RNAs extracted from protoplasts ( 3 × 104 cells ) , which were collected from protoplast transient assay . Total RNA extraction was performed with PicoPure RNA Isolation Kit ( Arcturus ) . 20μL of reverse transcript reaction was performed for the first cDNA strand synthesis using 500ng of total RNAs and Superscript III reverse transcriptase ( Invitrogen ) . For quantitative PCR reactions , a master mix was prepared using an iQ SYBR Green supermix ( Bio-Rad ) and PCR condition was programmed according to the manufacturer’s instructions for CFX96 Real-Time PCR machine ( Bio-Rad ) . Three technical replicate reactions were performed . Primer information for RT-PCR is available in S5 Table . Glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) was used as internal control gene . Microarray data have been submitted to the GEO database ( accession number GSE33015 ) . Sequence data from this article can be found in the Arabidopsis Genome Initiative or GenBank/EMBL databases under the following accession numbers: SHR ( At4G37650 ) , PHB ( At2G34710 ) , PHV ( At1G30490 ) , CNA ( At1G52150 ) , ATHB8 ( At4G32880 ) , REV ( At5G60690 ) , SCR ( At3G54220 ) , PLT1 ( At3G20840 ) , PLT2 ( At1G51190 ) , S32 ( At2g18380 ) , AGL42 ( At5g62165 ) , IPT3 ( At3G63110 ) , IPT7 ( At3G23630 ) , ARR7 ( At1G19050 ) , ARR10 ( At4G31920 ) and GAPDH ( At3g26650 ) . | Plant roots are programmed to grow continuously into the soil , searching for nutrients and water . The iterative process of cell division , elongation , and differentiation contributes to root growth . The quiescent center ( QC ) is known to maintain the root meristem , and thus ensure root growth . In this study , we report a novel aspect of root growth regulation controlled independently of the QC by PHABULOSA ( PHB ) . In shr mutant plants , PHB , which in the meristem is actively restricted to the central region of the stele by SHORTROOT ( SHR ) via miR165/6 , suppresses root meristem activity leading to root growth arrest . A high concentration of PHB in the stele does this by modulating B-ARR activity through a QC-independent pathway . Accordingly , we observed a significant recovery of root meristem activity and growth in the shr phb double mutant , while the QC remained absent . However , the presence of QC may be required to sustain continuous root growth . On the basis of our results , we propose that SHR maintains root growth via two separate pathways: by modulating PHB levels in the root stele , and by maintaining the QC identity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | PHABULOSA Controls the Quiescent Center-Independent Root Meristem Activities in Arabidopsis thaliana |
B-cells , in addition to antibody secretion , have emerged increasingly as effector and immunoregulatory cells in several chronic inflammatory diseases . Although Erythema Nodosum Leprosum ( ENL ) is an inflammatory complication of leprosy , the role of B- cell subsets has never been studied in this patient group . Therefore , it would be interesting to examine the contribution of B-cells in the pathogenesis of ENL . A case-control study design was used to recruit 30 untreated patients with ENL and 30 non-reactional lepromatous leprosy ( LL ) patient controls at ALERT Hospital , Ethiopia . Peripheral blood samples were obtained before , during and after treatment from each patient . Peripheral blood mononuclear cells ( PBMCs ) were isolated and used for immunophenotyping of B- cell subsets by flow cytometry . The kinetics of B-cells in patients with ENL before , during and after Prednisolone treatment of ENL was compared with LL patient controls as well as within ENL group . Total B-cells , mature B-cells and resting memory B-cells were not significantly different between patients with ENL reactions and LL controls before treatment . Interestingly , while the percentage of naive B-cells was significantly lower in untreated ENL patients than in LL patient controls , the percentage of activated memory B-cells was significantly higher in these untreated ENL patients than in LL controls . On the other hand , the percentage of tissue-like memory B-cells was considerably low in untreated ENL patients compared to LL controls . It appears that the lower frequency of tissue-like memory B-cells in untreated ENL could promote the B-cell/T-cell interaction in these patients through downregulation of inhibitory molecules unlike in LL patients . Conversely , the increased production of activated memory B-cells in ENL patients could imply the scale up of immune activation through antigen presentation to T-cells . However , the generation and differential function of these memory B-cells need further investigation . The finding of increased percentage of activated memory B-cells in untreated patients with ENL reactions suggests the association of these cells with the ENL pathology . The mechanism by which inflammatory reactions like ENL affecting these memory cells and contributing to the disease pathology is an interesting area to be explored for and could lead to the development of novel and highly efficacious drug for ENL treatment .
B-cells enable the antigen-specific humoral immunity by forming highly specific antibodies during primary immune response . B-cells within the lymphoid tissue of the body such as bone marrow , spleen and lymph nodes , are stimulated by antigenic substances to proliferate and transform into plasma cells and the plasma cells in turn produce immunoglobulins which bind to cognate antigen [1] . Although B-cells are traditionally known as precursors for antibody-secreting plasma cells , they may also act as antigen-presenting cells ( APC ) and play an important role in the initiation and regulation of T and B cell responses [1 , 2] . However , B-cells may also involve in disease pathology especially in autoimmune disorders . The pathogenic roles of B-cells in autoimmune diseases occur through several mechanistic pathways that include autoantibodies , immune-complexes , dendritic and T-cell activation , cytokine synthesis , chemokine-mediated functions , and ectopic neolymphogenesis [2] . Memory B-cells are B-cell sub-types that are formed within the germinal centres following primary infection and are important in generating an accelerated and more robust antibody-mediated immune response in the case of re-infection also known as a secondary immune response . Recent advances in tracking antigen-experienced memory B-cells have shown the existence of different classes of memory B-cells that have considerable functional differences . Currently there are three types of memory B-cells: resting , activated and tissue like memory B-cells , [3] . Activated memory B-cells have been shown to function as effective antigen presenting cells ( APCs ) to naive T-cells [4] . Tissue-like memory B-cells ( TLM ) expressed patterns of homing and inhibitory receptors similar to those described for antigen-specific T-cell exhaustion . Tissue like memory B-cells proliferate poorly in response to B-cell stimuli , which is consistent with high-level expression of multiple inhibitory receptors . Higher percentage of TLM has been reported in immunosuppressive diseases such as HIV [5 , 6] . Leprosy is a spectrum disease with the polar tuberculiod ( TT ) and lepromatous ( LL ) forms and the three borderlines forms including borderline tuberculoid ( BT ) , mid borderline ( BB ) and borderline lepromatous ( BL ) [7] . TT characterized by strong cell-mediated immune response which restricts the spread of M . leprae while the LL forms are characterized by lack of cell mediated immune response which allows the growth and spread of M . leprae in these patients [8] . Studies have shown that circulating high levels of antibodies to M . leprae specific antigens in LL patients although these antibodies are unable to control the growth and the spread of M . leprae [9] . The study of the humoral immunity in leprosy has largely been restricted to antibodies . Patients towards lepromatous leprosy ( LL ) pole of the spectrum have higher antibody concentration as compared with the tuberculoid ( TT ) pole . Elevation of the polyclonal isotypes of these classes of antibody types with the highest concentration has reported in patients with LL forms compared to the other clinical types of the spectrum [10–13] . However , the role of these antibodies in the pathogenesis of leprosy is poorly understood . Although the in-situ presences of plasma cells and B-cells have been reported in leprosy , the role of these cells in the pathology of leprosy lesions is unclear . Both Plasma cells and B-cells have been detected in tuberculoid and lepromatous leprosy lesions [14] . It was speculated that these lesional B-cells could influence T-cell responses and /or play a role in maintaining the inflammatory reaction in leprosy partly through the local secretion of antibodies . However , data supporting such hypothesis are lacking . It is generally thought that antibodies against M . leprae components do not play a significant role in protection against leprosy . However , antibodies may play a role in the uptake of M . leprae by mononuclear phagocytes and hence the pathogenesis of the diseases [15] . There are two types of leprosy reaction , type one and erythema nodosum leprosum ( ENL ) reactions . ENL is an immune-mediated inflammatory complication affecting about 50% of patients with lepromatous leprosy ( LL ) and 10% of borderline lepromatous ( BL ) patients [16–18] . ENL can occur before , during or after successful completion of multi-drug therapy ( MDT ) . The onset of ENL is acute , but it may pass into a chronic phase and can be recurrent [19] . B -cells are the least studied immune cells in leprosy in general and leprosy reactions in particular . An increased percentage and absolute count of B-cells in the sera from patients with ENL has been reported [20] , but normal numbers of circulating B-cells have also been reported [21] . A study looking at T-cell phenotypes in ENL lesions showed a normal proportion of B-cells in these patients [22] . In a prospective cohort study of 13 untreated patients with acute ENL reaction , polyclonal IgG1 antibody synthesis was elevated compared to patients with stable lepromatous leprosy and decreased after the disease had subsided . However , the concentration of polyclonal IgG2 had revealed the reverse trend: decreased before treatment and increased after treatment [23] . These authors also investigated the frequency of antibody secreting B-cells in the blood compartment of these patients with the Enzyme-Linked ImmunoSpot ( ELISPOT ) and found that the decrease in M . leprae specific IgG1 antibody was not related to the down-regulation of B-cell responses . In addition to antibody secretion , B-cells have emerged increasingly as both effector and immunoregulatory cells in several chronic inflammatory diseases [24] . The role of B-cells in the pathogenesis of autoimmune disorders such as rheumatoid arthritis ( RA ) and systemic lupus erythematosus ( SLE ) is now being re-examined [1] . It would therefore be interesting to examine the contribution of B-cells in the pathogenesis of ENL .
Informed written consent for blood and skin biopsies were obtained from patients following approval of the study by the Institutional Ethical Committee of London School of Hygiene and Tropical Medicine , UK , ( #6391 ) , AHRI/ALERT Ethics Review Committee , Ethiopia ( P032/12 ) and the National Research Ethics Review Committee , Ethiopia ( #310/450/06 ) . Under 18 years old patients were excluded from the study . Vulnerable and minor groups were also excluded from the study . All patient data analyzed and reported anonymously . A case-control study with follow-up after the initiation of prednisolone treatment was used to recruit 30 untreated patients with ENL reactions and 30 non-reactional LL patient controls between December 2014 and January 2016 at ALERT Hospital , Ethiopia . All patients recruited into this study were attending the ALERT Hospital , Addis Ababa , Ethiopia . The patients were classified clinically and histologically on the leprosy spectrum based on the Ridley-Jopling ( RG ) classification schemes [7] . ENL was clinically diagnosed when a patient with BL or LL leprosy had painful crops of tender cutaneous erythematous skin lesions [17] . New ENL was defined as the occurrence of ENL for the first time in a patient with LL or BL . Lepromatous leprosy was clinically diagnosed when a patient had widely disseminated nodular lesions with ill-defined borders and BI above 2 [19] . Patients with ENL were treated according to the World Health Organization ( WHO ) treatment guideline with steroids that initially consisted of 40mg oral prednisolone daily and the dose was tapered by 5mg every fortnight for 24 weeks . All patients were received WHO-recommended leprosy multidrug treatment ( MDT ) . Twenty micro-litter of venous blood was collected into sterile BD heparinised vacutainer tubes ( BD , Franklin , Lakes , NJ , USA ) before treatment , during treatment on week 12 and after treatment on week 24 from each patient and used for PBMC isolation . PBMCs were separated by density gradient centrifugation at 800g for 25 min on Ficoll-Hypaque ( Histopaque , Sigma Aldrich , UK ) as described earlier [25] . Cells were washed three times in sterile 1x phosphate buffered saline ( PBS , Sigma Aldrich , UK ) and re-suspended with 1mL of Roswell Park Memorial Institute ( RPMI medium 1640 ( 1x ) + GlutaMAX+ Pen-Strip GBICO , Life technologies , UK ) . Cell viability was determined by 0 . 4% sterile Trypan Blue solution ( Sigma Aldrich , UK ) ranged from 94–98% . PBMC freezing was performed using a cold freshly prepared freezing medium composed of 20% Foetal Bovine Serum ( FBS , heat inactivated , endotoxin tested ≤5 EU/ml , GIBCO Life technologies , UK ) , 20% dimethyl sulphoxide ( DMSO ) in RPMI medium 1640 ( 1x ) . Cells were kept at -80°C for 2–3 days and transferred to liquid nitrogen until use . Cell thawing was done as described [26] . The procedure is briefly described as: cells were incubated in a water bath ( 37°C ) for 30 to 40 seconds until thawed half way and re-suspended in 10% FBS in RPMI medium 1640 ( 1x ) ( 37°C ) containing 1/10 , 000 benzonase until completely thawed , washed 2 times ( 5 minutes each ) and counted . The percentage viability obtained was above 90% . Cells were harvested , transferred to round bottomed FACS tubes ( Falcon , BD , UK ) and washed twice at 400g for 5 minutes at room temperature . The cells were resuspended in 50μl of PBS and incubated in 1ml of 10% human AB serum ( Sigma Aldrich , UK ) for 10 minutes in the dark at room temperature to block nonspecific Fc-mediated interactions , and centrifuged at 400g for 5 minutes . After resuspending cells in 50μL PBS buffer , Life/dead staining was performed at a concentration of 1μl /1mL live/dead stain ( V500 Aqua , Invitrogen , Life technologies , UK ) for 15 minutes at 4°C in the dark . Cells were washed once and stained for surface markers directed against CD10 ( FITC ) , CD19 ( PerCp-Cy5 . 5 ) , CD27 ( V500 ) , CD21 ( V450 ) and Isotype control ( IgG1 ) ( all from BD Biosciences , UK ) , Live/dead ( eFluoro 780 , Invitrogen , Life technologies , UK ) . A single-stained OneComb eBeads ( affymetrix , eBioscience , UK ) for all fluorescence compensation except for the live dead stain were used . For the viability dye , cells rather than beads were stained and used for fluorescence compensation . Forward scatter height ( FSC-H ) versus Forward scatter area ( FSC-A ) plots were used to select singlets , and FSC-A versus dead cell marker plots identified viable cells . Side scatter area ( SSC-A ) versus FSC-A plots were used to discriminate lymphocytes from monocytes and residual granulocytes . The threshold for FSC was set to 5 , 000 . For each sample , 500 , 000–1 , 000 , 000 cells were acquired . The percentage of B-lymphocytes ( CD19+ ) , Mature B-cells ( CD19+CD10- ) , Nave B-cells ( CD19+CD10-C27-CD21+ ) , resting memory B- cells ( CD19+CD10-CD27+CD21+ ) , activated memory B cells ( CD19+CD10-CD27+CD21- ) and tissue like memory B-cells ( CD19+CD10-CD27-CD21- ) was defined relative to the parent population with FlowJo version 10 ( Tree Star , USA ) using logicle ( bi-exponential ) transformation as recommended [27 , 28] ( Fig 1 ) . Data were exported to Excel spreadsheet for each sample and compiled for further analysis . Differences in percentage of B-cell subsets were analyzed with either the two-tailed Mann-Whitney U test or the Wilcoxon signed rank non-parametric tests using STATA 14 version 2 ( San Diego California USA ) . Graphs were produced by GraphPad Prism version 5 . 01 for Windows ( GraphPad Software , San Diego California USA ) . The median and Hodges-Lehmann estimator were used for result presentation . Hodges–Lehmann is used to measure the effect size for non-parametric data [29] . P-values were corrected for multiple comparisons . The statistical significance level was set at p≤0 . 05 .
The median percentage of B-lymphocytes ( CD19+ ) in patients with ENL ( 9 . 5% ) and LL controls ( 11 . 6% ) was not statistically significantly different at recruitment . During treatment , the percentage of B-cells slightly increased to 10% in patients with ENL and to 14% in LL patient controls but did not show a statistically significant difference . However , after treatment , the median percentage of B-cells appreciably decreased to 5 . 7% in patients with ENL while it was slightly decreased in LL patient controls to 12 . 0% and the difference between the two groups was statistically significantly different ( P≤ 0 . 001; ΔHL = 6 . 02% ) ( Fig 2A ) . The kinetic analysis of B-lymphocytes within ENL group at different treatment time points has shown that the median percentage of B-lymphocytes before and during treatment of patients with ENL was 9 . 5% and 9 . 9% respectively ( P>0 . 05 ) . However , after treatment , the percentage of these cells was significantly decreased to 5 . 7% compared with before and during treatment ( P ≤ 0 . 05 ) ( Fig 2D ) . The median percentage of mature ( CD19+CD10- ) and immature ( CD19+CD10+ ) B-cells in patients with ENL and LL controls before and after treatment were also measured . A significant difference was not observed with regard to the median percentage of mature and immature B-cells in patients with ENL reactions and non-reactional LL patient controls before and after treatment ( Fig 2B ) . Similarly , the median percentage of mature B-cells before and after treatment was not statistically significantly different within ENL group ( Fig 2E ) . The median percentage of naive B-cells ( CD19+CD10-CD27-CD21+ ) in untreated ENL patients was significantly lower ( 76 . 0% ) than in LL patient controls ( 86 . 4% ) ( Fig 2C ) . This implies that more number of B-cells in untreated ENL patients are antigen experienced than those from LL patient controls . On the other hand , the percentage of naive B-cells within ENL group was not significantly changed before and after treatment ( Fig 2F ) . The median percentage of memory B-cell subtypes ( resting , activated and tissue-like memory B-cells ) in patients with ENL reactions and non-reactional LL controls as well as within ENL group was analysed in the unstimulated PBMCs before , during and after treatment . The median percentage of activated memory B-cells ( CD19+ CD10-CD27+CD21- ) was significantly higher in patients with ENL ( 2 . 6% ) than in LL patient controls ( 1 . 4% ) before treatment ( P≤ 0 . 005 ) . During treatment , the percentage of activated memory B-cells ( AM ) in patients with ENL and LL controls increased to 3 . 8% and 4 . 4% respectively and the difference was not statistically significantly different ( P> 0 . 05 ) . After treatment , the percentage of these memory cells did not change ( Fig 3B ) . A comparison within ENL has shown that the median percentage of activated memory B-cells in untreated ENL patients was higher ( 2 . 6% ) than after treatment ( 1 . 3% ) and the difference was statistically significantly different ( P≤ 0 . 005 ) ( Fig 3E ) . Hence , it seems that activated memories B-cell is associated with ENL reactions . The median percentage of resting memory B-cells ( CD19+ CD10-CD27+CD21+ ) was 5 . 8% in patients with ENL and 4 . 8% in LL patient controls before treatment and the result was not statistically significantly different . However , during treatment it was increased to 15 . 2% in patients with ENL and was higher than in LL patient controls ( 8 . 6% ) ( P≤0 . 05 ) . After treatment , the proportion of these memory cells was decreased to below 5% in both groups and was not statistically significantly different ( Fig 3A ) . Analysis within ENL group has shown that the proportion of resting memory B-cells ( RM ) was considerably lower ( 5 . 8% ) before treatment than during treatment ( 15 . 3% ) ( P≤0 . 001 ) . After treatment , the proportion of these resting memory B-cells was decreased to 3 . 7% and it was significantly lower than before and during treatment ( P≤ 0 . 05 ) ( Fig 3D ) . Interestingly , untreated ENL patients had significantly lower ( 5 . 2% ) median percentage of tissue-like memory B-cells ( TLM ) ( CD19+CD10-CD27-CD21- ) than the corresponding LL patient controls ( 10 . 7% ) ( P≤0 . 05 ) . However , after treatment a significant difference was not observed between these two groups ( Fig 3C ) . Similarly , comparison within ENL group has shown that the median percentage of TLM is significantly lower in untreated ENL patients than after treatment ( P≤ 0 . 05 ) ( Fig 3F ) .
Memory B-cells are subtypes of B-cells that are formed within the germinal centres following infection . They proliferate and differentiate into antibody producing plasma cells also called effector B-cells in response to re-infection . Memory B-cells rapidly differentiate into plasmablasts that produce class-switched antibodies which are capable of clearing the infection far more quickly than naive B-cells [3] . The different classes of memory B-cells have been studied in various chronic viral infections such as hepatitis and HIV and in several autoimmune diseases [30 , 31] . The role of B-cells in the pathogenesis of ENL has been speculated in several studies but has never been studied . For the first time , we studied B-cells and the memory B-cell sub-types in patients with ENL and LL controls at different time points ( before , during and after treatment ) to investigate the dynamics of these cells during the course of prednisolone treatment . The percentage of total B-cells was not significantly different in the two groups before treatment . However , after treatment , the proportion of B-cells was significantly reduced from 9 . 5% to 5 . 7% in patients with ENL . The reduction of B-cells after prednisolone treatment of patients with ENL could be either transitory or associated with the subsiding of the ENL reaction which needs further investigation . The success of Rituximab to deplete B-cells for the treatment of rheumatoid arthritis has stimulated investigation of its effects in several other immune disorders , and considerable interest in the potential of drugs that can modulate B-cell function for the treatment of such diseases [32] . Thus , the finding of reduced B-cells after ENL subsides poses the question whether depleting B-cells could be effective in the treatment of ENL . Patients with ENL had a significantly lower naïve B-cells ( 76 . 0% ) than LL controls ( 84 . 6% ) before treatment ( P≤0 . 05; L-H = 6 . 75 ) . It implies that more B-cells are antigen experienced in untreated ENL patients compared to in LL patient controls . However , the proportion of naïve B-cells was still unusually high in spite of the presence of abundant M . leprae antigens in these patients . A significant difference was not observed with regard to the frequency of resting memory B cells ( RM ) in the two groups before treatment . However , the median percentage of activated memory B-cells ( AM ) was significantly higher in patients with ENL ( 2 . 6% ) than in LL controls ( 1 . 4% ) before treatment . Several studies have shown that activated memory B-cells are increased in patients with disease flares in systemic lupus erythematosus ( SLE ) [33] and rheumatoid Arthritis [34] . However , the biology of ENL and autoimmune diseases is different and whether activated memory B-cells are undesirable or not in the pathogenesis of ENL should be further investigated to arrive at a conclusive evidence . Nevertheless , activated B-cells may be primed to plasma B-cells which in turn produce immunoglobulins [30] . These immunoglobulins could interact with the M . leprae antigen and thereby form excess immune-complexes beyond clearance or activated B-cells may serve as antigen presenting cells i . e . presenting M . leprae antigens to T-cells . Depending on the magnitude of antigen presentation , the T-cell response could be excessive and may cause tissue damage . Patients with ENL had lower percentage of tissue-like memory ( TLM ) B-cells ( 5 . 2% ) than LL controls ( 10 . 7% ) before treatment . Several studies have indicated that TLM B-cells represent the exhausted state of B-cells since they express several inhibitory receptors , including the immunoreceptor tyrosine-based inhibitory motif ( ITIM ) -containing inhibitory receptor Fc receptor-like protein 4 ( FcRL4 ) [35] . TLM B-cells show a reduced tendency to proliferate in response to cognate antigen [36] . The expression of FcRL4 on human B-cell lines disrupts immune synapse formation and blocks antigen induced B-cell receptor ( BCR ) signalling [37] . They also express not only FcRL4 but also a number of other inhibitory and chemokine receptors that would reduce the likelihood of B and T cell interaction [5] . It has been shown that a specific siRNA knockdown of FcRL4 and other inhibitory receptors may lead to a rescue of Ig secretion and proliferation in these tissue-like memory B-cells [38] . In contrary to our findings , increased proportion of TLM have been reported in chronic infections such as in hepatitis C virus and malaria infections , and in certain autoimmune diseases [39 , 40] . The decreasing tendency of TLM and increased secretion of activated memory B-cells may indicate the activation of B-cells when LL patients develop ENL reactions . The finding of increased percentage of TLM in LL patients in this study may partly explain why B-cells are unable to control M . leprae multiplication in spite of their abundance in LL patients . The T-cell unresponsiveness in LL may also be associated with the increased production of TLM B-cells . It appears that the higher frequency of TLM B-cells in LL could alter the B-cell/T-cell interaction through blocking B-cell receptors and this hypothesis could be a fertile area for future investigation . Once , this hypothesis is proved , the search for the B-cell immunomodulators that safely overcome this exhaustion phenotype may be necessary in order to develop proper immune response to LL patients . On the other hand , the significant reduction of TLM during ENL reaction suggests the down regulation of inhibitory molecules and thereby increases immune activation in LL patients leading to the onset ENL reaction . Thus , our finding implies that TLM B-cells could have a role in the initiation of ENL reactions which is a new exciting area for further investigation . | Some leprosy patients develop reactions which cause a significant morbidity and mortality in leprosy patients . There are two types of leprosy reactions , type 1 and type 2 reactions . Type 2 or Erythema nodosum leprosum ( ENL ) is an immune-mediated inflammatory complication of leprosy which occurs in lepromatous and borderline lepromatous leprosy patients . The exact cause of ENL is unknown . Immune-complexes and T-cells are suggested as the aetiology of ENL . However , the contribution of B-cells in ENL reactions has never been addressed . In the present study we described the role of B-cell subsets in ENL reaction and compared with non reactional LL patient controls before , during and after corticosteroids treatment . We found increased antigen experienced and activated B-cells in untreated ENL patients compared to those without the reaction ( LL patients ) . This implies that B-cells are associated with ENL pathology . Therefore , the finding provides a ground for future research targeting B-cells to develop effective drug for ENL treatment . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"tropical",
"diseases",
"bacterial",
"diseases",
"b",
"cell",
"receptors",
"neglected",
"tropical",
"diseases",
"memory",
"b",
"cells",
"antibodies",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"t",
"cells",
"pathogenesis",
"immune",
"response",
"antibody-producing",
"cells",
"biochemistry",
"signal",
"transduction",
"cell",
"biology",
"b",
"cells",
"physiology",
"leprosy",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"immune",
"receptors"
] | 2017 | Increased activated memory B-cells in the peripheral blood of patients with erythema nodosum leprosum reactions |
Cell size increases significantly with increasing ploidy . Differences in cell size and ploidy are associated with alterations in gene expression , although no direct connection has been made between cell size and transcription . Here we show that ploidy-associated changes in gene expression reflect transcriptional adjustment to a larger cell size , implicating cellular geometry as a key parameter in gene regulation . Using RNA-seq , we identified genes whose expression was altered in a tetraploid as compared with the isogenic haploid . A significant fraction of these genes encode cell surface proteins , suggesting an effect of the enlarged cell size on the differential regulation of these genes . To test this hypothesis , we examined expression of these genes in haploid mutants that also produce enlarged size . Surprisingly , many genes differentially regulated in the tetraploid are identically regulated in the enlarged haploids , and the magnitude of change in gene expression correlates with the degree of size enlargement . These results indicate a causal relationship between cell size and transcription , with a size-sensing mechanism that alters transcription in response to size . The genes responding to cell size are enriched for those regulated by two mitogen-activated protein kinase pathways , and components in those pathways were found to mediate size-dependent gene regulation . Transcriptional adjustment to enlarged cell size could underlie other cellular changes associated with polyploidy . The causal relationship between cell size and transcription suggests that cell size homeostasis serves a regulatory role in transcriptome maintenance .
The size of cells can vary significantly within an organism , and cells of the same type display pronounced increase in size with increasing ploidy [1] , [2] . During development , specific cell types in many diploid organisms perform endoreplication and differentiate into polyploid cells that are functionally distinct from their diploid progenitors [2] . Polyploidy also occurs as an intermediate state in aneuploid tumor formation [3] and as a mechanism to create substrates for evolution [4] , [5] . From yeast to mammals , polyploidy is associated with enlarged cell size and altered cellular physiology [2] , [6]–[8] . How polyploidy changes physiology is a long-standing question . Furthermore , a causal relationship between enlarged cell size and altered physiology has not been discovered . Yeast offers a unique advantage in studying the physiological consequences of polyploidy , because it is possible to construct isogenic strains of increasing ploidy . There were two previous analyses that compared transcription between cells of different ploidy . The first analysis of transcription in a yeast ploidy series identified a few genes whose transcript abundance in the transcriptome was altered by ploidy [6] . These included some genes that were strongly repressed and others that were strongly induced in polyploids . Although this study established a clear effect of ploidy on transcription , the limited set of identified genes did not reveal a functional relationship between ploidy and gene expression . The scope and sensitivity of this early investigation were hampered by technical limitations . Because the genome sequence of the studied yeast strain ( Σ1278b ) was not known at the time , microarrays designed for a related yeast strain ( S288c ) were employed . Recent genome analysis comparing these two yeast strain backgrounds has revealed many polymorphisms and changes in genomic organization [9] that compromised the power of detection by hybridization in the earlier study . A subsequent analysis of polyploid yeast detected no significant differences between the diploid and tetraploid transcriptomes by microarrays [8] , raising the possibility that the differences found in the first study were strain specific . Alternatively , experimental differences between the two studies could account for the different conclusions . Expanding the second study to compare strains with a greater difference in ploidy ( i . e . , between haploids and tetraploids ) might have uncovered significant transcriptional changes related to ploidy , as was observed in the first study . More importantly , this later study used a different laboratory strain ( S288c ) . Unlike the strain used in the first study ( Σ1278b ) , the S288c strain background does not express FLO11 [10] , the gene that was most affected by ploidy in the first study [6] . We address the issues raised by both of these studies by examining ploidy effects in different yeast strains with more sensitive assays . The recently acquired genome sequence of the Σ1278b strain [9] , combined with advances in transcriptome profiling by RNA-seq [11] , provides the resolution necessary for genome-wide determination of a functional connection among genes regulated by ploidy . The dynamic range of quantitative linearity in RNA-seq is at least 10-fold higher than that of microarrays , making RNA-seq superior at comparing transcript abundance [12] . In this study , RNA-seq enabled identification of a much larger set of differentially expressed genes between Σ1278b haploid and tetraploid strains and the discovery of related genes by Gene Ontology ( GO ) analysis . The enriched GO terms suggested a causal relationship between cell size and gene expression , and this relationship was then confirmed by analyzing gene expression patterns in cells of varying sizes . The genes repressed in large cells of the Σ1278b background were also found to be repressed in tetraploids of S288c , suggesting that the casual relationship between cell size and gene expression is a general feature .
To identify transcripts whose relative abundance in the transcriptome is changed by ploidy , poly ( A ) RNA transcripts isolated from isogenic haploids and tetraploids of the Σ1278b strain background were analyzed by RNA-seq ( Figure 1A ) . Approximately 9 million sequence reads were obtained from each cDNA library , and the majority of reads mapped to annotated ORFs ( Figure 1B ) . Pair-wise comparison of tetraploid samples with haploid samples revealed that ploidy affects the abundance of only a small proportion of the total transcripts ( Figure 1C ) . By comparison with haploids , 35 transcripts were significantly and reproducibly repressed and 30 transcripts were induced in tetraploids ( Figure 1D ) . The differentially expressed genes included several of the strongly regulated genes identified in the previous study on a Σ1278b ploidy series , and the majority of the remaining genes showed consistent regulatory trends in both studies ( Figure 2; Dataset S1 ) [6] . The disproportional expression of these genes appeared unrelated to the cell cycle , since there was no systematic bias for genes expressed in specific cell cycle stages ( Table S1 ) . Interestingly , the genes differentially expressed in tetraploid cells are significantly enriched not for those associated with chromosomes but for those encoding proteins localized to the cell surface ( cell wall , extracellular space , and plasma membrane ) and for genes that encode regulators of cell surface components ( Figure 2; Tables 1 and S2 , S3 , S4 , S5 ) . This compartmental bias suggests that the differential gene expression in tetraploids is not directly caused by an increase in the genome content , but by a difference in cell size/geometry: for a spherical cell , a 4-fold increase in volume corresponds to only a ∼2 . 5-fold increase in surface area . In other words , although tetraploid yeast cells are 4-fold larger in volume than haploid cells [13] , the ratio of surface area to volume is smaller in tetraploids than in haploids . Reduction in surface area relative to volume is likely to trigger differential regulation of components associated with the cell surface , where signaling and transport processes take place dynamically . A reduction in relative cell surface area could alter interactions between surface and cytoplasmic signaling pathway components and affect the cell's ability to transport metabolites across the plasma membrane . Either type of perturbation caused by a reduced surface area relative to volume could alter gene expression in enlarged cells . The relationship between cell size and transcription could be assessed by examining gene expression levels in haploid mutants with altered cell size relative to wild type ( WT ) . Cell size mutants ( Figure 3A ) previously identified in a genome-wide study [14] were selected because they effectively enlarge cell size without significantly affecting fitness or cell shape . In addition , the underlying mutations have no reported functional relationship with the differentially expressed genes identified in tetraploid cells . We considered additional mutations to increase cell size in haploids [15] , [16] but did not pursue them because of technical concerns that would preclude a clear interpretation of experimental results ( Text S1 ) . If differential expression of a gene in tetraploids is caused by enlarged cell size rather than directly by higher ploidy , the gene should be differentially expressed in haploid size mutants as well . Furthermore , the magnitude of change in the gene's expression level should , ideally , correspond to the magnitude of change in cell size . This cell size–transcription hypothesis was initially tested by investigating the effect of cell size on expression of FLO11 , a gene encoding a cell surface glycoprotein [17] , [18] . FLO11 showed the highest degree of repression in the tetraploid ( Figure 2A ) , providing a wide range of detection for changes in transcript levels . Expression levels of FLO11 were measured in WT Σ1278b and isogenic size mutant haploids ( Figure 3A ) treated with nocodazole for cell cycle arrest in M-phase , when the FLO11 transcript is most abundant ( Figure 3B ) . Because the size mutants manifest an altered cell cycle [1] , [14] , arresting the cell cycle was necessary to separate the transcriptional effect of cell size from that of cell cycle . We analyzed mutant alleles of the CLN3 gene that altered cell size significantly and arrested efficiently in the presence of nocodazole . The CLN3-2 mutant arrested as small cells ( 66% of WT volume ) , whereas the cln3Δ mutant arrested as large cells ( 185% of WT volume ) ( Figure 3C ) . In this haploid size series , we found an inverse correlation between FLO11 expression and cell size: FLO11 transcript abundance is highest in the small CLN3-2 haploid and lowest in the large cln3Δ haploid , the same relationship observed between FLO11 expression and cell size in haploid versus tetraploid cells . Expression of FLO11 was also significantly repressed in the bck2Δ and eap1Δ mutants that displayed enlarged cell size at 124% and 137% of WT volume , respectively ( Figure 3D and 3E ) . The reduced expression of FLO11 in these large mutants mimics the down-regulation of FLO11 in tetraploids ( Figure 2A ) . The results from the haploid size mutants demonstrate an inverse correlation between FLO11 expression and cell size , and this correlation is independent of ploidy . To see whether other genes differentially expressed in tetraploid cells were also influenced by cell size , we used quantitative PCR ( qPCR ) to compare expression of these genes in WT and cln3Δ haploids . Among the mutants we examined , the cln3Δ haploid displayed the most pronounced change in cell size and arrested in M-phase with efficiency most similar to WT ( Figure 3; Table S6 ) . The majority of differentially regulated genes in the tetraploid were regulated in the same direction in the cln3Δ haploid ( Tables 2 , S7 , and S8 ) , especially the top-ranking genes , i . e . , those that displayed the strongest differential expression in tetraploids ( Figure 2A and 2B ) . These top-ranking genes likely represent those that respond most robustly to changes in cell size , since the cln3Δ haploid ( 185% of WT haploid volume ) is still much smaller than the tetraploid ( 400% of WT haploid volume ) . Notably , the fold changes in expression levels of the top-ranking genes appeared to correlate with the increase in cell size: they were smaller in the cln3Δ haploid and larger in the WT tetraploid ( Tables S7 and S8 ) , a trend consistent with a functional relationship between cell size and gene expression . To determine whether the magnitude of change in transcription correlates with the magnitude of change in cell size , we compared expression of the top-ranking size-responsive genes in enlarged haploid mutants and the WT tetraploid . Expression levels in each enlarged strain were measured by qPCR and normalized to those in an isogenic WT haploid . The juxtaposed datasets show a negative correlation ( for repressed genes ) or a positive correlation ( for induced genes ) between gene expression levels and cell size ( Figures 4 and 5; Table S9 ) . The results indicate that the differential regulation of these genes is enhanced with increasing cell size . This observation strongly supports our cell size–transcription hypothesis and the idea that incremental changes in cell size can be sensed by the cell and lead to incremental transcriptional responses . GO analysis indicated that many of the genes repressed by large cell size are regulated by the mating and the filamentation mitogen-activated protein kinase ( MAPK ) pathways ( Table S3; see complete GO analysis results in Dataset S1 ) . Analysis of transcription factor binding motifs also suggested that these MAPK pathways mediate differential gene regulation in response to cell size: the binding motifs of Dig1 and Ste12 , transcription factors that function in both pathways , were significantly enriched ( Table S10; Dataset S2 ) . Ste12 is a transcriptional activator crucial for mating and filamentation [19] . Dig1 mediates transcriptional repression in both cellular processes by inhibiting the activity of Ste12 [20] , [21] . When either of the MAPK pathways is active , Ste12 is phosphorylated by the MAPK and released from inhibition by Dig1 . We compared transcription in haploid and tetraploid cells of a different strain background ( S288c ) to see whether the mating pathway , which is conserved among different species of yeast [22] , was affected by cell size in this background as it is in Σ1278b . The genes downstream of the mating pathway were differentially repressed in the S288c tetraploid ( Figure 6A ) , suggesting that the effect of cell size on the mating pathway could be a general characteristic in yeast . We also constructed isogenic haploid and tetraploid S288c cells expressing FLO11 [10] and found that expression of FLO11 was significantly repressed in this tetraploid strain ( Figure 6B ) , a result consistent with our finding in Σ1278b . Although genes up-regulated in large cells were significantly enriched for those containing the binding motifs of Ace2 , Swi5 , Rfx1 , and Yap7 in their promoters ( Dataset S2 ) , this group gave no obvious clues concerning the molecular pathway causing their differential regulation ( Text S2 ) . Moreover , when the transcription of this group of genes in haploids and tetraploids of the S288c background was compared , there was no difference in the levels of their expression ( data not shown ) . This difference between the two yeast strains probably reflects the many regulatory differences between them [9] ( Text S2 ) . To understand the roles of the mating and the filamentation MAPK pathways in mediating size-dependent gene regulation , we disrupted signaling in these pathways in enlarged cells by making mutations in key pathway components . In the absence of the transcriptional repressor Dig1 , several size-repressed genes were less repressed in large cells and became insensitive or much less responsive to size enlargement ( Figure 7A ) . The reduced effect of cell size on gene expression in the absence of Dig1 shows that this transcription factor is involved in gene regulation by cell size . Assessment of the role of the MAPK pathways required a double mutant lacking both Fus3 and Kss1 , MAPKs of the mating pathway and the filamentation pathway , respectively . The double mutant was necessary as these MAPKs have partially overlapping functions as transcriptional repressors [23] . The kss1Δ fus3Δ double mutation reduced the effect of enlarged cell size on the transcription of downstream genes ( Figure 7B ) . Results from the dig1Δ and kss1Δ fus3Δ mutants suggest that reduced activities in the mating and the filamentation pathways contribute to differential gene regulation in large cells .
Cell size homeostasis has been intensively studied and shown to be controlled by a complex coordination of cell growth and cell division [1] , [14] , [24]–[28] . However , the functional significance of this intricate and efficient maintenance of cell size has not been addressed . Moreover , when the genome is duplicated , as in polyploids , the accompanying increase in cell size is maintained upon cell division . Such whole genome duplication has been invoked to explain evolution , development , and diseases [2]–[5] , with little attention to the transcriptional consequences of the enlarged cell size that accompanies polyploidy . To explore the functional relationship between enlarged cell size and gene expression , we first profiled the transcriptomes of isogenic Σ1278b haploid and tetraploid strains by RNA-seq . This strategy revealed a more complete catalog of the genes influenced by ploidy than was possible to achieve in a previous study [6] . Genes encoding cell-surface-related proteins were overrepresented , suggesting an effect of cell size on gene expression . The top-ranking genes whose transcription was down-regulated by an increase in genome size were also down-regulated in another laboratory strain , S288c , that has significant physiological differences from Σ1278b [9] . The fact that these genes behaved similarly in two different strains supported the hypothesis that cell size was responsible for the transcriptional effects . This cell size–transcription hypothesis was plausible because the volume of yeast cells increases proportionally with ploidy [13] . Moreover , as the cell surface area relative to cell volume decreases with increasing cell size , the reduction in cell surface area could alter gene expression by affecting cell-surface-related signaling molecules and impairing molecular transport across the plasma membrane . The cell size–transcription hypothesis was supported by an independent assessment that measured gene expression in haploid mutants that make large cells . These large haploid mutants also showed a causal relationship between cell size and gene transcription , indicating the existence of a size-sensing mechanism that alters transcription independently of ploidy . An alternative model , in which polyploidy and size-altering mutations change transcription of size-regulated genes , whose altered expression then enlarges cell size , is unlikely . The set of size-regulated genes we identified do not themselves regulate cell size [14] . Moreover , the haploid size mutations and tetraploidy enlarge cell size by different mechanisms [1] , [13] and affect different cellular processes [29]–[32] . The similar transcriptional effects observed in these physiologically different contexts of increased cell size and the enhancing of transcriptional changes with increasing cell size support our hypothesis that cell size sensing is involved in differential transcriptional regulation . Our analysis of the transcriptional data suggests that the cell size signal may be transmitted by the mating and the filamentation MAPK pathways . Genes regulated by these pathways in Σ1278b were preferentially down-regulated in large cells and composed the most significant category in GO analysis . A previous study did not detect differential regulation of these genes , as the strains employed ( MATa/alpha S288c ) were inactive for mating and filamentation [8] . We showed that in S288c strains of a suitable mating type and with a greater difference in ploidy , genes downstream of the mating pathway were also repressed . Genetic disruptions in the mating and the filamentation MAPKs as well as their common downstream transcriptional repressor DIG1 reduced the effect of cell size on target gene expression . These results confirmed a decrease in activities of the mating and the filamentation pathways in large cells . The switch-like dual functions of the MAPKs [23] , [33] and the positive feedback loops involving downstream transcription factors [34]–[37] likely exacerbate differences in pathway activity between the active state ( in small cells ) and the inactive state ( in large cells ) . These attributes of the pathways may account for a nonlinear transcriptional response to changes in cell size . Consequently , the magnitude of change in gene expression did not appear to correlate with cell size or cell surface area in a simple fashion ( Table S9 ) . Although the exact signal that initiates a size-dependent change in transcription is not known , these MAPK pathways have an architecture that is well suited to transmitting a signal from the cell surface to the nucleus . In both MAPK pathways , plasma-membrane-bound G proteins recruit and activate the MAPKKK upon stimulation by the mating pheromone or by nutrient starvation . Through a series of further protein–protein interactions , the MAPKKK in turn activates downstream kinases including the MAPK , which then translocates from the cellular periphery to the nucleus to induce gene expression [38] . An enlarged cell size could affect one or more of the molecular events in the process of pathway activation . Because the nuclear size is proportional to overall cell size in yeast [39] , [40] , both the nuclear surface and the cell surface experience a reduction in area relative to the enclosed volume in large cells . The reduced relative nuclear surface area could impair translocation of the MAPKs . The reduced relative cell surface area could affect the interactions between plasma-membrane-bound and cytoplasmic components in the pathway . In summary , we showed that polyploidy-associated differential gene regulation is largely caused by an increase in cell size . The newly uncovered regulatory relationship between cell size and gene expression suggests that the uniformity of cell size in unicellular organisms and within tissues in multicellular organisms could be necessary to maintain the homeostasis of transcription . Our finding also suggests that cells monitor their geometric properties ( i . e . , size and shape ) and adjust transcription accordingly . These physical features have not been typically considered a regulatory factor in cellular biology , especially in gene expression studies . In metazoans , the control of gene expression by cell size could contribute to the altered development of large or polyploid cells in normal tissues or to the aberrant physiology of tumor cells .
Yeast strains used in this study are listed in Table 3 . Strains L6437 ( WT MATa haploid ) and L6440 ( WT MATa tetraploid ) were grown in synthetic complete medium plus 2% glucose at 30°C until mid-log phase for transcriptome profiling by RNA-seq . Cells were handled with caution to minimize passaging in order to avoid aneuploidy in the tetraploid . For cell cycle arrest using nocodazole , cultures were inoculated at low density in yeast extract peptone dextrose ( YPD ) plus 1% DMSO from overnight precultures and incubated at 30°C . After a few hours , the cultures in exponential phase were diluted to ∼0 . 15 O . D . 600 in prewarmed fresh medium and incubated for another 30 min . Nocodazole was added to a final concentration of 15 µg/ml to arrest cell cycle for 3 h . Enrichment of cells in M-phase was monitored by SWI5 RNA transcript abundance and by counting the percentage of large budded cells with DAPI-stained nuclei at the mother–bud junction . To compare gene expression levels in isogenic strains at different ploidies , cells were cultured in YPD at 30°C until mid-log phase . Cells were collected by centrifugation for RNA extraction and microscopy . Total RNA was extracted from yeast cultures in mid-log phase with acidic phenol . After enrichment of poly ( A ) RNA ( Qiagen Oligotex mRNA kit ) , the resultant mRNA was processed for cDNA library construction and sequencing as previously described [11] . The libraries were sequenced for 36 cycles on Illumina Genome Analyzer 2 using the standard protocol . Reads were mapped to the Σ1278b genome using the Bowtie alignment software ( version 0 . 10 . 0; http://bowtie-bio . sourceforge . net/index . shtml ) . Reads were either mapped uniquely ( bowtie –solexa-quals -k 1 -m 1 –best –strata -p 2 –strandfix ) or multiply ( bowtie –solexa-quals -k 100 -m 100 –best –strata -p 2 –strandfix ) . We used unique mappings only to look at differential gene expression . Multiple mappings were used to assess how many reads aligned to TY elements , the rDNA cluster , and other repetitive sequences; a read that mapped to n genomic locations was assigned a weight of 1/n , and the “number of reads” mapping to a repetitive element was the sum of the weights of the hits in that element . The complete RNA-seq data are available at the Gene Expression Omnibus ( http://www . ncbi . nlm . nih . gov/geo/ ) repository with the accession number GSE19685 . Mapped reads were stored in the David K . Gifford group's in-house ChIP/RNA-seq database and analyzed with the code provided in DifferentialExpression . java . This code performs the following procedure on each annotated ORF in the Σ1278b genome . ( 1 ) Determine the total number of uniquely mapped reads in the haploid and tetraploid experiments . ( 2 ) Determine the number of uniquely mapped reads on both strands in the ORF in the haploid and tetraploid experiments . ( 3 ) Compute frequency_haploid = ( haploid count for gene ) / ( total haploid count ) and frequency_tetraploid = ( tetraploid count for gene ) / ( total tetraploid count ) . ( 4 ) Use frequency_haploid to compute a p-value for the observed reads in the tetraploid experiment using a binomial model given the frequency in the haploid experiment . That isThe CMF is the cumulative mass function ( the discrete equivalent of a cumulative distribution function ) and is the sum of the probabilities for all counts less than or equal to the observed count . This is the p-value for the haploid observation given the tetraploid observation . ( 5 ) Compute the p-value for the tetraploid observation given the haploid observation . ( 6 ) Retain genes with p<0 . 001 . The worksheets “sample pair A” and “sample pair B” in Dataset S1 show the results . Based on read counts of known silenced genes ( hypoxia response and sporulation specific ) , a threshold of 15 was set as the minimal expressed level . In total , 5 , 613 genes were considered expressed and constituted the “background gene list” for subsequent GO analysis on Saccharomyces Genome Database ( SGD ) . The list of differentially expressed genes with read count of 15 or greater provided the set of candidate genes . The differentially regulated candidate genes were sorted by fold change after gene read counts had been normalized by the total number of reads in each dataset . Equal numbers of the top-ranking candidates from the two haploid–tetraploid replicates were compared , and the overlapping candidates were identified as differentially regulated . The numbers of top-ranking candidates from the replicates were selected to obtain a sufficient number of overlapping genes for GO analysis while ensuring that the overlap between replicates was highly statistically significant ( p<e−10 ) by hypergeomtric test using MATLAB ( MathWorks ) . Given the list of differentially expressed genes , we scanned the upstream promoter region for all motifs published [41] . For each gene , we determined the promoter region as the region ( 1 ) extending 50 bp downstream of the annotated coding start site and ( 2 ) extending upstream of the annotated coding start site to the first annotated feature ( ORF , Ty , tRNA , etc . ) or 5 kb away . Features that overlap the transcription start site are not used as the first upstream feature . For each motif , we determined the numerical score of the log-likelihood matrix match against all genes as the best match of the motif to the upstream promoter . Each log-likelihood matrix has a positive maximum score representing the highest possible value that the matrix can assign to a DNA sequence seen while scanning; as these are log-likelihood matrices , a score of zero indicates that the sequence in question matches the background model as well as it matches the matrix . The motif-scanning code then tested cutoff scores between 0 . 3 and 1 . 0 ( in increments of 0 . 05 ) times the maximum possible score for the motif to determine the score to use to call a motif “match”; the score cutoff used was the one that produced the most significant result when comparing the motif frequency in genes up-regulated in tetraploid against motif frequency in all genes using a binomial test . We performed a similar process on the down-regulated genes . We filtered the results to include only motifs that showed a p-value of 0 . 005 or less , were found in at least 25% of up-/down-regulated genes , and for which the fold change in frequency was at least 1 . 5 . We used these filters to retain motifs mostly likely to be relevant to the lists of identified genes . Dataset S2 shows the results of the motif scanning . Total RNA extracted using acidic phenol was processed for cDNA synthesis using the QuantiTect Reverse Transcription Kit ( Qiagen ) . Expression levels were measured on an Applied Biosystems 7500 Real-Time PCR System with SYBR Green in Absolute Quantification mode following manufacture's procedure . Unless specified , we used the abundance of ACT1 transcript to normalize expression levels of genes of interest . The representation of ACT1 transcript in total RNA is constant in all strains used in this study . Statistical treatment ( unpaired t test for two-tailed p-values ) of qPCR data was performed using GraphPad Prism ( GraphPad Software ) . Cells were fixed in 3 . 7% formaldehyde at 4°C overnight and digested with a mixture of zymolyase and glusulase in the presence of 1 . 2 M sorbitol citrate to relieve aggregation . Microscopy images of more than 50 cells per strain were analyzed using ImageJ ( United States National Institutes of Health ) . In experiments involving cell cycle arrest with nocodazole , cell size was calculated from the measured width and length of both mother and bud of large budded cells , assuming rotational symmetry about the long axis . For actively cycling cultures , cell size was calculated from the measured width and length of the mother of budded cells . Statistical comparison of cell size was performed using Student's t test . Haploid strains with suitable genotypes were transformed with a plasmid encoding the HO endonuclease inducible by galatose to enable mating-type switching . Transformed strains were pre-grown overnight in synthetic complete drop-out medium supplemented with 0 . 1% glucose . After sufficient washing with water , cells were resuspended in synthetic complete drop-out medium plus 2% galactose and grown for approximately one doubling time to enable mating-type switching before plating on YPD agar . Mating-type-switched candidates were passaged multiple times on YPD to ensure loss of the HO-encoding plasmid prior to mating-type assessment by auxotrophic marker complementation or a pheromone-dependent growth inhibition assay . | Cells of the same type , whether microbial , plant , or metazoan in origin , exhibit remarkable uniformity in size . This uniformity arises from control mechanisms that respond to internal cellular changes as well as external environmental factors . Although precise control of cell size is a universal phenomenon , its relationship to cellular physiology is underexplored . In this study using yeast we show a causal relationship between cell size and gene regulation: changes in cell size correlate with changes in the expression of a set of genes . Hence , the maintenance of uniformity in cell size could be a homeostatic mechanism for the maintenance of gene expression in a cell or in a population of cells within a tissue . The relationship between cell size and gene expression uncovered in this study may have fundamental implications in evolution , in the development of multicellular organisms , and in the formation of tumors , as these processes often involve genome duplication accompanied by enlarged cell size . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"cell",
"biology",
"cell",
"biology/gene",
"expression",
"microbiology"
] | 2010 | Control of Transcription by Cell Size |
While it is appreciated that population size changes can impact patterns of deleterious variation in natural populations , less attention has been paid to how gene flow affects and is affected by the dynamics of deleterious variation . Here we use population genetic simulations to examine how gene flow impacts deleterious variation under a variety of demographic scenarios , mating systems , dominance coefficients , and recombination rates . Our results show that admixture between populations can temporarily reduce the genetic load of smaller populations and cause increases in the frequency of introgressed ancestry , especially if deleterious mutations are recessive . Additionally , when fitness effects of new mutations are recessive , between-population differences in the sites at which deleterious variants exist creates heterosis in hybrid individuals . Together , these factors lead to an increase in introgressed ancestry , particularly when recombination rates are low . Under certain scenarios , introgressed ancestry can increase from an initial frequency of 5% to 30–75% and fix at many loci , even in the absence of beneficial mutations . Further , deleterious variation and admixture can generate correlations between the frequency of introgressed ancestry and recombination rate or exon density , even in the absence of other types of selection . The direction of these correlations is determined by the specific demography and whether mutations are additive or recessive . Therefore , it is essential that null models of admixture include both demography and deleterious variation before invoking other mechanisms to explain unusual patterns of genetic variation .
There is tremendous interest in quantifying the effects that demographic history has had on the patterns and dynamics of deleterious variation and genetic load [1–8] . Several studies have suggested that recent human demography has had little impact on load [9 , 10] while others have suggested weak , but subtle , differences between human populations [11–15] . All of these studies have typically focused on how population size changes , such as expansions and bottlenecks , have affected deleterious variation . Other types of complex demography , however , have received considerably less attention . In particular , gene flow may be important for shaping patterns of deleterious variation . Population admixture , or hybridization between closely related species , appears to be quite common in nature [16] and has had a significant role in shaping human genomes [17] . Gene flow alone can subtly change the effects of selection on deleterious variation [13] , but should have notable fitness consequences if deleterious variation is distributed differently between admixing populations . For example , Neanderthals likely had a higher genetic load than coincident human populations due to the former’s smaller long-term population size [18 , 19] . As a result , it is thought that gene flow from Neanderthals into the ancestors of modern humans could have increased the genetic load of some human populations by 0 . 5% [18] , and that linked selection removed much of Neanderthal ancestry from humans since that time . In contrast , domesticated species likely have increased genetic load due to domestication bottlenecks and hitchhiking of deleterious alleles with artificially selected variants [20–22] . Gene flow from wild populations could alleviate the genetic load of domesticated species , and increases in the frequency of wild-population ancestry should be observed in the domesticated population [23] . Such changes in patterns of introgression are important to consider when studying how natural selection shapes the evolution of hybrid ancestry , a major goal in evolutionary biology . Differences in the distribution of deleterious variation between hybridizing populations is one reason why natural selection may shape the evolution of hybrid ancestry . Hybridization can also decrease the fitness of a population , for instance , if the parent lineages have diverged significantly and evolved genomic incompatibilities , or if parent lineages have evolved under unique and strong selective pressures in different environments . In both cases , linked selection removes hybrid ancestry especially in regions of low recombination and high functional density [24–26] . This creates genome wide , negative correlations between the local recombination rate , or functional density , and the frequency of introgressed ancestry , a pattern that is observed in humans [24 , 26 , 27] , swordtail fish [26] , and mice [28] . The similar outcomes of both these processes mean that models of selection on deleterious variation should be considered before interpreting genomic patterns of introgression as evidence of divergence and speciation . Another complication to studying the effects of deleterious mutations on introgression is that strongly deleterious new mutations are more likely to be fully or partially recessive [29–31] . Furthermore , dominance coefficients vary between species . New mutations in humans [14] are more likely to be additive than new mutations with the same selection coefficient in Arabidopsis [31] . If some proportion of deleterious recessive variants is private to a population , admixed populations could experience heterosis when recessive variants are masked ( heterozygous ) in hybrid individuals [32] . As a result , heterosis may participate in a tug-of-war on hybrid ancestry with additive variants by increasing the frequency of linked ancestry [18] , increasing apparent migration rates in regions linked to selected variants [33 , 34] , particularly when gene flow occurs in a highly structured population [35] . Heterosis should also increase the probability that introgressed ancestry will persist in an admixed population , even if the introgressed ancestry contains more deleterious alleles [18] . Given the extent to which hybridization is thought to be common to all species [16] , with levels of shared polymorphism in taxa such as Arabidopsis motivating arguments for the bifurcating species concept to be revoked [36] , it is crucial to understand the contribution of heterosis to patterns of hybrid ancestry . Hybridization also transfers novel adaptive variants between evolutionarily distinct lineages [37] . In humans , many Neanderthal variants are thought to be adaptive [38] , possibly affecting phenotypes such as skin pigmentation [39 , 40] , the response to oxygen levels at high altitudes [41] , and immunity to pathogens [42 , 43] . In this case , the introduction of beneficial alleles via gene flow will also oppose the effect of linked selection from deleterious variation , since introgressed ancestry would increase in frequency by hitchhiking with adaptively introgressed variants . Interestingly , North American populations of Drosophila melanogaster exhibit an overall enrichment for introgressed African ancestry in genomic regions of low recombination [44 , 45] . The divergence time between these two D . melanogaster populations is small , and so selection on hybrid individuals may be driven by adaptive variants that arose over shorter time scales than genomic incompatibilities . On the other hand , no correlation between recombination rate and introgression is observed in invasive Californian sunflowers [46] . How selection against additive deleterious variation , selection for adaptive variants , and heterosis interact to determine these genomic patterns is unknown . The objective of this study is to develop a clearer idea for null models of the dynamics of introgression in hybridizing populations while considering the effect of deleterious variants on fitness . Specifically , we aim to understand how selection on introgressed ancestry is determined by differences in the effective population size , mating system , genome structure , recombination rate , distribution of fitness effects , and distribution of dominance coefficients . Previous simulation and empirical work have shown that for at least some systems , deleterious variation is a significant modulator of gene flow [18 , 19 , 23 , 25 , 26] , but few studies have investigated these questions outside of demographic models specific to a system . This study presents a series of simulations utilizing demographic models that generalize biological scenarios of interest by borrowing population genetic parameters and genomic structure from humans and Arabidopsis thaliana , two markedly different organisms with markedly different population genetic parameters . We include realistic distributions of fitness effects and simulate under various models of dominance . In addition , we examine how the relationship between the genomic landscape of introgressed ancestry and recombination rates or functional content is determined by the underlying demography .
We used SLiM 3 . 0 [47] in conjunction with tools from pyslim [48] and msprime [49] to simulate a series of five models of admixture in the presence of deleterious variation . Each of the five models was based on a divergence model where an ancestral population at equilibrium splits into two subpopulations . At some time after the split , a single pulse of admixture occurs at a proportion of 5% , in one direction and for a single generation . Due to practical considerations only an initial admixture proportion of 5% was simulated . Fig 1 provides a cartoon representation of these models and the specific model parameters can be found in S1 Table . All simulated sequence included genic structure ( exon/intron/intergenic regions ) , which was either randomly generated or incorporated from a reference genome as described in the following sections . Only new nonsynonymous mutations were assigned nonzero selection coefficients , which were drawn from a gamma distribution of fitness effects ( DFE ) with shape parameter 0 . 186 and average selection coefficient E[s] = -0 . 01314833 [50] except when specified otherwise . In other words , no positively selected mutations were simulated . Throughout , we will refer to the subpopulation that migrants originate from as the donor subpopulation , and the subpopulation that migrants join as the recipient subpopulation . Furthermore , we will refer to ancestry in the recipient subpopulation that originated in the donor subpopulation as introgression-derived ancestry . We use pI to denote the total proportion of ancestry that is introgression-derived in the recipient subpopulation . See the Methods for additional details on the simulations . To better understand how deleterious variants shape patterns of introgressed ancestry , we first simulated small genomic segments with randomly generated genic structure , of length ~5 Mb and selection coefficients from a gamma DFE . Two hundred simulation replicates using each of the 5 demographic models in Fig 1 ( parameters in S1 Table ) , each of the per base pair recombination rates r = 10−6 , 10−7 , 10−8 , and 10−9 , and additive ( h = 0 . 5 ) or recessive ( h = 0 . 0 ) fitness effects were generated , for a total 8 , 000 independent replicates . In the 20 , 000 generations between the population split and admixture event , deleterious mutations accumulate at different rates across subpopulations for each unique model ( S1 Fig ) , illustrated by the relative difference in subpopulation fitness in Fig 2 . We report subpopulation fitness while ignoring the deleterious variants that have fixed in both subpopulations , since selection will not act on globally monomorphic variants . Because some weakly deleterious variants will fix in one subpopulation yet be lost in the other , each subpopulation’s fitness also steadily decreases through time . In the additive fitness model , this relative difference in fitness is simply determined by relative differences in subpopulation size . When there are no differences in subpopulation size ( Model 0 ) , the fitness of both donor and recipient subpopulations decreases at approximately the same rate ( wR ≈ wD , Fig 2 ) . A similar pattern is observed for a short bottleneck in the recipient population ( wR ≈ wD , Model 1 , Fig 2 ) , reflecting the insensitivity of additive genetic load ( measured in terms of the number of deleterious variants per haplotype in S2 Fig ) to short-term changes in Ne [9] . In contrast , long-term differences in population size ( Models 2–4 , Fig 2 ) provide enough time for deleterious variants to drift to higher frequency in the smaller subpopulation , resulting in substantial differences ( approximately 5% ) in fitness between subpopulations . When deleterious mutations are recessive , a qualitatively similar relationship between subpopulation size and subpopulation fitness is generally observed . When there are no differences in population size ( Model 0 ) , the fitness of donor and recipient subpopulations decreases at a similar rate ( wR ≈ wD , Fig 2 ) . A short bottleneck in the recipient population ( Model 1 ) increases the frequency of homozygous , recessive genotypes immediately post-bottleneck ( S3 Fig ) which slightly decreases the recipient subpopulation’s fitness immediately before admixture ( Fig 2 ) . Finally , similar to the additive fitness model , long-term differences in population size result in substantial differences ( >10% ) in relative fitness between admixing populations . The recombination rate is a key factor in determining differences in fitness between the two subpopulations . When the recombination rate is low , the fitness of the smaller subpopulation decreases more quickly , reflecting the reduced efficacy of purifying selection in low recombination regions [51] . Relative subpopulation differences in fitness between high recombination ( r = 10−6 ) and low recombination ( r = 10−9 ) simulations are about 2% for the additive fitness model and about 8% for the recessive fitness model . Similar to the manner in which they affect subpopulation differences in fitness , recombination rates interact with demography to determine changes in subpopulation fitness after admixture . When fitness effects are additive , admixture is unlikely to cause immediate and large changes in fitness , while subpopulation differences in fitness lead to gradual changes in fitness over time . If admixing subpopulations have the same fitness ( Models 0 and 1 , Fig 2 and S1 Fig ) , admixture predictably has no impact on the recipient population’s fitness . If donor haplotypes have lower fitness than the recipient ( Model 2 , Fig 2 ) , the recipient population’s fitness is negligibly decreased by admixture ( S2 Fig ) , specifically because relative differences in donor and recipient are small ( <10% ) and the initial frequency of donor ancestry is always 5% . If instead the donor subpopulation has higher fitness ( Models 3 and 4 , Fig 2 ) , recipient fitness is relatively unaffected at the time of admixture but increases over time ( S2 Fig ) as the more fit haplotypes experience directional selection . The velocity and magnitude of these changes depends on the recombination rate , as variants originating from the same subpopulation are generally selected in the same direction , and these variants remain on the same haplotypes when recombination is low . When fitness effects are recessive , admixture instead causes immediate and large changes in fitness as recessive alleles are masked in heterozygous , hybrid individuals ( Fig 2 and S1 Fig ) . The qualitative patterns observed are consistent across all demographic models , but the magnitude of these changes is significantly larger in simulations where the recipient subpopulation has lower fitness . The recombination rate again plays a key role in determining fitness in the recipient subpopulation , with the largest changes in fitness occurring in simulations with low recombination . This occurs because the largest differences in pre-admixture fitness are observed when recombination is low ( Fig 2 ) , but also because the heterozygosity of hybrids is maximized if recombination does not occur between donor and recipient haplotypes . This linkage effect is particularly important as most of the variants under selection should have weak effects , since selection is likely to prevent strongly deleterious variants from drifting to high frequency even in a small population . We next explore changes in the frequency of introgressed ancestry ( pI ) over time in the different models . In the additive fitness case , changes in the frequency of introgression-derived ancestry are directly predictable from the differences in subpopulation fitness . When there are no differences in load ( wR ≈ wD , Models 0 and 1 , Fig 2 and S1 Fig ) between mixing haplotypes , selection does not favor a particular ancestry and donor subpopulation ancestry remains , on average , at the initial admixture proportion of 5% in the recipient ( Fig 3 ) . If donor subpopulation haplotypes have lower fitness as in Model 2 ( Fig 2 and S1 Fig ) deleterious donor ancestry is removed by selection , leading to a long-term pI of less than 5% . If instead the donor subpopulation has higher fitness ( Models 3 and 4 , Fig 2 ) , pI is increased above 5% by selection . This increase is greatest ( pI = 75% ) when there is an expansion after the time of admixture and in regions of low recombination ( Model 4 ) . In a recessive fitness model , selection initially favors donor ancestry in the recipient subpopulation . In all cases ( Models 0–4 , Fig 3 ) , the frequency of introgression-derived ancestry increases after admixture , regardless of whether the donor subpopulation’s fitness is less fit or more fit than the recipient . This effect is explained by heterosis , which occurs when recessive deleterious variants are masked as heterozygotes in hybrid individuals ( S3 Fig ) , particularly in the generations immediately following admixture . At this time point , recombination has had little chance to shuffle donor and recipient haplotypes and heterozygosity is maximized in admixed individuals . Again , the recombination rate is a key parameter that determines patterns of introgressed ancestry . As described previously , variants that are selected in the same direction remain linked when recombination is low ( r = 10−9 , Fig 3 ) , maximizing the effect of selection and minimizing selective interference between recombinant haplotypes . When recombination is high ( r = 10−6 ) , the proportion of donor ancestry is unaffected by selection post-admixture ( long-term pI = 5% , Fig 3 ) , as recombination quickly decouples variants under selection from their ancestry backgrounds . Importantly , when recombination rates are low ( r = 10−9 ) , the frequency of introgressed ancestry can increase substantially to up to 75% in the recipient population , despite the initial admixture proportion of 5% . Even with higher recombination rates , when deleterious mutations are recessive and there is a population expansion at the time of admixture ( Model 4 ) , introgressed ancestry can increase up to 25% frequency . So far , we have fixed the split time before admixture at 2N generations , a substantial time for differences in deleterious variation to accumulate between subpopulations . To further examine the relationship between split time and selection on introgression-derived ancestry , we simulated with Models 0 and 4 but also varied the time between the split and admixture ( ts ) . For simulations with a demography analogous to Model 0 , we simulated two divergent populations of equal size . For those analogous to Model 4 , the recipient subpopulation’s size was reduced to 1 , 000 diploids immediately after the split and recovered to the original size at the same time that gene flow occurred . The recombination rate was set to r = 10−9 in these simulations . Fig 4 depicts the long-term proportion of introgressed ancestry , pI , 10 , 000 generations after the admixture event for these two sets of models . We found that across our range of simulated ts , the long-term frequency of introgressed ancestry increases monotonically with ts regardless of the underlying demography . Longer split times result in more deleterious variation being unique to each subpopulation , causing heterosis after admixture as private deleterious variants are masked by introgressed ancestry ( S4 Fig ) . However , these differences appear to reach equilibrium after 20 , 000 generations ( Fig 4 ) , about when most deleterious variants are private to one subpopulation ( Fig 4 ) . We also found as a bottleneck increases in duration , differences in subpopulation fitness become a significant contributor to the increase in long-term pI , but note the apparent equilibrium at 20 , 000 generations . At a split time and thus bottleneck time of >20 , 000 generations , heterosis and differences in load increase long-term pI nearly 2-fold relative the model with no differences in load ( compare Model 0 to Model 4 in Fig 4 ) . When parametrizing the population split times in terms of the realized FST values computed from the SNPs in the simulation output , we find that even for low levels of differentiation ( FST<0 . 05 ) , there is a pronounced increase in introgressed ancestry ( Fig 4 ) . Interestingly , simulations with large long-term pI ( e . g . Model 4 at 1 , 000 generations or Model 0 at 5 , 000 generations ) can have a level of differentiation of FST<0 . 2 at the time of admixture , suggesting that even moderate levels of differentiation between subpopulations are sufficient to drive heterosis in low recombination regions ( Fig 4 ) . So far , we have shown how selection on load shapes introgression-derived ancestry in a set of simple simulations . However , recombination rates and gene density are heterogeneous across actual genomes , and our simulations suggest this variation also could influence the genomic landscape of introgression . To investigate how a realistic genomic structure affects patterns of introgression , we simulated with three of the demographic models described previously ( Models 0 , 2 , and 4 ) using exon definitions and recombination map for a 100 Mb segment of human chromosome 1 . We fixed the exon definitions and recombination map to be the same for all simulations . Only new nonsynonymous mutations were assigned non-zero selection coefficients drawn from a gamma DFE . In addition to simulating both additive and recessive fitness effects separately , we also simulated an inverse relationship between dominance coefficients and selection coefficients , which we will refer to as the h ( s ) relationship , using the function estimated by Henn et al . [14] . We generated 100 simulation replicates for each of the three demographic models . At the end of each simulation , we split the simulated chromosome into non-overlapping 100kb windows and computed the frequency of introgression-derived ancestry , exon density , and the average per base pair recombination rate in each window . The frequency of introgression-derived ancestry generally exhibited genome-wide increases after admixture when mutations were partially or fully recessive and varied in accordance with differences in population size between subpopulations when mutations were additive . In the model with equal subpopulation sizes ( Model 0 ) , we observed no average change in the frequency of introgression-derived ancestry when mutations were additive . When new deleterious mutations were partially or fully recessive , we observed an overall genome-wide increase in the frequency of introgression-derived ancestry ( Fig 5 ) , with many regions reaching high frequency ( >50% ) in single simulation replicates ( S5 Fig ) . This increase in frequency is only due to selection on recessive mutations and local variation in recombination rate , since no positively selected mutations were simulated . In the model where introgressing haplotypes carried a larger deleterious burden ( Model 2 ) and when deleterious mutations were not all recessive , we observed an overall depletion of introgressed ancestry consistent with the effects of purifying selection upon introgressed ancestry ( Fig 5 ) . However , in simulations with fully recessive mutations , the effects of heterosis were strong enough such that many genomic regions showed average increases in frequency of 1 . 5 to 2 times that of the initial introgression frequency of 5% . Importantly , Harris and Nielsen [18] predicted that heterosis would increase the frequency of introgressed ancestry by only a few percent , but our simulations with a similar demographic model show that larger increases in the frequency of introgressed ancestry are possible , especially in exon-dense and low recombination regions . Finally , when we simulated the introgression of haplotypes from a subpopulation with lower genetic load ( Model 4 ) , we observed drastic , genome-wide increases in the average frequency of introgressed ancestry in the recipient subpopulations ( Fig 5 ) as well as many fixed loci in individual simulations ( S5 Fig ) , regardless of whether fitness effects of mutations were additive or recessive . For example , local regions of the simulated chromosome showed an average increase in introgressed ancestry from an initial frequency of 5% up to 50–60% frequency . Furthermore , peaks of introgression are highly correlated between the simulations with different models of dominance , suggesting that the interplay between exon density and recombination strongly affects the way that selection acts on introgressed ancestry in this model . This is the type of signature that is unlikely to be generated under neutral demographic models and could be mistakenly attributed to adaptive introgression . It is also notable that the frequency of introgression-derived ancestry ( pI ) in each window appears to be driven not only by recombination but by exon density , or the local concentration of sites at which deleterious mutations can occur . For recessive mutations , pI is greatly increased on the left-hand side of the simulated chromosome , which tends to be more gene-rich than the right-hand side of the chromosome ( Fig 5 and S5 Fig ) . Importantly , the recombination rate was not significantly correlated with exon density ( Spearman’s ρ = -0 . 0457 , p = 0 . 149 ) in our simulations , showing these factors likely act independently to shape the landscape of introgression . To more formally explore these relationships , we examine the correlations between genomic features and the average frequency of introgressed ancestry across 100 simulation replicates , measured in 100 kb windows ( Figs 6 and 7 ) . In the model of equal subpopulation sizes ( Model 0 ) , the frequency of introgression-derived ancestry is not significantly related to the recombination rate or exon density when mutations have additive effects , but is positively correlated to exon density when fitness effects are fully or partially recessive ( Fig 7 ) . The h ( s ) relationship results in intermediate levels of introgression relative to simulations with strictly additive or fully recessive new mutations . For Model 2 , the frequency of introgression-derived ancestry is positively correlated to the recombination rate and negatively correlated to exon density when fitness is additive . When fitness effects are fully recessive for this model , the frequency of introgressed ancestry is negatively correlated to recombination rate ( middle panel in middle row in Fig 6 ) and positively correlated to exon density ( middle panel in middle row in Fig 6 ) . However , under the h ( s ) relationship , introgression derived ancestry is not significantly correlated to the recombination rate but is correlated with exon density . Lastly , when introgressed ancestry comes from a larger subpopulation with a lower deleterious burden than the recipient subpopulation ( Model 4 ) , the frequency of introgression-derived ancestry is always negatively correlated with recombination rate , and positively correlated with exon density . For Model 4 , these correlations are observed for all models of dominance . Using these same simulations , we examined how selection on deleterious variation after admixture might influence the distribution of introgression deserts , or long stretches of the genome of the recipient population devoid of introgressed ancestry ( S6 Fig ) . When subpopulation fitnesses are expected to be the same ( Model 0 ) , the distribution of introgression deserts for models with deleterious mutations is similar to a neutral model , suggesting that selection does not appreciably impact the distribution of deserts . When introgression-derived ancestry is expected to be deleterious ( Model 2 ) , simulations with additive fitness are enriched for longer ancestry deserts , though only slightly so . If instead introgression-derived ancestry is less deleterious than ancestry in the recipient population ( Model 4 ) , the length distribution of introgression deserts is shifted to be shorter , with the shortest introgression deserts occurring in models with recessive mutations ( h = 0 ) where both selection on load and heterosis act synergistically to increase the frequency of introgressed ancestry . The observation that human X chromosomes are five-fold more resistant to introgression than the human autosome has been interpreted as a signature of genomic incompatibility between Neanderthals and humans , caused by an overrepresentation of male hybrid sterility genes on the X chromosome [24] . However , the evolution of the X chromosome differs from the autosomes in a number of important aspects , particularly in the strength of selection on deleterious variants [52] , which may contribute to differences in patterns of introgression [18 , 19] . It is additionally unclear how selection on recessive variants might contribute , or counteract , the apparent resistance of the X chromosome to introgression . To investigate the expected patterns of introgression on the X chromosome , we modeled X chromosome admixture with the simulation framework previously described . Although we used the same DFE for all these simulations , we utilized an analogous model of fitness that accounts for dosage compensation and the hemizygous sex [52 , 53] . Chromosome structure , recombination rates , and the DFE were the same as the simulations of human chromosome 1 . See Methods for additional details on the calculation of fitness in these simulations . Our simulations show that deleterious variation alone can result in significant differences between introgression on the X and the autosomes ( Fig 8 ) . When fitness is additive , stronger overall selection occurs on the X chromosome because males cannot be heterozygous . This does not affect the X to autosome introgression ratio ( X/A ratio ) for Model 0 , since both populations carry a similar burden of deleterious variants . For Model 2 , selection removes introgressed ancestry from the X more quickly ( X/A < 1 ) , and for Model 4 , selection increases the frequency of introgressed ancestry more on the X than on the autosomes ( X/A > 1 ) . When fitness is recessive , the effect of heterosis is weaker for the X chromosome , since the hemizygous sex cannot be heterozygous . This effect also results in less observed introgression on the X than the autosome ( X/A < 1 ) for all considered models . Finally , under the h ( s ) relationship , our models predict amounts of introgression that are intermediate between strictly additive or strictly recessive models . Human-like demography and genomic parameters may not generalize well for the purpose of understanding introgression in other species . Functional density , recombination rates , effective population sizes , dominance , and the DFE can differ by an order of magnitude between species . To provide an alternative picture of how introgression dynamics are driven by deleterious variation in a natural system where dominance and selection have been estimated , we simulated Models 0 , 2 , and 4 using the genomic structure of Arabidopsis thaliana . While the simulated demography was similar to the ones described previously , we used exon definitions and a recombination map of most ( 29 . 1 out of 30 . 4 Mb ) of A . thaliana chromosome 1 , and chromosome structure was fixed to be the same in all 100 simulation replicates . Both exon density and recombination rates are higher in A . thaliana ( medians of 100kb windows 4 . 8×10−1 and 3 . 2×10-8 , respectively ) than humans ( medians of 100kb windows 1 . 6×10−2 and 8 . 04×10−9 , respectively ) . The ancestral population size was set to NA = 100 , 000 diploids , and the DFE to a gamma distribution with shape parameter 0 . 185 and E[s] = -0 . 0004866 [31] . We also assumed that dominance coefficients followed the h ( s ) relationship estimated by Huber et al . and did not simulate scenarios with only additive or only recessive new mutations . To the best of our knowledge , this is the only estimate of the h ( s ) relationship in a natural population other than humans . We split the simulated chromosome into non-overlapping 100kb windows and computed the frequency of introgression-derived ancestry , exon density , and the average recombination rate in each window . The genomic landscape of introgression in our simulated Arabidopsis population varied little ( Fig 9 ) , even in a single simulation replicate of the same demographic model ( S7 Fig ) . For Model 0 , introgressed ancestry rose quickly from an initial frequency of 5% to about 24% , NA generations after admixture . There was little spatial variation in the frequency of introgression-derived ancestry . For example , pI did not appear to be affected by the paucity of exons near the centromere ( Fig 9 ) . In Model 2 , introgression-derived ancestry was quickly removed from the recipient subpopulation . This meant that pI decreased to 0% across the whole chromosome . The converse was true for Model 4 , where introgression-derived ancestry was favorable , and selection resulted in a complete replacement of recipient population ancestry ( pI = 100% ) . A notable life history feature distinguishing Arabidopsis thaliana from its congeners is the capability to self-fertilize [54] . Populations that are capable of self-fertilization may experience an overall reduced Ne leading to an accumulation of weakly deleterious variants relative to an outcrossing population , and increased levels of inbreeding depression . On the other hand , strongly deleterious recessive mutations should be purged in a selfing population [55 , 56] . Relative differences in the types of deleterious variation between groups with different mating systems may then initiate another kind of selective tug-of-war after admixture . To investigate how deleterious mutations affect levels of introgression when admixture occurs between two populations with different mating systems , we simulated gene flow between a partially selfing and an outcrossing subpopulation using the same A . thaliana parameters as described in the previous section . We limited our simulated demographic model to Model 0 so that any difference in deleterious variation between subpopulations could be attributed to the mating system . Seven different gene flow scenarios were simulated , with selfing probabilities of 0% , 25% , 50% , and 75% in either subpopulation ( Fig 10 ) . Specifically , we simulated: first , with two outcrossing populations ( 0% to 0% ) ; then with the outcrosser ( 0% ) as the donor and the partial selfer ( selfing probabilities of 25% , 50% , 75% ) as the recipient , then the partial selfer ( 25% , 50% , 75% ) as the donor and the outcrosser ( 0% ) as the recipient . Self-incompatibility alleles were not simulated . Our simulations show that the long-term frequency of introgression ( 10 , 000 generations after admixture ) depends on the proportion of selfing individuals in the selfing subpopulation ( Fig 10 ) . In other words , selfing reduces Ne relative to an outcrosser , resulting in increased drift and a greater accumulation of deleterious mutations . These differences in load result in patterns of introgression qualitatively similar to those observed previously in this study . In the simulations between two outcrossing populations , pI increases from 5% to a long-term 20–25% , due to heterosis from the large proportion of recessive mutations predicted by the h ( s ) relationship . This is the same result as the simulations of Model 0 in the previous section . When the outcrosser is the donor , pI increases monotonically with the selfing probability of the recipient , this time above the fraction expected between two outcrossing populations . When the partially selfing population is the donor , long-term pI usually increases by heterosis from the initial 5% value , although the long-term pI monotonically decreases as the selfing probability increases . At a selfing probability of 75% , the outcrossing population is almost completely resistant to introgression . In the absence of fitness epistasis , it is likely that a combination of high recombination rates and strong initial selection from differences in deleterious mutations between populations counteracts any loss of donor ancestry from the purging of strongly deleterious recessive variants .
We have shown through simulations that deleterious variation can greatly influence the dynamics of introgression between admixing populations , in markedly different directions , magnitudes , and manners depending on the demographic model , mating system , models of selection , and genomic structure . In particular , the recombination rate is a key parameter that determines the way in which deleterious variants accumulate between populations and how selection acts on introgression-derived ancestry after admixture , ultimately determining the genomic landscape of introgression . Our work demonstrates how demography can shape patterns of deleterious variation in different populations . Previous studies have examined the role of population size changes [1 , 2 , 9 , 13 , 57] and serial founder effect models [14 , 58] on deleterious variation . Interpreting how differences in the distribution of deleterious variation impact fitness has been a contentious issue [5 , 6 , 8–10 , 12] . In this study , we observed that admixture can increase the fitness of the recipient population , sometimes drastically if the donor population is of larger long-term effective population size and thus carries lower genetic load . Generally , gene flow is observed to drive smaller , subtle changes in fitness . Nevertheless , the influx of new alleles can result in a rearrangement of deleterious variants in an admixed population ( S2 and S3 Figs ) , and subtle changes to fitness can lead to significant shifts in the frequency of introgressed ancestry ( e . g . see Model 0 , h = 0 . 0 , in Fig 3 ) . These effects can be long lasting , persisting for thousands of generations in some of our simulations ( Figs 2 and 3 , S1 Fig ) . If hybridization is a significant feature of a study population , studies concerning load should consider the fitness consequences of admixture as well as population size changes . That dynamics of introgression-derived ancestry can be driven by deleterious variation is also important for the study of selection on gene flow between populations or species . Patterns of introgression between hybridizing species are often asymmetric , vary across the genome , and can be driven by demography at expansion fronts [59] , dispersal processes [60] , or by natural selection . However , when natural selection is implicated as driving changes in introgression-derived ancestry , processes such as genomic incompatibility or adaptive introgression are invoked to explain variation in introgression across the genome . We have shown that differences in demography and mating system create between-population differences in standing deleterious variation , and that selection upon these differences provides an alternative hypothesis to selection on alleles transplanted onto a new genomic background or new environment . To the best of our knowledge , only a few studies have considered the contribution of selection on deleterious variation to observed patterns of introgression [13 , 25 , 33] , and mostly in specific systems [18 , 19 , 23 , 26] . Selection on deleterious variation may be particularly important for determining patterns of introgression in natural populations that are out of demographic equilibrium . Models of increased genetic drift predict accumulations of genetic load at the edges of expanding populations [14 , 58] which suggests introgression into the expanding population could be driven by selection on deleterious mutations . We have also shown that population bottlenecks can greatly affect patterns of introgression , particularly when assuming a recessive fitness model . If recessive deleterious variation also creates heterosis in admixed individuals , the effects of heterosis and population size will be synergistic , further enhancing introgression in genomic regions of low recombination . Our simulations also directly suggest heterosis may contribute to the pervasive patterns of introgression and shared polymorphism between different species in the genus Arabidopsis [36] even if hybridizing species have similar amounts of deleterious variation . Because selection can alter patterns of introgression even if hybrid ancestry is not explicitly deleterious , genome-wide inferences of admixture proportions that assume neutrality are likely to be biased . For instance , our simulations predict the amount of introgression is strongly influenced by deleterious mutations in Arabidopsis , and the manner in which this occurs is dependent on the demography . Observed proportions of ancestry range from 0% for Model 2 to 100% for Model 4 ( Fig 9 and S8 Fig ) , despite the true admixture proportion of 5% . Taking the observed proportion of introgressed ancestry at face value , researchers would not infer the true initial admixture proportion of 5% accurately . Similarly , linkage disequilibrium patterns are often used to infer the timing of admixture events and to test competing demographic hypotheses about admixture [61] . If the distribution of segments of introgressed ancestry can be altered by deleterious mutations relative to what is predicted under a neutral model ( e . g . Model 4 in S7 Fig ) , these inferences can also be biased . To circumvent this problem , we recommend focusing on putatively neutral regions of the genome far from genes . Likewise , our simulations may provide grounds for a plausible alternative explanation for the negative correlation between recombination rate and introgressed African ancestry observed in North American populations of D . melanogaster [44 , 45] , which is the opposite of what is usually observed by other empirical studies of hybridization . Corbett-Detig and Nielsen [45] proposed that widespread adaptive introgression could bring along larger linkage blocks in low recombination regions . If D . melanogaster has accumulated genetic load through the serial colonization of the world in association with humans [62 , 63] , selection may favor introgression of the origin population ( African ) haplotypes in low recombination regions , similar to what we observed in Model 4 of our simulations . This could act synergistically with the effect of heterosis , which can happen in significant amounts even when divergence is low ( Fig 4 ) , and the divergence for which significant increases in introgressed ancestry are observed is comparable to that between populations of D . melanogaster [64] . Admittedly , our models bear little resemblance to the estimated demography of D . melanogaster ( e . g . [63] ) . Similar to humans [9] , there may be little difference in additive load between populations due to recent demography , and we have not simulated with a DFE and model of dominance estimated from D . melanogaster . Further study of these population genetic features is necessary to estimate the relative contribution of these processes to the genomic pattern of introgression in D . melanogaster . Importantly , we do not claim that deleterious variation can explain all the patterns of introgression in any species , but rather that it is a plausible alternative explanation and therefore possible confounder that is important to consider when testing hypotheses about the nature of selection on gene flow . It is alternatively possible that colonizing populations of D . melanogaster experience a reduction in the rate of fixation of adaptive alleles due to reduced Ne , creating favorable conditions for the introgression of parent population haplotypes . Additionally , there is strong evidence for the role of sexual selection and fitness epistasis between the X and the autosomes in separating populations of D . melanogaster [65–67] . In hybridizing swordtail fish , recombination rates are positively correlated with the frequency of introgressed ancestry even when the minor parent population , analogous to the donor population in our simulations , has a larger effective population size [26] . This pattern suggests that hybrid ancestry has an overall deleterious effect , meaning that genomic incompatibility is the dominant force shaping hybrid genomes in that system . In humans , regions of high recombination rate are enriched for introgressed Neanderthal ancestry particularly in genes that code for virus-interacting proteins [43] , suggesting that in these regions putatively adaptive variants were more likely to recombine off the deleterious Neanderthal background and increase in frequency . In these two latter cases , selection on deleterious variation or heterosis may instead obscure genome-wide signals of incompatibility or adaptive introgression . Because selection on additive and recessive variation can act in complementary or opposing directions , our study also highlights the fundamental importance of understanding the distribution of selection coefficients and their relationship to dominance coefficients in natural populations ( i . e . the h ( s ) relationship ) . In this study , we simulated human genomic structure , where new mutations are more likely to have additive fitness effects [14] , and Arabidopsis genomic structure , where deleterious new mutations are likely to be more recessive [31] . In these two scenarios , we found that modes of dominance interacted with demography , recombination rates , and functional density in complex ways . Importantly , we observed an increase in introgressed ancestry as a result of the heterosis effect even when mutations were not completely recessive , that is , dominance was modeled with the h ( s ) relationship . While the effects observed in the present study may be applicable to real populations with realistic amounts of dominance , the h ( s ) relationship is unknown for virtually all natural systems . Therefore , we cannot easily predict the contribution of heterosis to introgression and shared polymorphism between closely related species . Nevertheless , the underlying demographic model will determine how additive and recessive new mutations should interact after gene flow . For example , the introgression of deleterious haplotypes in Model 2 was facilitated by heterosis but impeded by additive load , leading to uncertainty about the overall contribution of the effects of deleterious variation in certain scenarios , such as Neanderthal to human admixture [18] . In other demographic models , selection on additive and recessive variants should operate in the same direction . As another example , if admixture occurs between a partially selfing and outcrossing population , our simulations predict that selection works to remove ancestry from the selfing population , since it carries an overall larger burden of deleterious variants . It may yet be possible that strongly deleterious recessive variants , which should be purged in the selfer , play a role in preventing some introgression from the outcrossing to the selfing population . Without knowing the h ( s ) relationship for a specific system , it is difficult to disentangle the effects of selection on additive versus recessive variation . Our work further highlights the importance of considering deleterious variation when comparing complementary lines of evidence to make inferences about selection on hybrids . Even in the absence of fitness epistasis , our models predict an overall depletion of hybrid ancestry on the X chromosome compared to the autosomes . While the magnitude of this difference ( about 1 . 5-fold ) is far less than the 5-fold difference observed in humans [24] , our results clearly show that simpler models of deleterious variation have the potential to mimic some of the signals that are considered evidence of hybrid incompatibility . Granted , we have only provided a simple model of selection on sex chromosomes to contrast to previous simulations of the autosomes , while ignoring the fact that recombination , chromosome structure , and the DFE are unlikely to be the same between the X and the autosomes . Additionally , it has been shown that sex-biased demographic processes have occurred throughout human history [68–72] . Future work should test the extent to which our results hold across more realistic population genetic models . The recombination rate also plays a key role in determining the landscape of introgressed ancestry in the presence of deleterious variation . Models of Hill-Robertson interference [51 , 73] predict that deleterious mutations will not be removed as effectively in regions of the genome with low recombination rates when weakly selected mutations occur on different haplotypes , since selection on a particular site will weaken selection ( i . e . increase drift ) at other linked sites . We observe this effect in our simulations , where fitness declines the fastest when recombination rates are low , both pre- and post- admixture ( S2 Fig ) . However , we observe the opposite effect immediately after admixture . Specifically , in our simulations , the fitness in the admixed population increased the most for the lowest recombination rates , suggesting that deleterious mutations were most effectively eliminated when recombination rates were the lowest ( S2 Fig ) . This occurs because selection for a haplotype will be most effective when all alleles on a haplotype tend to have weak fitness effects in the same direction [18 , 19 , 26] . For example , if introgression-derived ancestry carries fewer deleterious variants than the other haplotypes in the recipient population , selection will act to increase the frequency of the protective alleles contained within the introgressed ancestry . This applies directly to our simulations of admixture since immediately following an admixture event , all the protective or deleterious variants are found on the same haplotype . Higher rates of recombination will shuffle selected variants onto different haplotypes , creating selective interference between recombinant haplotypes . One significant limitation of our study is that we have not considered all possible combinations of demographic , selective , and genomic parameters relevant for all species . For example , heterosis appears to stabilize long-term patterns of introgression at some frequency , but we only simulated an admixture fraction of 5% . It is possible that the magnitude or direction of observed changes may change with different major and minor parent ancestry proportions . It is therefore difficult to directly assess whether the specific conclusions seen for one combination of parameters will directly apply in a different specific system . Instead , our goal is to highlight the need to consider deleterious variation as a possible null model that should be investigated and rejected before attributing unusual patterns of introgressed ancestry to other evolutionary processes . That being said , we have observed some commonalities across models . For example , in Model 4 , when mutations are either fully recessive or have an intermediate dominance coefficient assigned as a function of the selection coefficient , we observe an increase in introgressed ancestry in the recipient populations when either using simple models ( Fig 3 ) , models relevant for human populations ( Fig 5 ) or models relevant for A . thaliana ( Fig 9 ) . This interplay between deleterious variation and recombination has substantial implications for detecting adaptive introgression . A major objective of genomic studies of hybridization is to identify loci that are adaptively introgressed and to ascertain the overall importance of introgression to adaptive evolution [38] . Genomic regions that contain introgressed haplotypes at high frequency are considered likely candidates for adaptive introgression [38 , 41 , 74 , 75] , but we have shown that selection on deleterious mutations can increase the frequency of introgression-derived ancestry , even in the absence of beneficial new mutations . Thus , outlier-based approaches that compare summary statistics computed for a particular window of the genome to a null distribution that does not account for deleterious variation may be misled . Linked deleterious variants may also impede positive selection on introgressed adaptive variants , particularly if they are recessive [76] . Because recombination can move an adaptive variant off of ancestry backgrounds of varying fitness , standard models of adaptive evolution , especially ones that do not consider deleterious variation , are unlikely to accurately describe genomic patterns generated by adaptive introgression . Finally , it may be difficult to differentiate heterosis due to the masking of deleterious recessive alleles from heterozygote advantage at introgressed loci , despite the fact that these are two very different evolutionary processes with dramatically different biological interpretations . Our results argue that new null models are needed in studies seeking to identify candidates of adaptive introgression . These new null models should include deleterious genetic variation , as well as complex demography . In order for these models to accurately capture the dynamics of deleterious variation , they should also include realistic parameters for the DFE of deleterious mutations and the relationship between dominance coefficients and selection coefficients . Lastly , the new null models should also include realistic models of the variation in recombination rate across the genome , as recombination rate is a key determinant of the dynamics of introgression ( Fig 3 ) . Failure to consider deleterious variation in a realistic way in studies of admixing populations or hybridizing species can mislead inferences about the evolution of hybrids .
All simulations were performed with SLiM 3 . 0 [47] . We chose to discard from our simulations , and therefore from calculations of fitness , mutations that were fixed in the ancestral or both subpopulations . Although fixed deleterious variants contribute to the overall genetic load of finite populations , they will have no effect on the relative differences between admixing subpopulations and no effect on the dynamics of introgression-derived ancestry . Therefore , each fitness calculation does not reflect the true fitness of each population , but rather the fitness components that are relevant during gene flow . An admixture event in SLiM is handled by modifying the way the parents in each generation are chosen ( SLiM manual 5 . 2 . 1 ) . For example , at an admixture proportion of 5% the recipient population reproduces as follows . Five percent of the parents of the recipient population , in that generation , are chosen from the donor population , and 95% of the parents are chosen from the recipient population . We rescaled simulation parameters by a scaling constant , c , to reduce the computational burden of forward simulations . Population sizes were scaled to be N/c , times to t/c , selection coefficients to sc , and the mutation rate to μc . Recombination rates were scaled as 0 . 5 ( 1- ( 1-2r ) c ) , which is approximately rc for small r and small c . The total length of simulated sequence was not changed in scaled simulations . Note , the simulation parameters we reference in this paper are always unscaled . The manner in which we scaled simulations follows Algorithm 1 in Uricchio and Hernandez [77] and is similar to how Lange and Pool [78] simulated populations of Drosophila melanogaster , although the primary features of interest in our simulations are related to the dynamics of introgression-derived ancestry through time . Because scaled simulations may not exactly recapitulate the dynamics of unscaled simulations , we used a set of test simulations to choose c = 5 for most simulations . The dynamics of pI for scaled simulations ( c = 2 , 5 , and 10 ) were compared to an unscaled simulation ( c = 1 ) , using the demography of Model 4 , a gamma DFE , and additive fitness ( h = 0 . 5 ) . Per base pair recombination rates of r = 10−7 and 10−8 were simulated separately . Although all scaled simulations exhibit slight differences from the unscaled simulations , a scaling factor of c = 5 provided a reasonably accurate representation of the unscaled dynamics of pI ( S8 Fig ) while keeping simulation run times within reasonable limits . We additionally note that our intent in this study is to understand qualitative patterns of introgression rather than to obtain accurate quantitative estimates from a particular system , and the qualitative patterns are consistent irrespective of the scaling factor . The proportion of admixture that is introgression-derived ( pI ) was tracked in one of two ways: by placing marker mutations at a fixed interval or by tracking the tree sequences ( genealogies ) across the simulated genome . In the former case , pI was estimated by placing marker mutations in the donor population immediately before the admixture event . These mutations were spaced at 500 base pair intervals over the genome of every individual . After admixture , pI was estimated in the recipient population by taking the averaged allele frequency of marker mutations per window , or throughout the whole simulated chromosome . In the latter case , the true ancestry proportions were calculated , since the information on start/end coordinates and the lineages that trace their ancestry back through donor and recipient populations is preserved . Although tracking tree sequences provides the most accurate estimate of pI , marker mutation tracking was used for computational efficiency in some simulations . The sequences from simulations with randomly generated chromosome structure were approximately 5Mb in length , and contained intergenic , intronic , and exonic regions , but only nonsynonymous new mutations experienced natural selection . The per base pair mutation rate was constant and set to μ = 1 . 5×10−8 and we set nonsynonymous and synonymous mutations to occur at a ratio of 2 . 31:1 [79] . The selection coefficients ( s ) of new nonsynonymous mutations were drawn from a gamma-distributed DFE with shape parameter 0 . 186 and expected selection coefficient E[s] = -0 . 01314833 [50] for both additive and recessive dominance models . The chromosomal structure of each simulation was randomly generated by drawing exon lengths from Lognormal ( μ = log ( 50 ) , σ2 = log ( 2 ) ) , intron lengths from Lognormal ( μ = log ( 100 ) , σ2 = log ( 1 . 5 ) ) , and the length of noncoding regions from Unif ( 100 , 5000 ) , following the specification in the SLiM 3 . 0 manual ( 7 . 3 ) , which is modeled after the distribution of intron and exon lengths in Deutsch and Long [80] . The per base pair per chromosome recombination rate ( r ) was fixed in each simulation , but we varied r between different sets of simulations where r ∈{10−6 , 10−7 , 10−8 , 10−9} . Lastly , we simulated 200 replicates for each set of simulations , of each specific h and r . Chromosome-wide FST was calculated for all variants from exons , introns , and intergenic regions by calculating FST at individual sites following Hudson et al . [81] and by combining FST across sites following Bhatia et al . [82] . In simulations of fixed chromosome structure from the human genome , we fixed the structure to 100 Mb randomly chosen from human genome build GRCh37 , chromosome 1 ( chr1:5 , 005 , 669–105 , 005 , 669 ) . The exon ranges were defined by the GENCODE v14 annotations [83] and the sex-averaged recombination map by Kong et al . [84] , averaged over a 10 kb scale . The per base pair mutation rate was constant and set to μ = 1 . 5×10−8 and we set nonsynonymous and synonymous mutations to occur at a ratio of 2 . 31:1 [79] . The selection coefficients ( s ) of new nonsynonymous mutations were drawn from a gamma-distributed DFE with shape parameter 0 . 186 and expected selection coefficient E[s] = -0 . 01314833 [50] for all models of dominance . All other new mutations were neutral . We simulated additive fitness ( h = 0 . 5 ) , recessive fitness ( h = 0 ) , and the h ( s ) = 0 . 5/ ( 1–7071 . 07s ) relationship [14] separately , using the same DFE for s for each simulation . All simulations were scaled by a factor of c = 5 . In simulations of fixed chromosome structure from the genome of Arabidopsis thaliana , we fixed the structure to 29 . 1 Mb from chromosome 1 ( chr1:488 , 426–29 , 588 , 426 ) . The exon ranges were defined using the Araport11 annotations [85] and the recombination map using Salomé et al . [86] . The per base pair mutation rate was constant and set to μ = 7×10−9 and we again set nonsynonymous and synonymous mutations to occur at a ratio of 2 . 31:1 . The selection coefficients ( s ) of new nonsynonymous mutations were drawn from the gamma distribution estimated by Huber et al . [31] ( shape parameter 0 . 185 and E[s] = -0 . 00048655 ) . We simulated dominance with the h ( s ) relationship estimated by that study: h ( s ) = 1 ( ( 1/0 . 987 ) – 39547s ) . Simulations were scaled at c = 100 , but we note that we could not test the difference between c = 100 and smaller scaling factors ( e . g . c = 50 ) due to limits on computational time . Computing fitness as additive ( h = 0 . 5 ) within a locus but multiplicative across loci was problematic for our simulations because it created heterosis in admixed individuals . This occurred because the product of a fitness decrease reduces fitness less than the sum of a fitness decrease . As a simple example , imagine two additive deleterious alleles are in a single individual , each with selection coefficient s where s is the absolute value of the selection coefficient . If they are found as a single homozygous site , the fitness decrease is usually computed as 1-s . If they are found as two heterozygous sites , the fitness would be computed as ( 1–0 . 5s ) 2 = 1-s+0 . 25s2 . The fitness of the heterozygous individual is larger than the homozygous individual by 0 . 25s2 , despite carrying the same number of deleterious variants . Because admixed individuals are more likely to carry deleterious alleles as heterozygotes than non-admixed individuals , the fitness of the admixed individuals is always higher than a non-admixed individual in the above computation of fitness , even when the number of deleterious variants per individual is the same . Our intent was to examine the contribution of deleterious variation to selection on introgressed ancestry , but we have identified an inherent advantage of heterozygosity in the additive model that biased the direction of selection to favor introgressed ancestry . To address this , we computed heterozygote fitness at a locus as 1-hs and homozygote fitness as ( 1–0 . 5s ) 2 , and the fitnesses across loci were computed multiplicatively . In the additive case ( h = 0 . 5 ) , an individual’s fitness was then multiplicative across all deleterious variants , such that an individual j carrying i variants each with selection coefficient si had fitness wi: wj=∏i ( 1+0 . 5si ) Fitness was then monotonically related to the number of deleterious variants regardless of zygosity while remaining approximately equivalent to additive fitness . This computation in essence created a slight underdominance-like effect , but importantly this effect was caused by the difference in homozygous fitness rather than a difference in heterozygote fitness ( i . e . the dominance coefficient ) . In practice , the difference in homozygous fitness is negligible for weakly deleterious alleles and strongly deleterious alleles are unlikely to be found as homozygotes . Therefore , the overall underdominance effect should be minimal . To confirm this , we simulated 100Mb of human chromosome 1 in an equilibrium population , with selection coefficients drawn from a gamma DFE with the two fitness models . The frequency spectrum was unaffected by our calculation of fitness ( S9 Fig ) , suggesting our simulations approximate the standard additive model well . We used the same calculation for additive and partially recessive fitness models for consistency when simulating the h ( s ) relationship . Completely recessive fitness ( h = 0 ) was computed the standard way , that is , as 1-si when homozygous for the deleterious allele and as 1 otherwise . We modeled fitness of the sex chromosomes following the framework described by Charlesworth et al . [53] and Veeramah et al . [52] , with a slight modification to preserve the multiplicative fitness scenario described for the autosome . The specific fitness models for each dominance scenario—additive , recessive , and with the h ( s ) function—are presented in S2 Table . Importantly , the fitness of females that are homozygous and males that have the selected allele are the same , and , in the additive model , heterozygous females have an intermediate fitness . This models dosage compensation in females , assuming levels of gene expression map to the same fitness values for males and females . | Individuals from distinct populations sometimes will produce fertile offspring and will exchange genetic material in a process called hybridization . Genomes of hybrid individuals often show non-random patterns of hybrid ancestry across the genome , where some regions have a high frequency of ancestry from the second population and other regions have less . Typically , this pattern has been attributed to adaptive introgression , where beneficial genetic variants are passed from one population to the other , or to genomic incompatibilities between these distinct species . However , other mechanisms could lead to these heterogeneous patterns of ancestry in hybrids . Here we use simulations to investigate whether deleterious mutations affect the patterns of introgressed ancestry across genomes . We show that when ancestry from a larger population is added to a smaller population , the ancestry from the larger population dramatically increases in frequency because it carries fewer deleterious mutations . This occurs even in the absence of beneficial mutations in either population . Additionally , we show that differences in sex chromosome evolution relative to autosomes , or differences in mating system , can affect patterns of introgression in similar ways . Our study argues that deleterious mutations should be included in population genetic models used to identify unusual regions of the genome that appear to be under selection in hybrids . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"biotechnology",
"deletion",
"mutation",
"engineering",
"and",
"technology",
"population",
"genetics",
"brassica",
"simulation",
"and",
"modeling",
"mutation",
"plant",
"science",
"model",
"organisms",
"experimental",
"organism",
"systems",
"plant",
"genomics",
"population",
"biology",
"plants",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"bioengineering",
"animal",
"studies",
"plant",
"genetics",
"introgression",
"heterosis",
"eukaryota",
"plant",
"and",
"algal",
"models",
"natural",
"selection",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"gene",
"flow",
"genomics",
"evolutionary",
"biology",
"plant",
"biotechnology",
"evolutionary",
"processes",
"organisms"
] | 2018 | Deleterious variation shapes the genomic landscape of introgression |
The dynamic of cancer is intimately linked to a dysregulation of the cell cycle and signalling pathways . It has been argued that selectivity of treatments could exploit loss of checkpoint function in cancer cells , a concept termed “cyclotherapy” . Quantitative approaches that describe these dysregulations can provide guidance in the design of novel or existing cancer therapies . We describe and illustrate this strategy via a mathematical model of the cell cycle that includes descriptions of the G1-S checkpoint and the spindle assembly checkpoint ( SAC ) , the EGF signalling pathway and apoptosis . We incorporated sites of action of four drugs ( palbociclib , gemcitabine , paclitaxel and actinomycin D ) to illustrate potential applications of this approach . We show how drug effects on multiple cell populations can be simulated , facilitating simultaneous prediction of effects on normal and transformed cells . The consequences of aberrant signalling pathways or of altered expression of pro- or anti-apoptotic proteins can thus be compared . We suggest that this approach , particularly if used in conjunction with pharmacokinetic modelling , could be used to predict effects of specific oncogene expression patterns on drug response . The strategy could be used to search for synthetic lethality and optimise combination protocol designs .
Pharmacokinetic and pharmacodynamic ( PK/PD ) models of anticancer drug action have many potential applications [1–3] . Among the most promising are the ability to match tumours with particular gene expression profiles to selective treatments [4] , the ability to search for potential synthetic lethalities [5] , and the ability to optimise combination protocols [6] . Thousands of treatment protocols can be screened in silico , and the most promising selected for experimental or clinical evaluation [7] . Modelling the cellular pharmacodynamics of anticancer drugs , whether they are cytotoxic agents or targeted agents requires , minimally , a description of three biological processes: the cell cycle , the associated signal transduction pathways , and the apoptotic cascade . There are published models of all of these processes , and our model includes descriptions of the cell cycle , the EGF signalling pathway and apoptosis . In a previous study we showed that the loss of the G1-S and/or SAC checkpoints were critical to the description of cancer [8] . This was consistent with Duesberg’s theory [9] which suggested that cancer is , in essence , a disease of chromosomal instability . According to this line of thought the phenotypic hallmarks of cancer that arise are the inevitable outcome of the selection process operating on the numerous chromosomal variants . The evidence linking defective SAC function with cancer has been reviewed by Kops , Weaver and Cleveland [10 , 11] , and by Musacchio and Salmon [12] . There are many deletions or mutations that can cause SAC over-ride , resulting in aneuploidy . One of the commonest SAC abnormalities in human cancer appears to result from over-expression of aurora kinase A [13 , 14] . Other mitotic proteins whose over-expression or mutation results in aneuploidy include Nek2 [15 , 16] , Hec1 [17] , and Mad2 [18] . Our model includes mathematical descriptions of the G1-S and SAC checkpoints where aurora kinase A expression can be manipulated . There have been a number of attempts to enhance the selectivity of cancer chemotherapy by exploiting loss of checkpoint function in cancer cells , a concept that has been termed “cyclotherapy” [19–21] . Cyclotherapy is an example of new biomarker-driven therapeutic strategies that will require more sophisticated pharmacodynamic modelling to realise their full potential . Here we illustrate how a modelling approach that incorporates the cell cycle oscillator and descriptions of the G1-S and SAC checkpoints , together with EGF signalling and apoptosis pathways , can help in developing such strategies . To study the effects of drugs in various cytokinetic configurations , the sites of action of different anticancer drugs can be incorporated ( Table A in S1 Text ) . We consider here four different drugs: palbociclib , gemcitabine , actinomycin D and paclitaxel . Sets of parameter values can also be used to describe different cell types ( Table B in S1 Text ) . We investigated here with the malignant cell line MiaPaca-2 and normal cell line ARPE-19 . Following a brief description of the implementation of this approach , we aim to demonstrate via a series of examples how the strategy can facilitate identifying drug selectivity by simulating drug effects on normal and transformed cells .
This approach allows the study of cohorts of normal and cancer cells and comparing the effects of drugs . Fig 3 shows the simulated growth of cells in the absence of drug treatment . For normal cells , the cell cycle dynamic can be modulated by the presence of ligands . The model simulations results in normal cells having a shorter doubling time ( from 25 to 13 hours; ( Fig 3A ) while MiaPaca-2 cells , in which MAP kinase signalling downstream from ras was modelled as constitutively activated , were indifferent to EGF concentration ( Fig 3B ) . As mentioned earlier , it has been argued that the selectivity of cancer chemotherapy could be enhanced by exploiting loss of checkpoint function in cancer cells , a concept termed cyclotherapy [19–21] . Since arrest in G1 is non-cytotoxic to normal cells , one approach to cyclotherapy is to treat with a drug that will cause G1 arrest . Cancer cells with impaired G1 checkpoint function , e . g . because of a p53 mutation , will progress into S phase , where they may be selectively killed by an S-phase-specific drug . The cdk4/cdk6 inhibitor palbociclib ( PD332991 ) [22 , 23] prevents normal cells from progressing through the G1 checkpoint and entering S phase . In cells with mutant ras such as MiaPaca-2 the G1 checkpoint is weakened by high levels of production of cyclin D , making them less prone to arrest following treatment with palbociclib . Fig 4A shows that the simulated effect of exposure for 48 hours to 30 , 100 and 300 nM palbociclib was to reduce proliferation but was non-cytotoxic to both cell lines ( thus having minor utility as a single agent at these concentrations ) . Nevertheless , progression from G1 to S was affected with these concentrations and resulted in a decrease of the population of cells in S phase which was more pronounced in normal cells ( Fig 4B ) . Combining gemcitabine with palbociclib . The anticancer drug gemcitabine is selectively cytotoxic to cells in S phase [24] . Gemcitabine might possess inherent anticancer selectivity since cells lacking G1 checkpoint function should have a higher proportion of cells reaching S phase . Fig 4C shows that consistent with this understanding , the model predicted that gemcitabine was more cytotoxic to MIA PaCa-2 cells compared to the normal cell line ( Fig 4C ) . We then investigated combining gemcitabine with palbociclib since palbociclib was differentially active in allowing cancer cells to enter the S-phase compared to normal cells . The simulated dose-responses showed that the combination affected more malignant cells than normal cells as per gemcitabine alone ( Fig 4D ) . In particular , antagonistic effect was observed with normal cells when adding palbociclib . We then wanted to investigate if delaying gemcitabine administration would enhance differential effect between normal and cancer cell lines . Thus , we simulated cell growth following administration of palbociclib and gemcitabine where gemcitabine administration was delayed ( Fig 4E ) . Concentrations of 30 nM for both drugs were used as these induce similar effects in normal and malignant cells ( Fig 4D ) . The simulations showed that delaying gemcitabine addition by approximately 12 hours from the start of palbociclib resulted in the best differential effect ( Fig 4F ) , with a ratio of normal cell viability to malignant cells viability of 17 . This illustrates from a mechanistic point of view the paradigm of cyclotherapy , in which cells with an intact G1 checkpoint can be selectively protected from cytotoxic agents acting later in the cell cycle . A few experimental studies have demonstrated this effect , generally exploiting the fact that tumour cells with mutant p53 lack a functional G1 checkpoint [25 , 26] . More advanced optimization approaches can also be employed to attempt the optimisation of both concentration and administration schedules in order to reach one or several pre-defined goals . For instance normal cell viability can be set above a specific threshold or target malignant cell viability can be set below a specific threshold while exposure constraints can also be taken into account . In addition to defects in the G1 checkpoint , most tumours may have impaired SAC function . One cause of this is over-expression of AK-A [13 , 14] . High levels of AK-A tend to be associated with low-level resistance to taxanes [13] . It is not clear , intuitively , whether there is a mechanistic relationship between the AK-A over-expression and the taxane sensitivity . [13] Paclitaxel causes M phase cell cycle arrest , and cells that remain arrested for several hours enter apoptosis . Simulations were used to illustrate how during a 24 hour treatment with 10 nM paclitaxel , cells with a functional SAC accumulate in M phase ( Fig 5A ) . If the drug is removed after 24 hours , a large cohort of cells moved synchronously into G1 ( Fig 5A ) . Malignant cells , which are modelled with higher AK-A levels , also showed an initial increase in the M phase fraction ( 12 hours; Fig 5B ) . However , after about 24 hours , the simulation captured the phenomenon of “checkpoint leakiness” , i . e . an increasing number of malignant cells entered cell division even though mitosis was not fully completed , thus appearing in G1 phase ( 24 hours; Fig 5B ) . In contrast to normal cells , if the drug was removed after 24 hours , only an additional small cohort of cells move into G1 ( Fig 5A ) . Actinomycin D is a transcription inhibitor that kills cells in all phases of the cell cycle except M phase ( because transcription is not active during mitosis ) . Because simulations highlighted a preferential arrest of normal cells in M phase by paclitaxel , we therefore asked the question if combining this drug with actinomycin D could generate preferential kill of tumor versus normal cells . Actinomycin D used as a single agent has similar activity in our modelled malignant and normal cell lines ( Fig 6A ) . We then simulated the combination of actinomycin D with paclitaxel which also resulted in a similar combination dose-response , although with slightly more antagonistic effects for normal cells against malignant ones at the highest concentrations ( Fig 6B; antagonistic scores of -17% and -12% respectively ) . Experiments confirmed differential combination effect , with slight antagonism for normal cells but mild synergy for malignant cells ( Fig A in S1 Text ) . Earlier , we showed that normal cells tended to accumulate more in M phase in the presence of paclitaxel between 12 and 24 hours ( Fig 5 ) , thus suggesting that a combination treatment in which actinomycin D was added after the start of paclitaxel could enhance this differential effect . Further simulations ( Fig 6C ) showed that antagonistic interactions were enhanced for both cells lines ( compared to concomitant treatment , Fig 6B ) , but indeed with greater protection achieved in normal versus malignant ( SAC-deficient ) cells ( Fig 6B ) . Nevertheless , the resulting predicted efficacy of the paclitaxel + actinomycin D combination for malignant cells was not strong enough ( 40% of control ) and only marginally better than for normal cells ( 48% of control ) and therefore is not likely to be a therapeutic option . Overall , these results highlight the potential to enhance the therapeutic window by considering inherent dynamical differences between malignant and normal cells .
Several hundred oncogenes and tumour suppressor genes have been identified . What they have in common is that all are involved in control of the cell cycle or its associated signalling pathways ( including the apoptosis pathways ) . [8 , 25–27]Understanding the dynamics of tumour growth and the pharmacodynamics of anticancer drugs could be greatly assisted by quantitative descriptions of these processes . It has been argued that malignant transformation involves , minimally , two kinds of somatic mutation . Moreover , it is believed that cancer is a disease of genetic instability[9 , 10 , 12 , 28]and that all human tumours have some degree of aneuploidy . Aneuploidy confers increased spontaneous cell loss , so that tumour cells can only survive and proliferate if they have a compensating growth advantage over competing normal cells . These changes in cancer are usually the result of mutations or changes in expression levels leading to over-ride of cell cycle checkpoints [8 , 10–12] . Modelling the cell cycle can therefore enable us to capture differences in dynamics between normal and cancer cells resulting from various mutations and associated phenotypes . Although the essential features of the mammalian cell cycle have been the subject of detailed dynamic models [29–32] , the significance of cell-cycle checkpoints has not been well studied . The present model builds upon these available models but also includes detailed kinetic descriptions of two of the major cell cycle checkpoints that are mutated or over-ridden in cancer: the G1 checkpoint and the spindle assembly checkpoint . This approach can be used to explore the potential anti-tumour selectivity of drugs that act on essential components of these checkpoints and the signalling pathways leading to them . This approach might facilitate exploring a therapeutic strategy termed”cyclotherapy”[19 , 20 , 33] , which attempts to optimise drug selectivity against cells with defective checkpoint function . The cell cycle , with its associated signalling pathways and apoptotic pathways , constitutes a complex interactive system . Analysing the dynamics of such systems requires that we take into account multiple positive and negative feedbacks , cross-talks , and effects that span multiple spatial compartments and multiple levels of hierarchical organization . Any model that attempts to describe the kinetics of the cell cycle must represent a compromise between this almost intractable complexity and over-simplifications so sweeping that the essential dynamics of the system are lost . The simulations described in this report were chosen to illustrate capabilities and limitations of the suggested approach . We also provide the tool developed for these studies ( CYCLOPS , https://sourceforge . net/projects/cyclops-simulations/ ) . The CYCLOPS model does not incorporate all known biological processes . Only major components of the cell cycle and simplified descriptions of apoptosis , EGF signalling , G1-S and SAC checkpoints are incorporated . Additionally , the model is based on specific kinetics which are only reasonable average values . It should be noted that great cell to cell variability and difficulties in quantifying temporal and spatial profiles of proteins and other cellular components currently excludes deriving models which are truly quantitative . 54 A more complete description of cell cycle dynamics could also consider the G2-M checkpoint , particularly as there is cross-talk between it and the spindle assembly checkpoint . As illustrated here , models such as CYCLOPS can facilitate the understanding of underlying cellular dynamics and how to develop or optimize therapeutic strategies . Mechanistic approaches as this one ( increasingly termed quantitative systems pharmacology ( QSP ) [34] ) bridges molecular and systems biology studies to traditional PK/PD studies . They offer the potential for accelerating the drug discovery process and making it more cost-effective . An obvious practical application is the design of rational drug combinations for which exhaustive experimental study can be impractical . Potential combinations can be better understood and optimized if assisted by mathematical models . In this context , modelling the cell cycle and its modulating components can facilitate the development of combinations , particularly within the scope of cyclotherapy as illustrated here . Additional features that can be investigated could include multiple tumour cell populations and a description of mutations from drug sensitivity to resistance , and vice versa . Double mutants with resistance to two drugs can also be modelled . Indeed , another important rationale for combination chemotherapy is the use of combinations of non-cross-resistant drugs to prevent or delay treatment failure resulting from acquired drug resistance . The approach presented here can be extended to facilitate the design and optimization of such combinations . Although ambitious , it is possible to envision a future where pharmacodynamic models of the cell cycle can also be used to develop personalised chemotherapies . Because each tumour is genetically unique and expressed against a unique genetic background , individualising therapy is essentially a multi-dimensional optimization problem . Eventually , CYCLOPS type and other QSP approaches , together with more traditional PK modelling , might provide powerful tools for matching treatment regimens to each tumour’s particular expression profile .
We have developed and coded ( C language , code available on https://sourceforge . net/projects/cyclops-simulations/ ) a mathematical model to investigate cyclotherapy pharmacodynamics strategies ( CYCLOPS ) . CYCLOPS allows simulating a cohort of cells in different phases of the cell cycle . For each cell the following processes are modelled: the basic cell cycle , the G1-S checkpoint , the spindle assembly checkpoint , part of the MAP kinase signal transduction pathway and apoptosis . We first describe each one of these cellular processes individually and how they have been incorporated . Then we explain how this model has been used to simulate large cohorts of cells . A comprehensive review of cell cycle modelling has been published by Csikásh-Nagy . [29] Since our underlying premise is that malignant transformation requires , minimally , loss of function of two cell cycle checkpoints , it was necessary to model these checkpoints , and their effects on the pharmacodynamics of anticancer drugs . The CYCLOPS model differs from previous models in incorporating a kinetic description of the spindle assembly checkpoint ( SAC ) and the G1-S checkpoint ( Fig 1A ) . It is an update of a classical cytokinetic model[35 , 36] to which has been added a version of the cell cycle oscillator based on that described by Novak and Tyson[30 , 37] and elaborated by Gérard and Goldbeter [31] . Portions of the cell cycle model have been published in Chassagnole et al . [32] . The description of the mitotic spindle assembly checkpoint is based on that described by Mistry et al . [38] and by Kamei et al . [39] . The normal G1-S checkpoint , as modelled by CYCLOPS , is summarised in Fig 1B . Transcription of most of the enzymes required for DNA replication is driven by the transcription factor c-myc , which in turn is under the control of the transcription factor E2F . E2F in G1 cells is bound to , and inactivated by the RB protein . When the RB protein is phosphorylated , active E2F is released . Phosphorylation of RB is catalysed by the cyclin-dependent kinases , cdk2 and cdk4 . Cdk4 is activated by cyclin D , which is produced by a number of signalling pathways , including the MAP kinase pathway . Cdk2 is activated by cyclin E [25] . For progression of cells from G1 phase into S phase it is essential that their DNA is intact and that they have sufficient DNA precursors for DNA synthesis to commence . In the presence of DNA damage , or if the nucleotide pools are depleted or unbalanced , the p53 protein is activated [26] . This results in transcription of p21 and p27 , which inhibit cdk2 , and thus prevents release of the G1-S checkpoint . When the DNA damage has been repaired , p27 transcription ceases , and the cell enters S phase . If the DNA damage cannot be repaired within a certain time ( usually 8 to 24 hours , depending upon cell type ) the cell enters apoptosis . About 50% of carcinomas have mutations or deletions of p53 while some tumours lack a functional retinoblastoma ( RB ) protein , resulting in a dysfunctional G1-S checkpoint . In CYCLOPS , this is modelled via over-ride of the G1-S checkpoint . Other tumours have mutations that also result in over-ride of the G1-S checkpoint: this can be caused by over-expression of cyclin D or cyclin E , or by mutations that result in constitutive activation of the EGF receptor , or of ras . Thus the CYCLOPS model is able to model mechanisms of G1-S checkpoint defect or over-ride . Nevertheless , this model represents a first approximation , and at present does not describe the effects of a number of physiological regulators , such as for instance CDKN2A . The checkpoint model could be elaborated as additional kinetic data becomes available . The SAC acts by sensing correct connections of kinetochores to the two opposite spindle poles . All kinetochores are initially modelled as emitting a “wait” signal . Once all kinetochores reach a state of tension this signal is stopped , thus allowing progression to anaphase ( Fig 1C ) . The wait signal is mediated by the bistable , tension-sensitive Aurora Kinase B ( AK-B ) [40] . Microtubules grow from the opposite spindle poles , and attach at random to the kinetochores . This results in various configurations which are modelled here: Syntelic attachments are not in a state of tension , which results in the second attachment being removed through activity of the enzyme aurora kinase B . Amphitelic attachments are in a state of tension which results in aurora kinase B being inactive . When all pairs of sister chromatids are correctly attached , the wait signal rapidly decays , and the cell progresses to anaphase . Failure of the SAC results in premature exit from mitosis and aneuploidy . Most tumours are aneuploid , but aneuploidy is never detected in normal replicating cells . [41] We modelled in CYCLOPS the effects of three drug classes on the SAC . Aurora kinase A ( AK-A ) inhibition , which slows the process of mitosis by increasing time to anaphase ( AK-A is essential for centrosome maturation ) . Paclitaxel , which stabilises microtubules against depolymerisation and also increases time to anaphase . Aurora kinase B ( AK-B ) inhibition , which slows down the removal of incorrect microtubule-kinetochore attachments . The current model of the MAP kinase pathway used by CYCLOPS is based on the model of Brightman and Fell [42] . Several other groups have modelled this pathway ( reviewed in Gilbert et al . [43] ) . The CYCLOPS model captures much of the essential dynamics of EGF signalling ( Fig 1D ) and includes sites of action of five classes of drugs . When EGF binds to its cell surface receptor , RAS is activated , and signals through RAF , MEK and ERK to up-regulate cyclin D and over-ride the G1-S checkpoint ( Fig 1D ) . Caspases are produced as inactive procaspases . One procaspase molecule , when activated ( by a cellular damage signal ) can then catalytically activate many other procaspase molecules . The process is thus autocatalytic . Like kinases , proteases can act as multi-stage amplifiers . In apoptosis , procaspase 9 is activated to caspase 9 , which catalyzes the conversion of procaspase 3 to caspase 3 , which is the proximal cause of cell death ( Fig 1E ) . Apoptosis has been modelled mathematically[44–46] and the CYCLOPS model is adapted from these published models . To model cancer cytokinetics requires that we can model asynchronous cell populations , which may contain millions of cells . To model the cell cycle oscillator individually in each cell would be impractical . Instead , cells are grouped into a succession of cohorts , assumed to be a few minutes apart . CYCLOPS treats the cell as a sequence of 63 states , with transition rules based upon a combination of elapsed time and biochemical values ( Fig 2 ) . Some of these quantities are modelled continually ( DNA , total protein ) , and others are calculated . In these cohorts , the apparent cell cycle time is modulated by biochemical parameter values . The 63 cytokinetic states are: 15 G1 states ( differing in total protein content and cyclin E level ) , 30 S phase states ( differing in DNA content ) , 10 G2 states ( differing in time elapsed from the start of G2 ) , 5 M states ( prophase , prometaphase , metaphase , anaphase , telophase ) , a single G0 phase , a single population of terminally differentiated and senescent cells , and a population of irreversibly damaged cells that are metabolically active but unable to replicate . These 63 compartments can contain any number of cells ( Fig 2 ) . In addition to progressing through the stages of the cell cycle , cells may leave the cycle irreversibly through cell death , differentiation or senescence . Spontaneous cell loss after cell division is treated as a cytokinetic parameter characteristic of particular cell lines , as are rates of differentiation/senescence ( Table 1 ) . Senescence , differentiation , and apoptosis may also be stimulated by drug treatment . Cells may leave the cell cycle reversibly and enter a quiescent ( G0 ) compartment ( Fig 2 ) . In the current study , a modelled MiaPaca-2 cancer cell line was used . A goal of the model is to optimise drug selectivity and the selection of an appropriate normal control cell is essential . Our approach is two-fold: ( a ) for many anticancer drugs it is possible to identify a particular drug-sensitive normal cell type that represents the site of dose-limiting toxicity . For most anticancer drugs this is bone marrow , intestinal mucosa , skin , or the immune system . There is sufficient cytokinetic information for these tissues to be modelled in detail , and we can describe the effects of many drugs on these tissues explicitly . We then assume that for the purposes of predicting efficacy and selectivity , drug effects on other cell types can be ignored . ( b ) Our underlying premise is that cancer is primarily a defect of cell cycle checkpoints . For modelling purposes we can then predict anticancer drug selectivity by assuming that normal cells differ from the cancer cell in having fully functional cell cycle checkpoints . Specifically , the MiaPaca-2 cells are modelled as having mutant , constitutively active RAS , resulting in up-regulation of cyclin D , and causing override of the G1 checkpoint . [47] . Second , a 3-fold over-production , relative to the normal control of aurora kinase A , causing impaired function of the SAC was also incorporated ( Table 1 ) [13 , 48 , 49] . Pharmacodynamic modelling using CYCLOPS . Four sites of drug action modelled in CYCLOPS were investigated here: Drug effects were modelled using a standard Hill equation whose parameter values were obtained from the DrugCARD database [53] . The rate of cytotoxicity was defined as a reduction of the cell count following treatment compared to the count at the start of treatment: Cytotox=100 ( 1−#cellsatt=24h#cellsatt=0h ) CYCLOPS was coded in C and is available online ( https://sourceforge . net/projects/cyclops-simulations/ ) . CYCLOPS generates graphical output using the open source program , gnuplot . [54] A flow chart of the model implementation is included in the supplementary material ( Fig B in S1 Text ) . The list of components is given in Supplementary Table B in S1 Text . In the present study the code was compiled using the free gcc compiler , release 4 . 6 . 2 . | Neoplastic transformation results from mutations , chromosomal abnormalities , or expression changes affecting components of the cell cycle , the signalling pathways leading into it , and the apoptosis pathways resulting from cell cycle arrest . Cytotoxic agents , but also newer drugs that target the cell cycle and its signalling pathways , perturb this complex system . Small differences in cell cycle control between normal and transformed cells could determine drug selectivity . Using cell cycle and representative signalling and apoptotic pathway simulations , we examine the influence of cell cycle checkpoints ( frequently defective in cancer ) on drug selectivity . We show that this approach can be used to derive insights in terms of drug combinations scheduling and selectivity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"cell",
"death",
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"and",
"cell",
"division",
"cancer",
"treatment",
"cell",
"processes",
"departures",
"from",
"diploidy",
"oncology",
"pharmacodynamics",
"pharmacology",
"synthesis",
"phase",
"aneuploidy",
"signal",
"transduction",
"apoptotic",
"signaling",
"cascade",
"cell",
"biology",
"apoptosis",
"genetics",
"biology",
"and",
"life",
"sciences",
"cell",
"signaling",
"signaling",
"cascades",
"cyclins"
] | 2017 | Modelling of the cancer cell cycle as a tool for rational drug development: A systems pharmacology approach to cyclotherapy |
Cellulosic plant biomass is a promising sustainable resource for generating alternative biofuels and biochemicals with microbial factories . But a remaining bottleneck is engineering microbes that are tolerant of toxins generated during biomass processing , because mechanisms of toxin defense are only beginning to emerge . Here , we exploited natural diversity in 165 Saccharomyces cerevisiae strains isolated from diverse geographical and ecological niches , to identify mechanisms of hydrolysate-toxin tolerance . We performed genome-wide association ( GWA ) analysis to identify genetic variants underlying toxin tolerance , and gene knockouts and allele-swap experiments to validate the involvement of implicated genes . In the process of this work , we uncovered a surprising difference in genetic architecture depending on strain background: in all but one case , knockout of implicated genes had a significant effect on toxin tolerance in one strain , but no significant effect in another strain . In fact , whether or not the gene was involved in tolerance in each strain background had a bigger contribution to strain-specific variation than allelic differences . Our results suggest a major difference in the underlying network of causal genes in different strains , suggesting that mechanisms of hydrolysate tolerance are very dependent on the genetic background . These results could have significant implications for interpreting GWA results and raise important considerations for engineering strategies for industrial strain improvement .
The increased interest in renewable energy has focused attention on non-food plant biomass for the production of biofuels and biochemicals [1] . Lignocellulosic plant material contains significant amounts of sugars that can be extracted through a variety of chemical pretreatments and used for microbial production of alcohols and other important molecules [2–5] . However , there are major challenges to making biofuel production from plant biomass economically viable [6] . One significant hurdle with regards to microbial fermentation is the presence of toxic compounds in the processed plant material , or hydrolysate , including weak acids , furans and phenolics released or generated by the pretreatment process [7–10] . The concentrations and composition of these inhibitors vary for different pretreatment methods and depend on the plant feedstocks [7 , 9 , 11] . These toxins decrease cell productivity by generating reactive oxygen species , damaging DNA , proteins , cell membranes [12–14] , and inhibiting important physiological processes , including enzymes required for fermentation [15] , de novo nucleotide biosynthesis [16] , and translation [17] . Despite knowledge of these targets , much remains to be learned about how the complete suite of hydrolysate toxins ( HTs ) acts synergistically to inhibit cells . Furthermore , how the effects of HTs are compounded by other industrial stresses such as high osmolarity , thermal stress , and end-product toxicity remains murky . Engineering strains with improved tolerance to industrial stresses including those in the plant hydrolysate is of the utmost importance for making biofuels competitive with fuels already in the market [6] . A goal in industrial strain engineering is to improve lignocellulosic stress tolerance , often through directed engineering . Many approaches have been utilized to identify genes and processes correlated with increased stress tolerance , including transcriptomic profiling of cells responding to industrial stresses [18–21] , genetic mapping in pairs of strains with divergent phenotypes [22–25] , and directed evolution to compare strains selected for stress tolerance with starting strains [26–29] . However , in many cases the genes identified from such studies do not have the intended effect when engineered into different genetic backgrounds [30–33] . One reason is that there are likely to be substantial epistatic interactions between the genes identified in one strain and the genetic background from which it was identified [34] . A better understanding of how tolerance mechanisms vary across genetic backgrounds is an important consideration in industrial engineering . Exploring variation in HT tolerance across strain background could also reveal additional defense mechanisms . The majority of functional studies in Saccharomyces cerevisiae are carried out in a small number of laboratory strains that do not represent the rich diversity found in this species [35 , 36] . The exploration of natural diversity in S . cerevisiae has revealed a wide range of genotypic and phenotypic variability within the species [36–40] . In some cases , trait variation is correlated with genetic lineage [36 , 41–43] , indicating a strong influence of population history . At least 6 defined lineages have been identified in the species , including strains from Malaysia , West Africa , North America , Europe/vineyards , and Asia [41] as well as recently identified populations from China [38 , 44] . In addition to genetic variation , phenotypic variation has cataloged natural differences across strains , in transcript abundance [37 , 45 , 46] , protein abundance [47–49] , metabolism [50–52] , and growth in various environments [32 , 36 , 37 , 42 , 52–54] . Thus , S . cerevisiae as a species presents a rich resource for dissecting how genetic variation contributes to phenotypic differences . In several cases this perspective has benefited industry in producing novel strains by combining genetic backgrounds or mapping the genetic basis for trait differences [25 , 55–59] . We used genome-wide association ( GWA ) in S . cerevisiae strains responding to synthetic hydrolysate ( SynH ) , both to identify new genes and processes important for HT tolerance and to explore the extent to which genetic background influences mechanism . We tested 20 genes associated with HT tolerance and swapped alleles across strains to validate several allele-specific effects . However , in the process of allele exchange we discovered striking differences in gene contributions to the phenotype: out of 14 gene knockouts tested in two strains with opposing phenotypes , 8 ( 57% ) had a statistically significant effect on HT tolerance in one of the backgrounds but little to no significant effect in the other background . In most of these cases , the specific allele had little observable contribution to the phenotype . Thus , although GWA successfully implicated new genes and processes involved in HT tolerance , the causal variation in the tested strains is not at the level of the allele but rather whether or not the gene’s function is important for the phenotype in that background . This raises important implications for considering natural variation in functional networks to explain phenotypic variation .
We obtained 165 Saccharomyces cerevisiae strains , representing a range of geographical and ecological niches , that have high quality whole genome sequencing reads ( coverage ~30X ) , coming from published sequencing projects across the yeast community [39 , 42 , 52 , 60] ( S1 Table ) . We identified 486 , 302 high quality SNPs ( see Methods ) . 68% of them had a minor allele frequency less than 5% . Nucleotide variation compared to the well-studied S288c-derived reference strain varied from as low as 0 . 08% for the closely related W303 lab strain and as high as 0 . 72% for the bakery strain YS4 ( S1 Table ) . The majority of strains were largely homozygous ( in some cases due to strain manipulation by sequencing projects ) ; however , we identified 21 strains with >20% heterozygous sites . Most of these were from natural environments ( 11 strains ) but they also included clinical samples ( 5 strains ) , baking strains ( 3 strains ) , a sugar cane fermenter ( 1 ) and a laboratory strain ( FL100 , which was scored as 98% heterozygous and may have mated with another strain in its recent history ( S1 Table ) ) . Sixty-three percent of the variants were present in coding regions ( S2 Table ) , which is lower than random expectation ( since 75% of the yeast genome is coding ) and consistent with purifying selection acting on most gene sequences . Indeed , coding variants predicted to have high impact , such as SNPs that introduce a stop codon , eliminate the start codon , or introduce a defect in the splicing region , were very rare ( 0 . 004% of genic SNPs ) –a third of these were in dubious ORFs ( 22% ) or genes of unknown function ( 8% ) [61] that are likely nonfunctional and under relaxed constraint . However , 54 genes with debilitating polymorphisms are reportedly essential in the S288 background; nearly half of these polymorphisms are present in at least 3 strains and in some cases are lineage specific ( S3 Table ) . Tolerance of these polymorphisms could arise through duplication of a functional gene copy [62] , but could also arise due to evolved epistatic effects as has been previously reported [63] , highlighting the complexity behind genetic networks and the role of genetic variation in determining their regulation . Principal component analysis of the genomic data recapitulated the known lineages represented in the collection , including the European/wine , Asian/sake , North American ( NA ) , Malaysian , West African ( WA ) , and mosaic groups [36 , 41 , 42 , 64] ( S1 Table ) . Our analysis split the West African population into three subgroups not previously defined ( Fig 1A ) . Construction of a simple neighbor-joining ( NJ ) tree broadly confirmed the population groups present in the 165-strain collection ( Fig 1B ) . We scored variation in lignocellulosic hydrolysate tolerance in several ways . Strains that are sensitive to hydrolysate grow slower and consume less sugars over time [65] , thus we measured final cell density and percent of glucose consumed after 24 hours to represent SynH tolerance . Growth and glucose consumption were significantly correlated ( R2 = 0 . 79 ) , although there was some disagreement for particular strains ( including flocculant strains ) ( S1 Table ) . We also determined tolerance to HTs specifically , to distinguish stress inflicted by HTs from effects of the base medium that has unusual nutrient composition and high osmolarity due to sugar concentration . To do this , we calculated the relative percent-glucose consumed and final OD600 in media with ( SynH ) and without HT toxins ( SynH–HTs , see Methods ) ( S1 Table ) . Tolerance to SynH base medium without toxins ( SynH–HT ) and SynH with the toxins was only partly correlated ( R2 = 0 . 48 ) ( S1 Fig ) , suggesting that there are separable mechanisms of growing in base medium and surviving the toxins . There is wide variation in tolerance to lignocellulosic hydrolysate that partly correlates with populations ( Fig 2 , S1 Fig ) . North American and Malaysian strains displayed the highest tolerance to SynH . As expected , phenotypic variation within each population was related to genetic variation , e . g . West African strains in Population 5 showed low genetic and phenotypic variation while mosaic strains with genetic admixture showed the widest range of phenotypes . We used GWA to map the genetic basis for the differences in SynH tolerance , for each of the four phenotypes introduced above . The population signatures in S . cerevisiae are problematic for GWA , since the strong correlations between phenotype and ancestry obscure the identification of causal polymorphisms [66 , 67] . To overcome this , we incorporated a large number of mosaic strains in the analysis and used a mixed-linear model to account for strain relationships , as implemented in the program GAPIT [68] ( see Methods ) . We used as input SNPs that were present in at least 3 strains , eliminating 42% of SNPs in the dataset ( see Methods ) . Of the remaining SNPs , 45% have a minor allele frequency of less than 5%; only those with an allele frequency >2% were used for GWA . GWA identified loci whose variation correlated with phenotypic variation . None of the GWA-implicated loci passed the stringent Bonferroni p-value correction based on the number of effective tests ( see Methods ) , which is not uncommon for GWA at this scale [42 , 69 , 70] . We therefore used a somewhat arbitrary p-value cutoff of 1e-04 and performed additional filtering to minimize false positive associations ( see Methods ) . The combined analysis yielded 76 SNPs that met our p-value threshold ( S4 Table , S2 Fig ) . Thirty-eight of these SNPs , linked to 33 genes , passed additional filtering ( See Methods , Table 1 ) . Of these , 17 SNPs are associated with growth in SynH , while 23 SNPs are associated with tolerance to HTs specifically ( Table 1 ) . Eight of the SNPs are intergenic and 20 are located within genes , with 13 of those predicted to change the coding sequence . Although we would expect that SNPs linked to HT tolerance should be identified in both sets of analyses , only 2 SNPs were significantly associated with both SynH and HT tolerance . This almost certainly highlights limited statistical power with the small set of strains used here . For most SNPs , the allele associated with tolerance was more frequent in our strain collection ( Fig 3A ) , but for some it was the allele associated with sensitivity that was nearly fixed . We carried out additional GWA filtering to ensure that results were not driven by population structure ( see Methods ) , since we note that many of sensitive alleles were prominent in the Asian population ( S3 Fig ) . As expected for a largely additive trait , there was a significant linear correlation between the number of deleterious alleles a strain harbored and its tolerance to hydrolysate ( R2 = 0 . 48 , p = 2 . 2e-16 , Fig 3B ) . Interestingly , the genes associated with the 38 implicated SNPs capture functionally related processes , suggesting mechanistic underpinnings of hydrolysate tolerance . Lignocellulosic hydrolysate contains a large number of toxins that affect multiple cellular functions and can target energy stores , membrane fluidity , protein and DNA integrity , and other processes [10 , 65] . Our analysis implicated several genes involved in redox reactions ( ADH4 , ALD3 ) , protein folding or modification ( CYM1 , UBP5 , UFD2 , AOS1 ) , ergosterol or fatty acid synthesis ( ERG12 , HMG2 , NSG2 ) , DNA metabolism and repair ( REV1 , DAT1 , MCM5 , SHE1 ) , mRNA transcription and export ( LEU3 , SIR3 , ELF1 , RIM20 , MEX67 ) , mitochondrial function ( MNE1 , MAS1 ) , and flocculation ( FLO1 , FLO10 ) . Several of these processes were already known to be associated with hydrolysate stress , including flocculation [71] , ubiquitin-dependent processes that may be linked to protein folding challenges [13 , 72 , 73] , and sterol biosynthesis which affects tolerance to multiple stresses present in this media [32 , 74 , 75] . Nearly a third of these genes were identified as differentially expressed in our previous study comparing strain responses to SynH and rich medium [32] , although this was not enriched above what is expected by chance . Thus , although gene expression differences can be informative in suggesting affected cellular processes , many of the genes implicated by GWA cannot be predicted by expression differences , especially SNPs that affect function without altering gene expression . Additional genes identified here belong to functional groups previously identified in our differential expression analysis , such as amino-acid and NAD biosynthesis . We sought to confirm the importance of the GWA-implicated genes in SynH tolerance , first through gene-knockout analysis and then with allelic replacement in two different strains backgrounds . We began by knocking out 19 of the implicated genes in the tolerant North American strain , YPS128 . Of these , 37% ( 7/19 ) of the knockout mutants had a significant phenotype when grown on SynH: four displayed decreased SynH tolerance , while 3 showed increased performance ( Fig 4A ) . We note that 4 of the 7 knockouts had a mild phenotypic effect in standard growth medium ( that was generally exacerbated in SynH ) , while 3 of these had a phenotype only in response to SynH ( S4 Fig ) . The most significant knockouts decreasing tolerance in the YPS128 strain included the transcription factor LEU3 , ribosomal protein RPL21B , protein phosphatase subunit SAP190 , and to a milder extend the mitotic spindle protein SHE1 . None of these genes has been directly implicated in tolerance to hydrolysate in previous studies . The effect of deleting LEU3 , which encodes the leucine-responsive transcription factor , was intriguing , since our prior work reported that amino-acid biosynthesis genes are induced specifically in response to HTs [32] . To confirm that this response was due to the toxicity found in the media and not due to amino acid shortage in SynH , we compared growth in synthetic complete ( SC ) medium , which has similar levels of branched-chain amino acids compared to SynH . The LEU3 knockout strain grew as well as the wild type in SC , but it grew to 54% lower final density in SynH–HT medium and 79% lower density in SynH medium with the toxins added ( Fig 5A ) . The defect was not fully complemented by supplementing synthetic hydrolysate with 10X the normal amino acid mix ( Fig 5B ) , indicating that amino acid shortage in the medium is unlikely to fully explain the growth defect . The most striking phenotypic improvement was caused by deletion MNE1 , encoding a splicing factor for the cytochrome c oxidase-encoding COX1 mRNA [76] . Aerobically , the mutant grew to roughly similar cell densities but consumed 44 . 7% more glucose and generated 64% more ethanol than the wild type , generating significantly more ethanol per cell ( S5 Fig ) . A logical hypothesis is that this mutant has a defect in respiration and thus relies more on glycolysis to generate ATP and ethanol than wild-type cells [76] . Under this hypothesis , the effect of the mutation should be normalized when cells are grown anaerobically because both the mutant and wild type must rely on fermentation . However , under anaerobic conditions the mutant grew significantly better than the wild type ( Fig 6A ) , consumed 70% more glucose ( Fig 6B ) , and produced 63% more ethanol after 24-hour growth ( Fig 6C ) . Thus , a simple defect in respiration is unlikely to explain the result , suggesting that Mne1 may have a separable role relating anaerobic toxin tolerance and/or metabolism . We next knocked out 16 genes in the sensitive strain YJM1444 , with the intention of allelic exchange ( Fig 4B ) . We were unable to recover knockouts for some of the genes tested in YPS128 , but of those we acquired 14 overlapped the YPS128 knockouts , and two ( REV2 and HMG2 ) that we were unable to knock out in the tolerant strain were added . Remarkably , knockouts had strikingly different effects between the two genetic backgrounds–while three of the gene deletions affected hydrolysate tolerance in YJM1444 , there was no overlap with the gene deletions causing a statistically significant effect in YPS128 ( although some mild effects may be below our statistical power to detect ) . The three knockouts specific to YJM1444 improved SynH tolerance and included two genes involved in sterol biosynthesis ( NSG2 and HMG2 ) and one involved in flocculation ( Fig 4B ) . In fact , deletion of FLO1 dramatically reduced the flocculation phenotype of YJM1444 and resulted in >236% increased glucose consumption in SynH . This single mutation converted YJM1444 tolerance to the level of SynH tolerance seen in YPS128 ( S6 Fig ) . To test that this phenotypic effect was directly caused by the FLO1 allele , we deleted its paralog FLO5 , which caused neither a change in flocculation nor increased glucose consumption of the culture ( S7 Fig ) . There appeared to be subtle , but not significant , effects of the MNE1 deletion in YJM1444 and we wondered if the was obscured by flocculation . Therefore , we measured glucose consumption in high-rpm shake flasks that disrupt flocculation . Indeed , MNE1 deletion had a significant benefit under these conditions; however , the magnitude of the effect was more subtle than MNE1 deletion in YPS128 ( S8A Fig ) . We also tested this deletion in an industrial strain , Ethanol Red ( E . Red ) . Deletion of MNE1 in a haploid spore derived from E . Red produced a minor , reproducible benefit although it was not statistically significant ( S8A Fig ) . Nonetheless , these results show that MNE1 plays a role in SynH tolerant , albeit to different levels , in three different strain backgrounds . We tested allelic effects in two ways . First , we introduced a plasmid-borne copy of the tolerant allele or sensitive allele ( S5 Table ) into YPS128 lacking the native gene , and measured percent final glucose consumption in SynH ( S9 Fig ) in synthetic complete medium ( required to allow drug-based plasmid selection ) with HTs ( Fig 7A ) . The assay was fairly noisy , nonetheless , there was a clear effect for the FLO1 allele , which caused YPS128 to become flocculant and dramatically decreased growth in the SC with HTs . We did not observe other allele-dependent effects that overcame the variability of the assay , including for the genes whose knockout produced a defect in YPS128 . Second , we performed reciprocal hemizygosity analysis for six genes , including three genes that whose deletion produced differential effects in YPS128 and YJM1444 . We crossed the YPS128 and YJM1444 backgrounds such that the resulting diploid was hemizigous for either the tolerant or sensitive allele ( Fig 7B ) . In this case , none of the six genes had an allele-specific effect–surprisingly , this included FLO1 for which there was clear allelic impact in the haploid backgrounds . We realized a unique phenotype in the YPS128-YJM1444 hybrid: whereas the strain is heterozygous for the functional FLO1 allele , the hybrid lost much of the flocculence of the YJM1444 strain ( S8B Fig ) . FLO1 expression is known to be repressed in some diploid strains [77] . Thus , simply mating the strains in effect created a new genetic background that changed the allelic impact of the gene . We wondered if this effect explained the lack of allele-specific phenotypes for other implicated genes . We therefore created homozygous deletions in the diploid hybrid for six genes whose deletion had strain-specific impacts in the haploids ( Fig 7C ) . Two of the knockouts ( leu3Δ and sap190Δ ) produced a defect in the hybrid , similar to the effect seen in YPS128 . Homozygous deletion of MNE1 produced a unique growth defect in 24-well plates that was not seen in the haploids or the hemizigous diploids . This appeared to be due to increased flocculation in the hybrid diploid; growth in shake flasks to disrupt flocculation resulted in a mild but statistically insignificant benefit to the hybrid when grown in flasks , similar to that seen for YJM1444 . In contrast , deletion of RIM20 or FLO1 had no effect under these growth conditions–this explains the lack of allele specific effect , because the genes are no longer important in this background and under these growth conditions . Mating YJM128 and YJM1444 created a new background that surpassed performance of YPS128 ( Fig 7D ) . We wondered if hybridization could benefit other strains as well . We mated industrial strain E . Red crossed to YJM1444 and YPS128 . E . Red and YJM1444 were both scored as sensitive and perform similarly in SynH ( Fig 7D ) . However , the hybrid had a striking jump in SynH tolerance , exceeding the tolerance of YPS128 . This benefit may be in part because the new diploid background changes the flocculation phenotype . On the other hand , YJM1444 and E . Red harbor alternate alleles at 71% of the SNPs implicated by GWA , raising the possibility that complementation of recessive alleles could also contribute to the strain improvement ( see Discussion ) .
Engineering strains for tolerance to lignocellulosic hydrolysate has proven difficult due to the complex stress responses required to deal with the combinatory effects of toxins , high osmolarity , and end products such as alcohols and other chemicals . Even when the cellular targets of stressors are known , the mechanisms for increasing tolerance are not always clear . We leveraged phenotypic and genetic variation to implicate new mechanisms of hydrolysate tolerance , by finding correlations between phenotypic and genetic differences among a collection of Saccharomyces cerevisiae strains , which allowed us to implicate specific genes and alleles involved in hydrolysate tolerance . The results indicate several important points relevant to engineering improved hydrolysate tolerance and genetic architecture of tolerance more broadly . Perhaps the most striking result is the level to which gene involvement varies across the strains in our study . We expected that swapping alleles of implicated SNPs should contribute to variation in the phenotype . Most alleles did not detectably affect tolerance , although it is likely that they may have a minor contribution below our limit of detection . Indeed , strains that harbor more deleterious alleles are significantly more sensitive to SynH ( Fig 3B ) , as expected for an additive trait . But at the same time , we uncovered significant variation in whether the underlying gene was involved in the phenotype . Among the genes that we were able to knockout in both strains ( 14 genes ) , 57% produced a phenotype ( to varying levels and significance ) in one of the two strains we tested . This indicates substantial epistatic interactions with the genetic background , such that the gene is important in one strain and but dispensable in another . Even more striking is the case of FLO1: knocking out the functional gene in YJM1444 produced a major benefit to that strain , whereas introducing the functional allele to YPS128 was very detrimental to SynH tolerance . Yet neither the allele nor the gene itself influenced SynH tolerance in the hybrid , because the hybrid is much less flocculant under these conditions ( despite carrying functional YJM1444 FLO1 gene ) . While it may not be surprising that gene knockouts result in quantitatively different phenotypes , we did not expect that most knockouts would have no detectible effect in specific backgrounds . It will be important to investigate the extent to which this effect is true in other organisms and for other phenotypes . However , evidence in the literature hints at the breadth of this result: several genes are required for viability in one yeast strain but not another [63 , 78] , while overexpression of other genes produces a phenotype in one background but not others [32] . Genetic background effects on gene contributions have been reported before , in yeast and other organisms [35 , 79–84]; however , the extent to which different genes appear to be involved in toxin tolerance in the different strains studied here suggests an important consideration that is underappreciated in GWA analysis: that the network of genes contributing to the phenotype could be largely different depending on genomic context . Dissecting these epistatic interactions is likely to be daunting , since a major challenge in most GWA studies remains identifying the epistatic interactions due to the high statistical hurdle [34 , 85 , 86] . We propose that emerging network-based approaches to augment linear contributions will be an important area in identifying genetic contributions in the context of background-specific effects . QTL mapping has allowed the characterization of the genetic architecture of industrially relevant stresses , including tolerance to ethanol [22 , 87] , acetic acid [23 , 56] , and plant hydrolysate [25] among many others [24 , 88–90] . But while this method exploits the genetic diversity between two strains , with GWA we were able to study a much larger collection of genetic diversity , providing unique insights . SynH tolerance is clearly a complex trait , with many genes likely contributing . Previous studies have shown that part of the growth inhibition can be explained by a re-routing of resources to convert toxins into less inhibitory compounds [18 , 19 , 91–94] and to repair damage generated by reactive oxygen species in membranes and proteins [13 , 14 , 95] . One of the most significant effects was caused by deletion of LEU3 , the transcription factor regulating genes involved in branched amino acid biosynthesis . Interestingly , weak acids have been shown to inhibit uptake of aromatic amino acids causing growth arrest [96] , and it is possible that Leu3 is required to combat this effect . Chemical genomic experiments suggest an additional role for Leu3 in managing oxidative stress in the cell [97] , which could relate to oxidative stressors in hydrolysate [13 , 14 , 32] . We also uncovered a gene , Mne1 , that when deleted significantly increases ethanol production in SynH . Mne1 aids the splicing of COX1 mRNA [76] and has not been previously linked to stress tolerance . Interestingly , MNE1 mutants produced more ethanol per cell aerobically , but also grew substantially better in SynH anaerobically , raising the possibility that Mne1 plays an additional , unknown role in cellular physiology that can be utilized to increase fermentation yields . Finally , although flocculation has been previously shown to increase cell survival in hydrolysate [71] , our study showed that flocculation reduced the rate of sugar consumption in the culture , likely because cells in the middle of the clump are nutrient restricted . Together , these results shed new light on SynH tolerance and mechanisms for future engineering . Our results raise broader implications for strain engineering , based on the genetic architectures uncovered here . Given the implication of gene-by-background interactions , the best route for improving strain performance may be crossing strains for hybrid vigor [98–100] . Indeed , we unexpectedly generated a strain that outperformed the tolerant YPS128 , by crossing two poor performers in SynH . This improved vigor could emerge if the hybrid complements recessive deleterious alleles in each strain , or if mating creates a new genetic background that changes the requirements ( and fitness ) of the strain . We believe that both models–weak but additive allelic contributions in the context of epistatic background effects–are at work in our study . For additive traits , GWA and genomic studies can have significant practical power , by predicting where individual strains fall on the genotype-phenotype spectrum and by suggesting which strains should be crossed for maximal phenotypic effect .
Strains used in the GWA are listed in S1 Table Gene knockouts were performed in strains derived from North American strain YPS128 and mosaic strain YJM1444 . The homozygous diploid parental strains were first engineered into stable haploids by knocking out the homothallic switching endonuclease ( HO ) locus with the KAN-MX antibiotic marker [101] , followed by sporulation in 1% potassium acetate plates and dissection of tetrads to attain heterothallic MATa and MATα derivatives . Gene knockouts were generated through homologous recombination with the HERP1 . 1 drug resistance cassette [102] and verified by 3 or 4 diagnostic PCRs ( validating that the cassette was integrated into the correct locus and that no PCR product was generated from within the gene that was deleted ) . Most knockouts removed the gene from ATG to stop codon , but in some cases ( e . g . kdx1 ) additional flanking sequence was removed , without removing neighboring genes . Genes from YPS128 or strains carrying the sensitive allele ( S5 Table ) were cloned by homologous recombination onto a CEN plasmid , taking approximately 1 , 000 bps upstream and 600 bps downstream from each genome , and verified by diagnostic PCR . Phenotyping of strains harboring alternate alleles on plasmids was performed in as previously described , except that the pre-culture was grown in YPD with 100 mg/L nourseothricin ( Werner BioAgents , Germany ) to maintain the plasmid expressing each allele . We note that plasmid-bourn expression of the gene complemented the gene-deletion phenotype , where applicable , in all cases tested ( not shown ) . Allele specific effects were additionally tested by reciprocal hemizygosity analysis ( RHA ) [103] . The HO locus was replaced with the nourseothricin resistance cassette ( NAT-MX ) for each mating type of YPS128 and YJM1444 . These were then crossed with the appropriate deletion strain of opposite mating type and harboring the KanMX cassette , selecting for mated cells resistant to both drugs , to generate heterozygous strains that were hemizigous for the gene in question ( crosses shown in S6 Table ) . Synthetic Hydrolysate ( SynH ) medium mimics the lignocellulosic hydrolysate generated from AFEX ammonium treated corn stover with 90 g glucan/L loading and was prepared as in Sardi et al . ( 2016 ) . Two versions were prepared to represent the complete hydrolysate ( SynH ) and the hydrolysate without the hydrolysate toxin cocktail ( HT ) ( SynH—HT ) , as previously published [32] . Phenotyping for GWA , gene deletion assessment , and RHA , was performed using high throughput growth assays in 24 well plates ( TPP® tissue culture plates , Sigma-Aldrich , St . Louis , MO ) . To prepare the cultures , 10 μl of thawed frozen cell stock were pinned onto YPD agar plates ( 1% yeast extract , 2% peptone , 2% dextrose , 2% agar ) and grown for 3 days at 30°C . Cells were then pre-cultured in 24 well plates containing 1 . 5 ml of YPD liquid , sealed with breathable tape ( AeraSeal , Sigma-Aldrich , St . Louis , MO ) , covered with a lid and incubated at 30°C while shaking for 24 h . Next , 10 μl of saturated culture was transferred to a 24 well plate containing 1 . 5 ml of SynH or SynH-HT where indicated , and grown as the preculture for 24 h . Cell density was measured by optical density at 600 nm ( OD600 ) as ‘final OD’ . Culture medium collected after cells were removed by centrifugation was used to determine glucose concentrations by YSI 2700 Select high performance liquid chromatography ( HPLC ) and refractive index detection ( RID ) ( YSI Incorporated , Yellow Springs , OH ) . Biological replicates were performed on different days . For GWA , we used four different but related phenotype measures of cells growing in SynH or SynH–HTs: 1 ) final OD600 as a measure of growth , 2 ) percent of starting glucose consumed after 24 hours in SynH , 3 ) HT tolerance based on OD600 ( calculated as the ratio of final OD600 in SynH versus final OD600 in SynH -HTs ) , and 4 ) HT tolerance based on glucose consumption ( calculated as the ratio of glucose consumed in SynH versus in SynH -HTs ) . Strains and phenotype scores are listed in S1 Table . Initial phenotyping for GWA was performed in biological duplicates; knockout strains and hemizigous strains were phenotyped in five biological replicates to increase statistical power , whereas homozygous deletion strains were phenotyped in triplicate . Replicates for each batch of strains shown in each figure were performed on separate days , for paired statistical analysis . Experiments done for allele replacements expressed on plasmids were performed in glass tubes using modified synthetic complete medium ( SC ) with high sugar concentrations and the toxin cocktail where indicated ( Sardi et al 2016 ) to mimic SynH but with no ammonium to support nourseothricin selection [104] ( 1 . 7 g/L YNB w/o ammonia sulfate and amino acids , 1 g/L monosodium glutamic acid , 2 g/L amino acid drop-out lacking leucine , 48 μg/L leucine , 90 g/L dextrose , 45 g/L xylose ) . This was required since nourseothricin selection does not work in high-ammonium containing SynH . First , we precultured strains carrying plasmids in SC medium with nourseothricin ( 200 ug/ml ) for 24 h . Next , we inoculated a fresh culture at a starting OD600 of 0 . 1 in 7 ml of the modified synthetic complete medium with nourseothricin ( 200 ug/ml ) and HTs . Cultures were grown for 24 h and phenotyped as described above . Replicates were performed on different days , and thus samples were paired by replicate date for t-test analysis . Anaerobic phenotyping was performed in the anaerobic chamber , where cells were grown in flasks containing 25 ml SynH or SynH-HT and maintained in suspension using a magnetic stir bar . Ethanol production was measured over time by HPLC RID analysis . Paired t-test analysis was performed to determine significance , pairing samples by replicate date . We obtained publicly available whole genome sequencing reads from Saccharomyces cerevisiae sequencing projects [39 , 42 , 52 , 60] . Sequencing reads were mapped to reference genome S288C ( NC_001133 , version 64 [105] ) using bwa-mem [106] with default settings . Single nucleotide polymorphisms ( SNPs ) were identified using GATK [107] Unified Genotyper , analyzing all the strains together to increase detection power . GATK pipeline included base quality score calibration , indel realignment , duplicate removal , and depth coverage analysis . Default parameters were used except for -mbq 25 to reduce false positives . Variants were filtered using GATK suggested criteria: QD < 2 , FS > 60 , MQ < 40 . A dataset with high quality SNPs was generated using VCFtools [108] by applying additional filters of a quality value above 2000 and excluding sites with more that 80% missing data . Genetic variant annotation was performed using SNPEff [109] . Principal component analysis and the neighbor-joining tree were performed with the R package Adegenet 1 . 3–1 [110] using the entire collection of high quality SNPs ( 486 , 302 SNPs ) . SNP data are available in the EBI under accession number PRJEB24747 . Correlations between genotype and phenotype were performed using a mixed linear model implemented in the software GAPIT [68] . Only SNPs with a minor allele frequency ( MAF ) of at least 2% were used for this analysis ( 282 , 150 SNPs ) . Multiple models , each incorporating a different number of principal components to capture population structure ( from 0–3 ) , were analyzed . The final model was manually chosen as the one with the greatest overall agreement between the distribution of expected and the observed p-values , i . e . based on QQ plots with the least skew across the majority of SNPs . We performed four analyses , one for each for the four related phenotypes measured . The model used to map SynH final OD600 and SynH percent glucose consumed used 0 principal components , with population structure corrected using only the kinship generated by GAPIT . The model used to map HT tolerance based on relative final OD600 used 2 principal components , and the model to map HT tolerance based on glucose consumed incorporated 1 principal component . The threshold for significance accounting for multiple-test correction was identified by dividing the critical p-value cutoff of 0 . 05 by the number of independent tests estimated by the SimpleM method [111] , which decreased the number of tests from 282 , 150 to 137 , 398 to produce a p-value threshold of 3 . 6e-7 [112] . However , none of our tests passed this threshold , which is likely overly conservative . We therefore used a p-value threshold of 1e-04 to identify genes for detailed follow-up analysis . We realized that the extreme phenotypes of Asian/sake strains coupled with their strong population structure might be confounding the analysis [66] . Therefore , to further reduce the chance of false positives due to residual population influences , we reran the analyses without the 11 sake strains and removed from the original list of significant SNPs those with p>5e-3 . For each locus carrying a significant SNP , we plotted phenotypic distributions for each possible genotype . We focused subsequent downstream analysis on individual SNPs whose effects were additive across strains that were heterozygous and homozygous at that site , assessed visually . Genes affected by each SNP were determined by the SNPEff annotation , which predicted the effect of variants on genes . | Understanding the genetic architecture of complex traits is important for elucidating the genotype-phenotype relationship . Many studies have sought genetic variants that underlie phenotypic variation across individuals , both to implicate causal variants and to inform on architecture . Here we used genome-wide association analysis to identify genes and processes involved in tolerance of toxins found in plant-biomass hydrolysate , an important substrate for sustainable biofuel production . We found substantial variation in whether or not individual genes were important for tolerance across genetic backgrounds . Whether or not a gene was important in a given strain background explained more variation than the alleleic differences in the gene . These results suggest substantial variation in gene contributions , and perhaps underlying mechanisms , of toxin tolerance . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"biotechnology",
"genome-wide",
"association",
"studies",
"carbohydrate",
"metabolism",
"medicine",
"and",
"health",
"sciences",
"toxins",
"pathology",
"and",
"laboratory",
"medicine",
"glucose",
"metabolism",
"alleles",
"toxic",
"agents",
"toxicology",
"glucose",
"tolerance",
"fungi",
"model",
"organisms",
"genome",
"analysis",
"experimental",
"organism",
"systems",
"genetic",
"engineering",
"research",
"and",
"analysis",
"methods",
"saccharomyces",
"genetic",
"loci",
"yeast",
"biochemistry",
"eukaryota",
"phenotypes",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae",
"genomics",
"metabolism",
"computational",
"biology",
"organisms",
"human",
"genetics"
] | 2018 | Genome-wide association across Saccharomyces cerevisiae strains reveals substantial variation in underlying gene requirements for toxin tolerance |
Meiotic recombination permits exchange of genetic material between homologous chromosomes . The replication protein A ( RPA ) complex , the predominant ssDNA-binding complex , is required for nearly all aspects of DNA metabolism , but its role in mammalian meiotic recombination remains unknown due to the embryonic lethality of RPA mutant mice . RPA is a heterotrimer of RPA1 , RPA2 , and RPA3 . We find that loss of RPA1 , the largest subunit , leads to disappearance of RPA2 and RPA3 , resulting in the absence of the RPA complex . Using an inducible germline-specific inactivation strategy , we find that loss of RPA completely abrogates loading of RAD51/DMC1 recombinases to programmed meiotic DNA double strand breaks , thus blocking strand invasion required for chromosome pairing and synapsis . Surprisingly , loading of MEIOB , SPATA22 , and ATR to DNA double strand breaks is RPA-independent and does not promote RAD51/DMC1 recruitment in the absence of RPA . Finally , inactivation of RPA reduces crossover formation . Our results demonstrate that RPA plays two distinct roles in meiotic recombination: an essential role in recombinase recruitment at early stages and an important role in promoting crossover formation at later stages .
During sexual reproduction , meiotic recombination permits reciprocal exchange of genetic materials between homologous chromosomes and ensures faithful chromosome segregation [1 , 2] . Abnormalities in meiotic recombination are a leading cause of aneuploidy , infertility , and pregnancy loss in humans [3] . Meiotic recombination is initiated by the formation of programmed DNA double strand breaks ( DSBs ) in germ cells and involves a large number of single-stranded DNA ( ssDNA ) -binding proteins [2] . These DSBs undergo end resection to generate 3’ ssDNA overhangs; subsequent loading of RAD51 and DMC1 recombinases and other proteins enables strand invasion into duplex DNA for homologue pairing and recombination intermediate formation [4–7] . All meiotic DSBs are repaired but only a subset lead to crossovers , which are critical for proper segregation of homologous chromosomes during the first meiotic cell division . The replication protein A ( RPA ) complex , comprised of RPA1 , RPA2 , and RPA3 , is the predominant ssDNA-binding heterotrimeric complex in DNA metabolism [8] . RPA1 , the largest subunit , is responsible for the majority of ssDNA-binding activity of the RPA complex . RPA protects ssDNA from degradation and prevents secondary structure formation . RPA interacts with both RAD51 and DMC1 [9] , suggesting that RPA may direct RAD51 and DMC1 to these ssDNA overhangs . RAD51 and DMC1 form nuclear complexes on meiotic chromosomes [10 , 11] . The Hop2-Mnd1 heterodimer interacts with the RAD51/DMC1 recombinases and stimulates their enzymatic activity [12] . MEIOB is a meiosis-specific ssDNA-binding RPA1 homologue [13 , 14] . The MEIOB/SPATA22 heterodimer interacts with RPA and could also position critical proteins for meiotic recombination [15] . These interactions suggest that RPA may play multiple roles in meiotic recombination [16] . A long-standing question remains as to whether RPA is required for loading of these ssDNA-binding proteins to DSBs in vivo . RPA is ubiquitously expressed and essential for DNA replication , repair , and recombination [8] . However , the physiological role of RPA in mammalian meiosis remains elusive , due to embryonic lethality of the Rpa1 mutation in mice [17] . As a result , the interplay of RPA with other ssDNA-binding proteins during meiotic recombination in vivo remains enigmatic . Here we employed a germ cell-specific inducible deletion approach and uncovered the functions of RPA in meiosis .
Using super-resolution imaging microscopy ( SIM ) , we examined the localization pattern of RPA and other meiotic proteins in mouse spermatocytes . RPA is a heterotrimer of RPA1 , RPA2 , and RPA3 [8] . These three RPA subunits are expected to colocalize in the same foci . Indeed , by conventional immunofluorescence microscopy , we previously showed that RPA1 always colocalized with MEIOB and that RPA2 also always colocalized with MEIOB [13] . By SIM , we found that RPA1 foci were present in leptotene , zygotene , early pachytene , and mid pachytene spermatocytes , but absent in late pachytene and diplotene spermatocytes ( Fig 1A–1F ) . We next examined colocalization of RPA2 with other DNA-binding proteins that are known to be involved in meiosis . While most RPA2 foci did not overlap with DMC1 foci ( Fig 1G ) , RPA2 foci always colocalized with MEIOB foci ( Fig 1H ) . These results were consistent with previous conventional immunolocalization findings , suggesting that RPA may function in meiotic recombination in mouse [13 , 18] . To overcome the expected embryonic lethality of Rpa1 inactivation , we generated an Rpa1 floxed ( Rpa1fl ) allele that allows for germ cell-specific inactivation of Rpa1 . We first produced Rpa1fl/- Ddx4-Cre males . Ddx4-Cre begins to express only in germ cells at embryonic day 15 [19] . Rpa1fl/- Ddx4-Cre males were viable but sterile . Testes from 10-week-old Rpa1fl/- Ddx4-Cre males lacked all germ cells ( S1A Fig ) . We also generated Rpa1fl/- Stra8-Cre males . Stra8-Cre begins to express in spermatogonia postnatally , prior to meiosis [20] . Testes from adult Rpa1fl/- Stra8-Cre males displayed a heterogeneous phenotype: some tubules were nearly depleted of all germ cells , while other tubules had apparently normal spermatogenesis ( S1B Fig ) , presumably due to inefficient Stra8-Cre-mediated deletion as observed previously [21] . Although these genetic studies revealed an essential role for RPA1 in spermatogenesis , the loss of all germ cells in testes from Rpa1fl/- Ddx4-Cre males or the presence of all germ cells in testes from Rpa1fl/- Stra8-Cre males precluded investigating its function in meiosis . To circumvent these hurdles , we performed tamoxifen-inducible inactivation of Rpa1 following crossing with Ddx4-CreERT2 mice ( Fig 2A ) [22] . Intraperitoneal injection of 8-week-old Rpa1fl/fl Ddx4-CreERT2 ( referred to as Rpa1cKO ) male mice with tamoxifen resulted in the deletion of exon 8 , which led to a frame shift in the resulting mutant transcript ( Fig 2A and 2B ) . The abundance of the Rpa1 transcript ( both pre and post exon 8 regions ) decreased at days 2 , 4 , and 6 post tamoxifen treatment , suggesting that the Rpa1 transcript is depleted due to nonsense mediated decay and/or depletion of spermatocytes ( S2 Fig ) . Immunofluorescence analysis of spread nuclei revealed the absence of RPA1 foci in mutant spermatocytes at 4 days post-tamoxifen treatment ( dpt ) , showing efficient depletion of Rpa1 in zygotene-like spermatocytes under this regimen ( Fig 2C ) . Interestingly , although RPA2 and RPA3 formed foci in wild type spermatocytes , they failed to form foci on meiotic chromosomes in zygotene-like Rpa1 mutant spermatocytes ( Fig 2C ) . These results were confirmed by the dramatic reduction in the abundance of RPA1 , RPA2 , and RPA3 proteins in tamoxifen-treated testes ( Fig 2D ) . These data demonstrate that the stability of both RPA2 and RPA3 depends on RPA1 and that loss of RPA1 causes the depletion of the RPA complex . Inactivation of Rpa1 results in severe defects in meiotic progression in males ( Fig 3A ) . At 4 dpt , Rpa1cKO seminiferous tubules ( Stage IX ) were partially devoid of leptotene spermatocytes ( Fig 3A ) . At 6 dpt , both leptotene ( stage IX ) and zygotene ( stage XII ) spermatocytes were lost in Rpa1cKO testes . At 8 dpt , in addition to the loss of leptotene and zygotene spermatocytes , there was a partial loss of pachytene spermatocytes ( Fig 3A and 3C ) . At 12 dpt , pachytene spermatocytes exhibited a significant loss ( Fig 3A ) . Progressive loss of spermatocytes was expected to contribute partially to the reduction in protein abundance of RPA and other proteins in the mutant testes ( Fig 2D ) . We monitored meiotic progression and determined the spermatocyte composition in Rpa1cKO males by nuclear spread analysis ( Fig 3B and 3C ) . At 4 dpt , Rpa1cKO testes lacked normal zygotene spermatocytes , characterized by partial chromosomal synapsis , but contained defective zygotene-like spermatocytes , which were characterized by lack of synapsis and presence of unusually strong γH2AX ( Fig 3B ) . The γH2AX intensity in both Rpa1-deficient leptotene and zygotene-like spermatocytes was significantly higher than wild type , suggesting that DSBs were generated but not repaired in the absence of RPA1 ( Fig 3B ) . Zygotene-like spermatocytes were only present in Rpa1cKO testes at 2 to 6 days post tamoxifen treatment with a peak at 2 dpt and 4 dpt ( Fig 3C and 3D ) . Consistent with histological analysis ( Fig 3A ) , only pachytene and diplotene spermatocytes were present in Rpa1 mutant testes at 8 dpt ( Fig 3C and 3D ) . Apoptosis in spermatocytes increased dramatically in stage IV tubules from Rpa1cKO testes ( S3 Fig ) , suggesting that Rpa1-deficient zygotene-like spermatocytes were eliminated by the pachytene checkpoint in response to early meiotic defects ( Fig 3D ) . The synchronized depletion of meiotic germ cells in mutant testes suggests that RPA1 is required for pre-meiotic S phase DNA replication in preleptotene spermatocytes . To test this , we performed a BrdU incorporation ( pulse-labelling ) assay after tamoxifen treatment ( 0 dpt ) ( Fig 4A ) . Preleptotene spermatocytes were only present in stage VII-VIII tubules and were nearly all BrdU-positive in ~80% of such tubules in wild type testes , whereas preleptotene spermatocytes in the remaining 20% of wild type stage VII-VIII tubules were all BrdU-negative ( Fig 4B ) . Partial positive tubules with a mixture of BrdU-positive and BrdU-negative preleptotene spermatocytes were not observed in wild type testes but accounted for 50% of tubules in Rpa1cKO testes ( Fig 4A and 4C ) , supporting the requirement for RPA1 in pre-meiotic S phase DNA replication in BrdU-negative preleptotene spermatocytes . While not observed in wild type testes , nearly 10% of stage VII-VIII Rpa1cKO tubules lacked preleptotene spermatocytes , suggesting that RPA1 is also essential for mitotic DNA replication in spermatogonia , direct precursors of preleptotene spermatocytes . During meiotic recombination , DNA recombinases RAD51 and DMC1 form nuclear complexes on meiotic chromatin [10 , 11] and generate presynaptic filaments on ssDNA that drive strand invasion into homologous DNA duplex [23 , 24] . While RAD51 and DMC1 formed foci on meiotic chromosomes in wild type mice ( Fig 5A ) , such foci were surprisingly absent in Rpa1-deficient zygotene-like spermatocytes ( Fig 5A and 5B ) , although both proteins were present at reduced abundance in the mutant testes as shown by Western blotting ( Fig 2D ) . This result raises the possibility that RPA binding to ssDNA precedes and is a pre-requisite for loading of RAD51 and DMC1 recombinases ( Fig 5C ) . In combination with our colocalization data ( Fig 1 ) , our results suggest that the interaction between RPA and RAD51/DMC1 is transient and that RPA is replaced by RAD51/DMC1 in the foci that form on DSBs . As such , loss of RPA blocks the critical strand invasion step in meiotic recombination and leads to a complete failure in chromosomal synapsis as observed in zygotene-like spermatocytes ( Fig 5C ) . ATR promotes loading of RAD51 and DMC1 to DSBs during meiotic recombination [25 , 26] . In wild type spermatocytes , ATR formed foci on the synaptonemal complex at the zygotene stage and localized to the XY body at the pachytene stage ( S4 Fig ) , consistent with the previous reports [26–28] . In pachytene spermatocytes from Rpa1cKO testes ( 4 dpt ) , ATR still localized to the XY body ( S4B Fig ) . In addition , ATR still formed foci in zygotene-like Rpa1 mutant spermatocytes ( S4A Fig ) . These results show that ATR is not sufficient for loading of RAD51 and DMC1 to DSBs in the absence of RPA . We next assessed if MEIOB contributes to RPA-dependent regulation of meiotic recombination . In wild type mice , the MEIOB/SPATA22 dimer always colocalized with RPA in foci on meiotic chromosomes ( Fig 1H ) [13] . In sharp contrast with RAD51/DMC1 , MEIOB and SPATA22 still co-localized in Rpa1-deficient zygotene-like spermatocytes , suggesting that the formation and loading of MEIOB/SPATA22 dimers on DSBs are independent of RPA ( Fig 5 ) . In addition , this result demonstrates that MEIOB/SPATA22 dimers are insufficient to recruit RAD51/DMC1 to DSBs ( Fig 5C ) . Taken together with the previous finding on the presence of RPA foci in Meiob-deficient spermatocytes [13] , the RPA trimer and the MEIOB/SPATA22 dimer localize to DSBs independently and function non-redundantly in meiotic recombination . Following strand invasion , a subset of DSBs are repaired into crossovers at the pachytene stage . At the subsequent diplotene and metaphase stages , crossover sites become chiasmata to physically link bivalent homologs till chromosome segregation . Strand invasion into the homologue duplex creates the displacement loop ( D-loop ) , which is single stranded . RPA formed foci in early-to-mid pachynema , presumably binding to the D loop and the single stranded second end ( Fig 1 ) . We first examined the RPA depletion efficiency in pachytene spermatocytes at 2 dpt and 6 dpt by nuclear spread analysis . At 2 dpt , the number of RPA2 foci in early-mid pachytene spermatocytes was comparable between control and Rpa1cKO testes ( S5 Fig ) , even though the RPA2 protein abundance was reduced in Rpa1cKO testes ( Fig 2D ) . However , at 6 dpt , in contrast with control early-mid pachytene spermatocytes , foci of RPA1 , RPA2 , or RPA3 were dramatically reduced in number and intensity in the early-mid pachytene mutant spermatocytes ( S6 Fig ) . Therefore , to investigate whether RPA plays a role in crossover formation , we analysed pachytene spermatocytes from Rpa1cKO testes at 6 or 8 days post-tamoxifen treatment . We examined proteins involved in meiotic recombination that form foci at the pachytene stage . MEIOB and SPATA22 foci were still present , but RAD51 and DMC1 foci were absent in Rpa1-deficient early-mid pachytene spermatocytes ( S7 Fig ) . This localization result was similar to that in zygotene-like Rpa1-deficient spermatocytes ( Fig 5 ) , confirming the unique role of RPA in RAD51/DMC1 localization throughout meiosis . We next analysed two meiosis-specific factors , TEX11 and MSH4 [29–31] , which modulate crossover formation . These two factors formed foci in control early-to-mid pachytene spermatocytes ( Fig 6 ) . However , the number of TEX11 foci was significantly reduced in early-mid pachytene spermatocytes from Rpa1 cKO testes at 6 dpt ( Fig 6A ) . Likewise , the number of MSH4 foci decreased significantly ( Fig 6B ) . The reduction in the number of TEX11 and MSH4 foci in Rpa1-deficient pachynema ( Fig 6 ) suggests a role of RPA in crossover formation . To directly assess the effect of RPA on crossovers , we immunostained for MLH1 , which localizes specifically to future crossover sites . MLH1 foci were present but significantly reduced in number in Rpa1-deficient pachynema compared with control at both 6 dpt and 8 dpt ( Fig 7A and 7B ) . Analysis of metaphase nuclei revealed that 20% of metaphase I spermatocytes had univalent chromosomes in Rpa1cKO testes ( Fig 7C and 7D ) . Univalents were observed for both autosomes and sex chromosomes . The presence of univalent chromosomes leads to chromosome mis-segregation and triggers the spindle checkpoint ( Fig 3D ) . Indeed , we detected massive apoptosis of spermatocytes in stage XII seminiferous tubules from Rpa1cKO testes , which contained metaphase and anaphase spermatocytes ( Figs 7E , 7F and 3A—12 dpt ) . These results demonstrate that RPA promotes crossover formation and thus is required for proper chromosome segregation during male meiosis ( Fig 7G ) .
RPA is often included in in vitro recombination reactions but its specific requirement is unknown . Recombinases RAD51 and DMC1 form helical nucleoprotein filaments on ssDNA [7 , 32] and RPA is necessary for efficient RAD51 filament formation [32] . The DNA strand exchange activities catalysed by RAD51 or DMC1 nucleoprotein filaments are rather inefficient but strongly stimulated by RPA [6 , 7 , 32] . These biochemical studies support RPA as an important accessary factor in recombination . While studies of yeast hypomorphic Rpa1 mutants also support a role for RPA in meiotic recombination [33] , these studies did not address whether RPA is required for pre-synaptic filament formation . Using loss of function Rpa1 mouse mutants , we demonstrate that RPA loading to meiotic DSBs is a pre-requisite for RAD51/DMC1 loading in vivo ( Fig 5C ) . One possible explanation is that RPA loading prevents formation of ssDNA secondary structure and recruits RAD51/DMC1 to these sites of recombination . An alternative but less likely explanation is that RAD51/DMC1 nucleation on ssDNA is independent of RPA , but that their localization to DSBs is stabilized by RPA . Finally , the fact that MEIOB/SPATA22 still localizes as foci in Rpa1-null spermatocytes suggests that the requirement of RPA in RAD51/DMC1 loading is unique and cannot be compensated for by other ssDNA-binding complexes such as MEIOB/SPATA22 . MEIOB , SPATA22 , and RPA colocalize in foci on meiotic chromosomes [13 , 34] . In Meiob or Spata22-deficient germ cells , RPA foci are still present [13 , 14 , 34 , 35] . Here we find that MEIOB and SPATA22 still form foci in RPA1-deficient spermatocytes . Collectively , although the MEIOB/SPATA22 dimer and the RPA trimer colocalize in foci and interact with each other [15] , their localizations to DSBs are independent . A previous study suggests that SPATA22 is required for the maintenance but not formation of RAD51 foci in rat spermatocytes [35] . RAD51 and DMC1 foci are present in Meiob-deficient spermatocytes [13] . Therefore , the presence of MEIOB/SPATA22 foci and the absence of RAD51/DMC1 foci in RPA-deficient spermatocytes support that RPA but not MEIOB/SPATA22 is required for formation of RAD51/DMC1 foci . Our genetic study identifies an essential role for RPA in meiotic recombination—strand invasion at the zygotene stage ( Fig 7G ) . Failure in strand invasion causes persistence of DSBs and DNA damage in early Rpa1-deficient zygotene-like spermatocytes , which are eliminated by the first round of apoptosis in stage IV seminiferous tubules due to the pachytene checkpoint ( Fig 3D , Fig 7G and S3 Fig ) . Therefore , Rpa1-deficient zygotene-like spermatocytes would not be expected to progress through the pachytene stage . Our study uncovers a second role for RPA in meiotic recombination—crossover formation at the pachytene stage ( Fig 7G ) . We postulated that the pachytene spermatocytes from Rpa1cKO testes were already at the pachytene stage and lost RPA1 upon tamoxifen injection , and thus were not likely to be derived from the defective zygotene-like spermatocytes . Inactivation of RPA1 in pachytene spermatocytes reduces crossover formation but does not completely abrogate it . This partial effect was likely due to the conditional deletion approach . RPA was dramatically depleted in Rpa1cKO pachytene spermatocytes at 6 dpt , but was still detectable as extremely weak foci at a greatly reduced number ( S6 Fig ) . A second possibility is that the existing RPA1 protein may have played a role in crossover formation before being degraded . Lastly , MEIOB/SPATA22 may exert a more important role in crossover formation than does RPA , since MEIOB/SPATA22 still forms foci in Rpa1-deficient pachytene spermatocytes . We hypothesize that , in the absence of RPA , MEIOB/SPATA22 binds to the recombination intermediates to promote the formation of most but not all crossovers ( Fig 7G ) . Regardless of these scenarios , reduction in crossover formation in Rpa1-deficient pachytene spermatocytes leads to chromosome mis-segregation at the anaphase I stage spermatocytes , which are depleted by the second round of apoptosis due to the spindle checkpoint activation ( Figs 3D and 7G ) . Previously , a knockdown approach was employed to investigate the function of an essential gene Rad51 in mouse spermatogenesis [36] . In this study , we have used a powerful inducible germline-specific inactivation strategy to elucidate the function of RPA , an essential gene involved in DNA metabolism . This approach can be used to study the functions of any somatically essential genes in meiosis . Our results demonstrate that RPA plays a dual function in meiotic recombination: an essential role in presynaptic filament formation and an important role in crossover formation .
Mice were maintained and used for experimentation according to the protocol approved by the Institutional Care and Use Committee of the University of Pennsylvania . Rpa1tm1a mice ( EUCOMM consortium ) [37] were mated with ROSA26-FLPo mice ( Stock number: 007844 , Jackson Laboratory ) [38] to generate a floxed ( Rpa1fl ) allele ( Fig 2 ) . Rpa1fl/fl females were crossed with Ddx4-CreERT2 males ( Stock number: 024760 , Jackson Laboratory ) to generate Rpa1fl/+ Ddx4-CreERT2 mice . Rpa1fl/+ Ddx4-CreERT2 males were crossed with Rpa1fl/fl females to obtain Rpa1fl/fl Ddx4-CreERT2 mice . Rpa1fl/fl Ddx4-CreERT2 mice were fertile and thus were used to breed with Rpa1fl/fl mice to produce more Rpa1fl/fl Ddx4-CreERT2/+ mice for experiments . To generate germ cell-specific Rpa1 knockout mice , injection of tamoxifen was used to induce Cre-mediated deletion of the floxed exon . In brief , tamoxifen ( Sigma , T5648 ) was dissolved in corn oil ( Sigma , C8267 ) at a concentration of 20 mg/ml and injected intraperitoneally into the 8-week-old Rpa1fl/fl Ddx4-CreERT2/+ male mice at a dose of 4 mg/30g body weight for five consecutive days . Untreated Rpa1fl/fl Ddx4-CreERT2/+ littermates were used as controls . Tamoxifen treatment in combination with Ddx4-CreERT2 in otherwise wild type males does not adversely affect meiosis as previously reported [39] . Ddx4-Cre ( Stock number: 006954 ) and Stra8-Cre ( Stock number: 008208 ) mice were obtained from the Jackson laboratory . Mice were genotyped by PCR analysis of tail genomic DNA . Wild-type allele ( 264 bp ) and Rpa1 floxed allele ( 277 bp ) was assayed by PCR with primers GATTATGACACCCTTTGGGACT and TGGCCAAATTAAACCACAGTAACACG . Ddx4-CreERT2 allele ( ~205 bp ) was assayed by PCR with primers ATACCGGAGATCATGCAAGC and GGCCAGGCTGTTCTTCTTAG . The mouse RPA2 full-length and RPA3 full-length were expressed as 6xHis-RPA2 and 6xHis-RPA3 fusion proteins in E . coli using the pQE-30 vector and affinity purified with Ni-NTA agarose . Two rabbits were immunized with each fusion protein ( Cocalico Biologicals Inc . ) . Two guinea pigs were also immunized with the RPA2 recombinant protein . The resulting antisera UP2436 ( rabbit anti-RPA2 ) , GP111 ( guinea pig anti-RPA2 ) , and UP2439 ( anti-RPA3 ) were used for immunofluorescence and western blotting analyses . For histological analysis , testes were fixed in Bouin’s solution , embedded with paraffin , and sectioned . Sections were stained with hematoxylin and eosin . For TUNEL analysis , testes were fixed in 4% paraformaldehyde overnight at 4°C , dehydrated in 30% sucrose and sectioned . Sections were performed with the TUNEL Enzyme and Label Kit ( Roche Boehringer Mannheim ) . Nuclear spread analysis of spermatocytes was performed as previously described [40] . Primary antibodies that were used for immunofluorescence were listed in S1 Table . For quantification of foci , images of spermatocytes from two to four animals were captured and analysed . Axial element markers SYCP2 or SYCP3 were used to classify the stage of meiotic prophase on nuclear spreads [43 , 44] . Early and mid-pachytene spermatocytes were distinguished by the morphology of XY chromosomal axis and the intensity of synaptonemal complex [41] . Characteristics of early pachynema: relatively low intensity of synaptonemal complex staining , short synapsed pseudoautosomal regions , and/or unsynapsed ends of a few autosomes . Characteristics of mid pachynema: strong intensity of synaptonemal complex staining , full synapsis of all autosomal pairs , and often U-shaped XY axis . Characteristics of late pachynema: accumulation of SYCP2/3 proteins at the synaptonemal complex ends and figure -8 shaped XY chromosomes . The abnormal zygotene-like spermatocytes from Rpa1 mutant testes were characterized by extremely strong γH2AX signals and clusters of ends of axial elements . Metaphase spread cells were stained with 4% Gurr Giemsa . For Western blot analysis , testes were homogenized in 500 μl protein extraction buffer ( 62 . 5 mM Tris-HCl ( pH 6 . 8 ) , 3% SDS , 10% glycerol , 5% 2-mercaptoethanol ) . Samples were boiled in 2x loading buffer for 10–15 min to obtain soluble testicular protein extracts . About 10–20 μl of testicular extracts were resolved by SDS-PAGE , transferred onto nitrocellulose membranes using iBlot ( Invitrogen ) , and immunoblotted with indicated antibodies ( S1 Table ) . Color histological images were captured on a Leica DM5500B microscope with a DFC450 digital color camera ( Leica Microsystems ) . Immunolabeled chromosome spreads and testis TUNEL assay images were captured with an ORCA Flash4 . 0 digital monochrome camera ( Hamamatsu Photonics ) on a Leica DM5500B microscope ( Leica Microsystems ) and processed using Photoshop ( Adobe ) software packages . Super-resolution imaging microscopy analysis was performed using a Nikon NSIM super-resolution microscope system and NIS-Elements 2 image processing software . Adult control ( wild type ) and Rpa1cKO male mice were intraperitoneally injected with 50 mg/kg body weight of BrdU ( Sigma , B5002 ) . Adult Rpa1fl/fl Ddx4-CreERT2/+ male mice were treated with daily injection of tamoxifen for five consecutive days as described above . 12 hours after the last tamoxifen injection , a single dose of BrdU was injected . After 2 h , the mice were sacrificed . Testes were collected and fixed in the fixative solution ( 30% formaldehyde , 15% ethyl alcohol , 5% glacial acetic acid ) overnight , embedded in paraffin , and cut in 5 μm sections . After deparaffinization and rehydration , slides were immersed in 20 mM Tris-HCl , pH9 . 0 , at 95°C for 15 min . The slides were blocked with 1% BSA for 1h at room temperature followed by incubating with anti-BrdU and anti-SP10 antibodies at 37°C overnight . The morphology of acrosome revealed by anti-SP10 ( ACRV1 ) antibody was used to precisely identify stage VII-VIII tubules , in which preleptotene spermatocytes are present [42] . Slides were rinsed with PBS and then incubated with rabbit anti-rat ( Vector , FI-4001 ) and goat anti-guinea pig secondary antibodies ( Novus , NB-1206906 ) at 37°C for 1h . Mounting medium with DAPI was added to the slides for imaging . Three pairs of control and Rpa1cKO males ( 0 day post tamoxifen treatment ) were analyzed . At least thirty five stage VII-VIII tubules were counted for each mouse ( Fig 4B and 4C ) . Adult wild type and Rpa1fl/fl Ddx4-CreERT2/+ male mice were treated with daily injection of tamoxifen for five consecutive days as described above . Testes were collected at 2 dpt , 4 dpt , and 6 dpt for semi-quantitative RT-PCR analysis ( S2 Fig ) . The pre-exon 8 Rpa1 region was assayed by RT-PCR ( 212 bp ) using primers GAACACGCTTTCCTCGTTCATGCTG ( exons 3/4 ) and CTTCATTATAGGGCACTGGATTCCC ( exon 6 ) . The post-exon 8 Rpa1 region was assayed by RT-PCR ( 256 bp ) using primers AGAGCTACTGCTTTCAATGAGCAAG ( exon 10 ) and TGTCTACTAGTGCGTCTTTAGCCTT ( exons 11/12 ) . Actb ( 382 bp ) was assayed with primers AGAAGAGCTATGAGCTGCCT and TCATCGTACTCCTGCTTGCT . Statistical analysis was performed with Student's t-test . | Meiosis , a process unique to germ cells , results in production of haploid gametes . Meiotic recombination , a hallmark of meiosis , together with random segregation of homologous chromosomes , generates genetic diversity in haploid gametes at every generation so that each gamete has a unique genetic composition . Such genetic diversity in gametes is important for evolution . Here we report the functional requirement of RPA in meiotic recombination in mouse . RPA is a ubiquitously expressed ssDNA-binding complex and is essential for DNA replication . Mutations in RPA cause lethality . Using an inducible Cre-mediated deletion approach , we find that RPA is required for meiotic recombination in mouse . Inactivation of RPA causes absence of DNA recombinases RAD51 and DMC1 at DNA double-strand breaks , resulting in a block in meiotic recombination at the zygotene stage . In contrast , the ssDNA-binding MEIOB/SPATA22 heterodimers and ATR still form foci on meiotic chromosomes in the absence of RPA . Moreover , inactivation of RPA reduces crossover formation in pachytene spermatocytes . In conclusion , RPA plays two stage-specific functions in the early recombinase recruitment and the late crossover formation respectively during meiotic recombination . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"meiosis",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"spermatocytes",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"germ",
"cells",
"molecular",
"biology",
"techniques",
"dna",
"sperm",
"homologous",
"recombination",
"research",
"and",
"analysis",
"methods",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"cells",
"chromosome",
"biology",
"molecular",
"biology",
"testes",
"biochemistry",
"cell",
"biology",
"anatomy",
"nucleic",
"acids",
"recombinase",
"polymerase",
"amplification",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"dna",
"recombination",
"seminiferous",
"tubules",
"genital",
"anatomy"
] | 2019 | Dual functions for the ssDNA-binding protein RPA in meiotic recombination |
Neural maps are emergent , highly ordered structures that are essential for organizing and presenting synaptic information . Within the embryonic nervous system of Drosophila motoneuron dendrites are organized topographically as a myotopic map that reflects their pattern of innervation in the muscle field . Here we reveal that this fundamental organizational principle exists in adult Drosophila , where the dendrites of leg motoneurons also generate a myotopic map . A single postembryonic neuroblast sequentially generates different leg motoneuron subtypes , starting with those innervating proximal targets and medial neuropil regions and producing progeny that innervate distal muscle targets and lateral neuropil later in the lineage . Thus the cellular distinctions in peripheral targets and central dendritic domains , which make up the myotopic map , are linked to the birth-order of these motoneurons . Our developmental analysis of dendrite growth reveals that this myotopic map is generated by targeting . We demonstrate that the medio-lateral positioning of motoneuron dendrites in the leg neuropil is controlled by the midline signalling systems Slit-Robo and Netrin-Fra . These results reveal that dendritic targeting plays a major role in the formation of myotopic maps and suggests that the coordinate spatial control of both pre- and postsynaptic elements by global neuropilar signals may be an important mechanism for establishing the specificity of synaptic connections .
The fidelity with which connections are made between neurons is a striking feature of nervous system design and essential for proper function [1] . How appropriate presynaptic and postsynaptic elements are brought together during development to generate such ordered connectivity is still a major unanswered question in neurobiology [2] . Most developmental studies investigating the generation of neural maps [3] or synaptic laminae [4] have focussed on the role that presynaptic elements play in establishing normal connectivity and the mechanisms that guide axons [5] . This “axonocentric” bias is understandable as the orderly growth of axons to their targets often reveals an explicit anatomical framework , upon which one can ask questions about mechanisms of network formation [6] . The role that dendrites , the major postsynaptic elements , play in the development of connectivity has been much less explored [7] . Dendrite shape is known to have important implications for neuron function as it determines a cell's integrative properties [8] and dictates the synaptic inputs it will receive [9] , [10] . Thus cell-type-specific programs of dendrite development ultimately have a profound effect on the role a cell plays within a network [11] . Two very different modes of growth can generate a dendritic tree of the same basic shape: neurons can either profusely elaborate dendrites across a wide field and then selectively remove branches from inappropriate territories , or alternatively , dendrite growth can be targeted into distinct territories using guidance mechanisms similar to those found in axons [12] . Examples of both types of growth have been observed . The first mode of growth is seen in mammalian retinal ganglion cells to generate ON and OFF sub-laminae of the Inner Plexiform Layer [13] . The second mode of growth , “dendritic targeting” , is seen in the generation of both neural maps and synaptic laminae . In Drosophila , second-order projection neurons , which convey olfactory information to higher brain centres , target their dendrites to form a “protomap” prior to the arrival of presynaptic olfactory receptor neurons [14] suggesting targeted outgrowth . Similarly , imaging studies in zebrafish show that retinal ganglion cells target the growth of their dendrites to generate lamina-specific projections patterns [15] . Our understanding of neural network structure and function has benefited greatly from studies on the sensorimotor systems of vertebrates [16] and invertebrates [17] . Recent data reveal that the embryonic motoneurons of Drosophila generate a dendritic map within the CNS that represents the innervation of body wall muscles [18] . These central projections are highly ordered and likely reflect some underlying organization of pre-motor interneurons within the network . The map develops in the absence of target muscles , glial cells , or competitive interactions with adjacent dendrites , suggesting that coordinated cell-intrinsic programs for targeting are likely to be important for its assembly [18] . Although our understanding of the molecular mechanisms that control dendritogenesis is still incomplete , a number of transcription factors have been identified that coordinate the patterning of dendritic maps [19] , [20] , [21] . At present however the only downstream “effector” molecule known to be required for dendritic map development is Semaphorin-1a . Both loss- and gain-of-function experiments demonstrate that the levels of Semaphorin-1a , acting cell-autonomously as a receptor or part of a receptor complex , direct the dendritic targeting of projection neurons along the dorsolateral to ventromedial axis of the antennal lobe during map formation [22] . Here we investigate how the dendrites of leg motoneurons are targeted to distinct neuropil territories and how these mechanisms can collectively generate a neural map . The majority of leg motoneurons are born during larval life and the bulk of those are derived from a single neuroblast lineage , lineage 15 [23 , unpublished data] . The neurons of lineage 15 form stereotyped projection patterns , dependent on their birth-order within the lineage . Early-born cells innervate proximal muscle targets and elaborate dendrites from medial to lateral territories , whereas late-born cells innervate more distal muscle groups within the leg and establish dendritic arborizations that are largely confined to lateral territories in the neuropil . Here we show how two subtypes , within this lineage , generate their distinct dendritic arborizations by targeting growth into specific territories using the midline signalling systems of Slit-Robo and Netrin-Fra . These data suggest that cell intrinsic blends of guidance molecules marshal the dendrites of this lineage into appropriate territories in a coordinated fashion to generate a myotopic map . Previous studies in Drosophila have revealed that both sensory neurons and interneurons position their synaptic terminals using the midline signalling cues [24] , [25] . We propose that during development the targeting of both pre- and postsynaptic elements into the same space using global , third-party guidance signals could provide a simple way of establishing the specificity of synaptic connections .
To identify the axonal and dendritic projection patterns of leg motoneurons , we performed MARCM analysis [26] with the OK371-GAL4 driver line , which robustly labels motoneurons [27] . Our previous work revealed that a single postembryonic neuroblast ( insect neural precursor cell ) generates a lineage containing solely motoneurons ( lineage 15 ) [23] . Most neuroblasts in Drosophila divide asymmetrically to generate themselves and a ganglion mother cell , which in turn divide once to give two neurons . With MARCM analysis the precise timing of clone induction by heatshock allows “snapshots” of the type of neuron born at different points within a lineage . To establish which muscles lineage 15 motoneurons innervate , we imaged GFP labelled axons of neuroblast clones directly through the body wall and in the legs of adult flies between 2 and 4 days after eclosion . For lineage 15 motoneurons the most proximal targets are a series of body wall muscles that control the leg , including the extracoxal leg depressor ( unpublished data ) . Axons from the remainder of lineage 15 then travel through the coxa and trochanter to the femur , where they establish connections proximally with the long tendon muscle 2 , ltm2 ( the pretarsal flexor ) , and distally with tibia reductor muscle , tirm ( the accessory tibial flexor muscle ) ( Figure 1A and 1D ) . For appendicular muscle description , see Soler et al . ( 2004 ) [28] . The remaining axons pass the femoral-tibial joint and innervate all muscles in the tibia ( Figure 1A and 1D ) . To gain insight into the central organization of this population , we visualized the projections of lineage 15 clones within the prothoracic neuromere ( Figure 1H and 1K ) of nervous systems counterstained with anti-neuroglian ( Figure 1I ) . Neuroglian is a transmembrane protein that is enriched on the fasiculated primary neurites of adult-specific neurons . In the adult CNS , anti-neuroglian staining reveals a stereotyped scaffold of neurite bundles and tracts that allows us to define positions relative to the midline ( Figure 1H and 1I ) . Neuroblast clones of lineage 15 insert their primary neurites into the anterior region of the leg neuropil ( Figure 1E and 1I ) and then arborize extensively within it ( Figure 1E and 1H ) . In the anterior neuropil the bulk of the dendrites cover lateral and intermediate territories sending a few processes to the midline , whereas in the posterior there are also branches that extend to and cross the midline ( Figure 1E ) . To better understand the organization of lineage 15 and to establish whether its members generate a neural map , we visualized individual neurons . The induction of single-cell MARCM clones during larval life not only allowed us to reveal single cells but also enabled us to ask whether there was a sequential generation of neuronal subtypes . Focusing on the periphery first , we found that the progeny of lineage 15 innervate the proximo-distal axis of the leg in a birth-order-dependent fashion , with the first-born neurons targeting body wall muscles ( unpublished data ) and subsequent neurons innervating more distal targets within the leg . The neurons born following a heatshock at 48 h AH ( after hatching ) make synapses on ltm2 [28] . This cell appears to be a member of a subtype of neurons , of which ∼4 innervate this muscle ( Figure 1A , 1B , and 1D ) . This observation is based on comparisons of the axonal arbors of neuroblast clones versus single neuron clones ( Figure 1A and 1B ) , along with data from a detailed clonal analysis ( Brierley et al . , unpublished ) . The neurons generated by heatshocks at 72 h AH innervate tirm in the distal femur; this neuron is also likely to be a member of a pool of ∼4 neurons ( unpublished data ) . The remaining neurons within lineage 15 innervate the distal most leg segment containing muscles , the tibia . The cell induced at 96 h AH is a member group of ∼3 neurons that innervates the tarsal depressor muscle of the tibia ( tadm ) ( Figure 1A , 1C , and 1D ) . The observation of multiple isomorphic neurons innervating muscles in the leg is supported by a recent independent study [29] and consistent with data from larger insects [30] . Single-cell clones allowed us to visualize the dendritic projections of lineage 15 neurons with unprecedented detail . The arborizations of the first-born neurons in this lineage , labelled by heatshocks of newly hatched larvae , cover much of the leg neuropil with branches extending from the anterior medial territories ( arrowhead in Figure 1E ) through to the lateral edge ( unpublished data ) . The neurons generated from a 48 h AH heatshock also span a large part of the leg neuropil running from medial territories to the lateral edge , with the majority of dendrites in an intermediate position ( Figure 1F and 1H ) . The cells generated following a heatshock at 72 h AH span the intermediate and lateral territory but do not send branches to the midline ( unpublished data ) . Contrasting with earlier born cells , neurons generated at 96 h AH have dendrites that are located in lateral neuropil , with some branches in the intermediate territory ( Figure 1G ) . We focused our studies on the neurons generated by heatshocks 48 h and 96 h AH as these are representative of two subtypes of leg motoneurons that cover clearly distinct territories within the map ( Figure 1H ) . A more detailed description of our studies of the developmental origins and architecture of leg motoneurons will be published elsewhere ( Brierley et al . , in prep ) . We used a plot profile tool to perform quantitative analysis on the dendrites of single-cell clones . The distances along the medio-lateral ( x ) axis were first scaled from the interval 0 ( midline ) to 100 ( lateral edge ) , and then weighted means were computed ( using average pixel intensity as weights ) . These weighted means ( i . e . , centres of mass ) were then used to compute statistical significance between profiles . We also computed the location of the mean 33rd percentile pixel intensity , which was chosen as a proxy for the spread and asymmetry of the intensity profiles . Each histogram presents the accumulated data of 16 neurons ( two groups of eight ) and is divided into medial ( M ) , intermediate ( I ) , and lateral ( L ) neuropil territories . Analysis of 48 h AH and 96 h AH single-cell clones show how these neurons distribute their dendritic processes in a stereotyped manner across the medio-lateral axis of the leg neuropil . The mean centre of arbor mass is significantly different between the two groups ( p = 0 . 00003 ) , and the 33rd percentile of the 48 h AH neurons is closer to the midline than the 96 h AH neurons ( 51% versus 64% ) . These data revealed to us the existence of a birth-order-dependent organization of projection patterns generated by lineage 15 . We have found four subtypes within this lineage based on the morphology of the dendritic arborizations and the target areas they innervate . Early-born cells innervate proximal muscle targets and elaborate dendrites from medial to lateral territories , whereas late-born cells innervate distal muscle groups within the leg and position dendrites in the lateral neuropil . Cells born between these two extremes generate dendrites that occupy intermediate positions . Taken together these data show that lineage 15 generates a myotopic map in the medio-lateral axis of the leg neuropil by birth-order-based developmental mechanisms . To gain insight into the developmental mechanisms that control the formation of myotopic maps , we looked at the way that the 48 and 96 h AH subtypes establish their dendritic arborizations . Two quite different modes of dendritic elaboration could be envisaged: dendrites could grow exuberantly initially and then eliminate branches from inappropriate territories; alternatively , they could restrict their growth to within appropriate target territories throughout their development . To establish which mode of growth these motoneurons use to generate their dendritic trees , we performed a timeline analysis through the pupal-adult transition . At 20 h after puparium formation ( APF ) the axons of all leg motoneurons have left the nervous system . Both the 48 and 96 h AH neurons have generated filopodia along their primary neurites in the lateral neuropil ( Figure 2A and 2F ) . At 30 h APF the two neurons begin to show differences in their coverage of the neuropil , with the 48 h AH neuron generating filopodia and small branches that extend into the intermediate territory , and large numbers of filopodia and small branches appear at sites where the anterior and posterior branches will form ( Figure 2B ) . The 96 h AH cell generates a similar quantity of filopodia and small branches , but these are focused in the lateral neuropil , one oriented towards the anterior and the other to the posterior , at a site where major lateral branches will ultimately form ( Figure 2G ) . At 40 h APF the dendrites of both neurons have significantly increased their size and complexity and already show features that are characteristic of the adult neuron . The 48 h AH neuron elaborates a branch on the anterior length of the primary neurite , i . e . , proximal to the cell body and also two growth zones at sites that will become the posterior branches , including the large one that crosses the midline ( Figure 2C ) . At this time the 96 h AH cell has established higher order branches on both the anterior and posterior lateral branches . There are also a large number of small branches along the length of the primary neurite , many of which have filopodia extending in the direction of the midline ( Figure 2H ) . At 60 h APF the 48 h AH neuron has increased its number of higher order branches and the major posterior branch has reached the midline but has not crossed it . In the 96 h AH cell the dendrites covers a large part of the lateral neuropil . Higher order branches can be seen oriented towards the midline , but these never extend into medial territories ( Figure 2D and 2I ) . At 80 h APF , which equates to ∼80% through the pupal-adult transition , the overall morphology of both neurons is indistinguishable from those seen in adults ( Figure 2E and 2J ) . It is possible that local refinements of the dendritic tree take place , but we did not focus on those . These data reveal two important features of myotopic map formation in the leg neuropil: firstly , that the different subtypes of neurons within lineage 15 initiate dendritogenesis synchronously , not sequentially , regardless of their actual birth date; secondly , that both generate subtype-specific dendritic morphologies by growing into distinct territories throughout development , rather elaborating extensively and undergoing large-scale remodelling ( Figure 2E and 2J ) . Thus , these data reveal that dendritic targeting plays a major role in generating a myotopic map in the leg neuropil .
The majority of leg motoneurons in a fly are born postembryonically and most of those are derived from a single neuroblast lineage , termed lineage 15 [23 , Brierley et al . , in prep] . Perhaps the most striking feature of this lineage is its birth-order-based pattern of innervation along the proximo-distal axis of the leg . Using mosaic analysis we observed the sequential production of four neuronal subtypes during larval life , each elaborating stereotyped axonal and dendritic projections in the adult . The axon of the first-born neuron innervates a muscle in the body wall and subsequent neurons innervate more distal targets in the leg . This organization has also been recently reported by Baek and Mann ( 2009 ) [29] . This birth-order-based peripheral pattern of lineage 15 is mirrored in the CNS , where dendrites generate a stereotyped anatomical organization . Dendrites of early-born cells span medial to lateral territories , whereas late-born cells elaborate dendrites in the lateral neuropil and cells born between these times occupy intermediate territories . The sequential production of neuronal subtypes by neural precursor cells is a common mechanism for generating a diversity of circuit components [45] . A similar birth-order-based specification of axonal and dendritic projection patterns has previously been described for projection neurons in the fly's olfactory system [46] , [47] . Our data reveal the existence of a myotopic map in the adult fly and supports the proposition that dendritic maps are a common organizing principle of all motor systems [18] . Mauss et al . ( a companion paper ) also reveal a map in the embryonic CNS of Drosophila , where the dendrites of motoneurons are organized along the medio-lateral axis of the neuropil representing dorsoventral patterns of innervation in the body wall muscles . How are dendritic maps built ? The myotopic map we see in the leg neuropil could be generated by two distinctly different mechanisms . Neurons could elaborate their dendrites profusely across a wide field and then remove branches from inappropriate regions or , alternatively , they could target the growth of dendrites into a distinct region of neuropil throughout development . Both mechanisms can generate cell-type-specific projection patterns as seen in the vertebrate retina [13] , [15] , [48] . To reveal which mechanism is deployed in the leg motor system of Drosophila , we imaged single-cell clones of motoneuron subtypes generated by heatshocks at 48 and 96 h AH , as their final dendritic arborizations cover clearly distinct territories within the map . The dendrites of both elaborate branches only in territories where the mature arborizations eventually reside , which strongly supports the notion that this myotopic map is generated by targeting and not large-scale branch elimination . Importantly , this developmental timeline also revealed that the motoneurons elaborate their dendrites synchronously , regardless of the birth date of the cell . This observation suggests that a “space-filling/occupancy based” model , where later-born neurons are excluded from medial territories by competitive interactions is unlikely . Similarly , heterochronic mechanisms where different members of the lineage experience different signalling landscapes due to differences in the timing of outgrowth are not likely either . With synchronous outgrowth dendrites experience the same set of extracellular signals , suggesting that the intrinsic properties of cells , defined by their birth order , may be more important for the generation of subtype-specific projections . Such intrinsic properties could include cell-cell recognition systems such as adhesion molecules , e . g . , Dscams [49] or classical guidance receptors [5] , [7] , that could interpret extracellular signals . In the Drosophila embryo motoneurons also use dendritic targeting to generate a myotopic map [see Mauss et al . ] . It is emerging that dendrites are guided by the same molecules that control axon pathfinding [7] , [12] , [33] . The medio-lateral organization of leg motoneuron dendrites within the leg neuropil prompted us to ask whether the midline signalling molecules Slit and Netrin and their respective receptors Roundabout and Frazzled could be involved in targeting growth to specific territories . Using mosaic analysis we found that both the 48 and 96 h AH motoneuron subtypes require Robo to generate their appropriate shape and position within the medio-lateral axis . When we removed Robo from the 48 h AH subtype the mean centre of arbor mass was shifted toward the midline . The dendrites of 96 h AH neurons showed a shift in distribution in the absence of Robo but still failed to reach the midline , suggesting that only part of this cell's targeting is due to repulsive cues mediated by the Robo receptor . We predicted that if Robo levels played an instructive role in dendrite targeting we would be able to shift dendrites laterally by cell autonomously increasing Robo . We found this to be the case in both subtypes . Taken together these data suggest that differences in the level of Robo signalling may provide a mechanism by which Slit could be differentially interpreted to allow subtype-specific targeting along the medio-lateral axis . The Robo receptor is part of a larger family of receptors that includes Robo2 and Robo3 [50] , [51] . This family of receptors have been found to be important for targeting axons to the appropriate longitudinal pathway in the embryonic CNS [50] , [51] . Comm plays a key role in allowing contralaterally projecting neurons to cross the midline [35] , and its ectopic expression ( CommGOF ) is known to robustly knock down Robo [31] and Robo2 and 3 [52] . We cell autonomously expressed Comm in both lineage 15 subtypes and found shifts to the midline in both 48 and 96 h AH neurons . For the 48 h AH neurons , Robo LOF data and CommGOF data are comparable , suggesting that Robo alone plays a major role in the positioning dendrites of these cells . In contrast , in the 96 h AH subtype we found RoboLOF and CommGOF effects to be significantly different , suggesting that the 96 h AH subtype may not only use the Robo receptor but additional Robos as well . Our knockdown of Slit also supports this idea , as we occasionally find the branches of late-born neurons reaching the midline ( Figure 6C ) , something we never see in RoboLOF clones . Thus , one way of establishing differences in the medio-lateral position could be through a dendritic “Robo code” where early-born cells express Robo and late-born cell express multiple Robo receptors . With Netrin being expressed in the midline cells during the pupal-adult transition we wondered whether attractive Netrin-Fra signalling could also contribute to positioning dendrites in the leg neuropil . When Fra was removed from the 48 h AH subtype we found that the arborization was shifted laterally , whereas removing it from the 96 h AH subtype had little effect , neither did the removal of Netrin A and B from the midline ( Figure 6E ) , suggesting that Netrin-Fra signalling may not play a role in dendritic targeting in the later-born cell . It may be that Fra is expressed in early-born cells within the lineage and then down-regulated , although we cannot exclude that Netrin-Fra signalling masked by the repulsion from Slit-Robo signalling . These data are consistent with Fra being a major player in targeting the dendrites of the 48 h AH cell . The fact that both Fra and Robo are required for normal morphogenesis of 48 h AH neurons raises the possibility that members of lineage 15 could use a “push–pull” mechanism for positioning their dendrites , where the blend of receptors within a cell dictates the territory within the map that they will innervate . How could such subtype-specific blends of receptors be established ? A number of studies have revealed that spatial codes of transcription factors are important for specifying the identity of motoneuron populations [53]–[55] . Within lineage 15 it is possible that temporal , rather than spatial , transcription factor codes are important for regulating the blend of guidance receptors . A number of molecules have been identified that control the sequential generation of cell types within neuroblast lineages [45] , [56] . Chief amongst these are a series of transcription factors that include Hunchback , Krüppel , Pdm , Castor and Seven-up . These temporal transcription factors are transiently expressed within neuroblasts and endow daughter neurons with distinct “temporal identities” . Castor and Seven-up are known to schedule transitions in postembryonic lineages , regulating the neuronal expression of BTB-POZ transcription factors Chinmo and Broad [57] , [58] . It is possible that the temporal transcription factors Broad and Chinmo could control the subtype-specific expression of different Robo receptors or the Netrin receptor Frazzled in leg motoneurons . There is a precedent for this in the Drosophila embryo , where motoneuron axon guidance decisions to distal ( dorsal ) versus proximal ( ventral ) targets are orchestrated by Even-Skipped , a homeobox transcription factor [59] , which in turn controls the expression of distinct Netrin receptor combinations [60] . Studies focusing on the growth of olfactory projection neuron dendrites in Drosophila reveal that they elaborate a glomerular protomap prior to the arrival of olfactory receptor neurons [14] suggesting that target/partner-derived factors may not be necessary for establishing coarse patterning of synaptic specificity . The global nature of the signals we describe here and their origin in a third-party tissue is a fundamentally different situation to that where target-derived factors instruct partner cells , such as presynaptic amacrine cells signalling to retinal ganglion cell dendrites in the zebrafish retina [15] . Furthermore , although we show that Slit and Netrin control the positioning of dendrites across the medio-lateral axis of the CNS in this study , it may be that other similar guidance signals are important for patterning dendrites in other axes [61] . There is a striking conservation of the molecular mechanisms that build myotopic maps in the embryo and pupae [see article by Mauss et al . ] . Understanding the similarities and differences between these myotopic maps , from an anatomical , developmental , and functional perspective , may give us insight into the evolution of motor systems and neural networks in general . In our study we found that individual leg motoneurons that lacked Robo signalling appeared to have more complex dendritic arborizations . Our working hypothesis , that dendrites invaded medial territories because of a failure of Slit-Robo guidance function , did not take into account the possibility that cells may generate more dendrites due to a change in a cell-intrinsic growth program . Thus the changes we see in dendrite distribution relative to the midline could formally be a result of “spill-over” from that increase in cell size/mass . To determine whether this was the case we generated larger cells by activating the insulin pathway in single motoneurons . We found the dendrites of these “large cells” remained within their normal neuropil territory , supporting the idea that the removal of Robo-Slit signalling results in a disruption in guidance , not growth . These data underline the fundamental importance of midline signals in controlling the spatial coordinates that these motoneuron dendrites occupy , i . e . , that a neuron twice the size/mass of a wild-type cell is still marshalled into the same volume of neuropil . When we reconstructed our image stacks to look at the distribution of the dendrites in the dorso-ventral axis , we found that the apparent increase in size was in fact a redistribution of the dendrites from ventral territories into more dorsal medial domains . This was unexpected and suggests that changes in midline signalling can also impact the organization of dendrites in the dorso-ventral axis . So CommGOF 96 h AH neurons may not only encounter novel synaptic inputs by projecting into medial territories , but they may also lose inputs from the ventral domains of neuropil they have vacated . Our observations suggest that motoneurons within lineage 15 have a fixed quota of dendrites and where it is distributed in space depends on cell-intrinsic blends of guidance receptors . Taken together these data support the idea that growth and guidance mechanisms are genetically separable programs [62] , [63] . In identified embryonic motoneurons where Slit-Robo and Netrin-Fra signalling has been disrupted , quantitative analysis reveals dendrites also show no measurable difference in their total number of branch tips or length [see accompanying article Mauss et al . ] . Moreover , recent computational studies in larger flies reveal that dendritic arborizations generated by the same branching programs can generate very different shapes depending on how their “dendritic span” restricted within the neuropil [64] . Previous work in both vertebrates and Drosophila has shown that a loss of Slit-Robo signalling results in a reduction in dendrite growth and complexity [65]–[67] , but we do not find evidence to support this . Neural maps and synaptic laminae are universal features of nervous system design and are essential for organizing and presenting synaptic information . How the appropriate pre- and postsynaptic elements within such structures are brought together remains a major unanswered question in neurobiology . Studies in recent years have shown that neural network development involves both hardwired molecular guidance mechanisms and activity-dependent processes; the relative contribution that each makes is still unclear [3] , [10] . Work by Li and colleagues [68] on the spinal cord network of Xenopus embryos revealed that seven identifiable neuron subtypes can establish connections with one another and that the key predictor of connectivity was their anatomical overlap . This could be interpreted to mean that connectivity is promiscuous and that the major requirement for the generation of synaptic specificity is the proximity of axons and dendrites . This is particularly interesting in light of our dendrite targeting data and the observation that both sensory neurons and interneurons in Drosophila use the same midline cues to position their pre-synaptic terminals in the CNS [24] , [25] . Moreover , a recent study has shown that Semaphorins control the positioning of axons within the dorso-ventral axis [61] . Taken together these observations suggest that during development the coordinated targeting of both pre- and postsynaptic elements into the same space using global , third-party guidance signals could provide a simple way of establishing the specificity of synaptic connections within neural networks . This idea is akin to “meeting places” such as the traditional rendezvous underneath the four-sided clock at Waterloo railway station where two interested parties organize to meet . Understanding how morphogenetic programs contribute to the generation of synaptic specificity is likely to be key to solving the problem of neural network formation .
The following stocks were used for this study: The MARCM system [26] was used to fractionate the GAL4 enhancer trap line OK371 , which strongly labels excitatory motoneurons [27] . To generate MARCM clones , eggs were collected on standard fly food for 7 h and allowed to age to the required stage ( at 25°C ) . Following this staging heat shocks were performed by incubating vials at 37°C for 45 min , followed by a period of 30 min at 25°C and then 37°C for 30 min . Three time points were used during this study: Newly Hatched Larval stage , 48 h AH , and 96 h AH . After heatshocks , larvae were reared on standard fly food at either 25°C or room temperature when precise staging was not required . Nervous systems were dissected from larvae , pupae , and adults in phosphate buffered saline ( pH 7 . 8 ) ( PBS ) and fixed in 3 . 7% buffered formaldehyde for 45 min at room temperature and then washed three times in PBS containing 1% Triton-X100 ( PBS-TX ) . Fixed samples were blocked in PBS-TX containing 2% normal donkey serum ( Jackson ImmunoResearch Laboratories , West Grove , PA , USA ) for 30 min and then incubated in combinations of primary antibodies for 1 to 2 d at 4°C . After washing with PBS-TX over 8 h , tissues were incubated overnight at 4°C in combinations of secondary antibodies for 1 to 2 d at 4°C . Following repeated washes with PBS-TX , tissues were mounted on poly-L-lysine coated coverslips , dehydrated , cleared in xylene , and mounted in DPX ( Fluka , Bachs , Switzerland ) . For the detergent free immunohistochemical method , the above protocol was repeated replacing PBS-TX with PBS at all stages . The following primary antibodies were used: ( 1 ) rabbit ( rb ) anti-GFP ( 1∶500; Invitrogen ) , ( 2 ) rat ( rt ) anti-mCD8 ( 1∶500; Caltag Laboratories , Burlingame , CA , USA ) , ( 3 ) mouse ( ms ) anti-Neuroglian ( BP104 , 1∶5; DSHB ) , ( 4 ) ms anti-robo ( 1∶100 ) , ( 5 ) ms anti-myc ( 9E 10 , 1∶10; DSHB ) , ( 6 ) ms anti-Slit ( C555 . 6D , 1∶300; DSHB ) . The following secondary antibodies were used: ( 1 ) FITC conjugated donkey anti-rabbit IgG ( 1∶500; Invitrogen ) , ( 2 ) Texas Red conjugated donkey anti-mouse IgG ( 1∶500; Invitrogen ) , ( 3 ) Alexa Fluor 488 donkey anti-Rb ( 1∶500; Invitrogen ) , and ( 4 ) FITC conjugated donkey anti-Rat IgG ( 1∶500; Invitrogen ) . Fluorescently stained nervous systems were imaged at 40× using a Zeiss LSM510 confocal microscope . Z-stacks were collected with optical sections at 1 . 5 µm intervals . Raw data stacks were imported into NIH Image J ( http://rsb . info . nih . gov . nih-image/ ) . In cases where multiple cells were labeled in nearby but non-overlapping regions , we used the lasso tool to remove any processes that obscured MN clones . In any situation where there was a possible conflict , stacks were discarded from analysis . The maximum z-projections were then imported into Photoshop ( Adobe , San Jose , CA , USA ) and minor adjustments were made to the brightness and contrast where required . To quantify the distribution of the dendrites of single-cell clones along the medio-lateral axis of the neuropil , we used the plot profile tool in ImageJ . Analysis was performed on maximum z-projections . Care was taken to obtain images that had equivalent grayscale ranges . A box was drawn around the whole arborization excluding the soma , the neuroglian counterstain was used for registration , and the distances along the x-axis were first scaled from the interval 0 ( midline ) to 100 ( lateral edge ) . Plot profile displays a two-dimensional histogram of average pixel intensities along the x-axis . Weighted means were computed ( using average pixel intensity as weights ) , i . e . , centres of arbor mass were then used to compute statistical significance between profiles . We also computed the location of the mean 33rd percentile pixel intensity to give an indication of the asymmetry of profiles . Nervous systems were measured for each experimental condition and individual profiles were overlayed using Photoshop ( Adobe , San Jose , CA , USA ) ; each histogram presents the accumulated data of 16 neurons ( two groups of eight ) . The medio-lateral axis of the neuropil was divided into three neuropil territories: medial ( M ) , intermediate ( I ) , and lateral ( L ) . Statistical analysis was done with two-tailed t test . Using the Welch correction , p denotes the significance ( * p<0 . 05 , ** p<0 . 01 , and *** p<0 . 001 ) . A quantitative measure of dendritic arborization complexity of individual ddaC neurons was determined using Sholl analysis [35] . The number of intersections between dendritic processes and Sholl-rings were counted in Photoshop using a template of 12 concentric circles ( each 22 µm apart ) centred on the cell body . The volume of a cell was measured using Volocity Acquisition software ( version 4 . 2 , Improvision ) . Raw data stacks were imported and 3D projections of the cell body constructed . The lasso tool was then used to capture the cell body and the volume subsequently measured . The VGN9281-GAL4 line was generated by cloning 849 bp of wild-type upstream DNA of the Drosophila DVGlut gene from position 2L: 2397582 , 2410668 . This corresponds to from −3 . 8 Kb to −4 . 6 Kb upstream of the translation start site ( used as a reference point ) . This region was cloned into the EcoR1 and BamH1 sites of a modified GAL4 P-element vector , pPTGAL-attB . The pPTGAL vector , a gift from Daniel F . Eberl [70] , was modified by cloning 285 bp of attB sequence from pUASTB , a gift from Michele P . Calos , into the XbaI and BglII sites of pPTGAL . The phiC31 site-specific integration system [71] was used to target the construct to chromosome 3L . Primers used to amplify attB sequence from pUASTB were: Forward primer , attBFXb- 5′CAGTCTCTAGAGTCGACGATGTAGGTCACGGTC 3′ and reverse primer , attBRBg- 5′CAGTCAGATCTGTCGACATGCCCGCCGTGACCG 3′ . Primers used to generate VGN9281-Gal4 line were: Forward primer , 9281EcoF-5′CAGTCGAATTCTAAGGCGATTCCTCCAAGTG 3′ and reverse primer , 9281BamR-5′ CAGTCGGATCCGAATCGGGCGAGGACTTC 3′ . | During development the axons of sensory neurons generate highly ordered ”sensory maps„ within the nervous system that represent specific qualities of the environment . Much less is known about the anatomical organization and development of motor systems . Here , we show that the leg motoneurons of Drosophila organize their dendrites within the central nervous system in a way that reflects the position of the muscles they innervate . These motoneurons generate a ‘myotopic map’ by targeting the growth of their dendrites ( sites of synaptic input ) into discrete territories during development . The precise targeting of dendrites along the mediolateral axis is controlled by the signaling molecules Slit and Netrin , which are secreted by midline cells . These proteins act as global guidance cues and exert their effects via distinct signaling pathways using receptors called Roundabout and Frazzled , respectively . Previous studies have shown that Slit also helps to position the termini of axons ( sites of synaptic output ) , independent of their synaptic partners . We suggest that the coordinated targeting of both input and output elements of a neural system into a common space using shared global guidance cues could be a simple way of establishing the specificity of synaptic connections within neural networks . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"neuroscience/motor",
"systems",
"neuroscience/neurodevelopment",
"developmental",
"biology/neurodevelopment"
] | 2009 | Dendritic Targeting in the Leg Neuropil of Drosophila: The Role of Midline Signalling Molecules in Generating a Myotopic Map |
Airway inflammation plays a major role in the pathogenesis of influenza viruses and can lead to a fatal outcome . One of the challenging objectives in the field of influenza research is the identification of the molecular bases associated to the immunopathological disorders developed during infection . While its precise function in the virus cycle is still unclear , the viral protein PB1-F2 is proposed to exert a deleterious activity within the infected host . Using an engineered recombinant virus unable to express PB1-F2 and its wild-type homolog , we analyzed and compared the pathogenicity and host response developed by the two viruses in a mouse model . We confirmed that the deletion of PB1-F2 renders the virus less virulent . The global transcriptomic analyses of the infected lungs revealed a potent impact of PB1-F2 on the response developed by the host . Thus , after two days post-infection , PB1-F2 invalidation severely decreased the number of genes activated by the host . PB1-F2 expression induced an increase in the number and level of expression of activated genes linked to cell death , inflammatory response and neutrophil chemotaxis . When generating interactive gene networks specific to PB1-F2 , we identified IFN-γ as a central regulator of PB1-F2-regulated genes . The enhanced cell death of airway-recruited leukocytes was evidenced using an apoptosis assay , confirming the pro-apoptotic properties of PB1-F2 . Using a NF-kB luciferase adenoviral vector , we were able to quantify in vivo the implication of NF-kB in the inflammation mediated by the influenza virus infection; we found that PB1-F2 expression intensifies the NF-kB activity . Finally , we quantified the neutrophil recruitment within the airways , and showed that this type of leukocyte is more abundant during the infection of the wild-type virus . Collectively , these data demonstrate that PB1-F2 strongly influences the early host response during IAV infection and provides new insights into the mechanisms by which PB1-F2 mediates virulence .
Influenza A virus ( IAV ) commonly causes acute respiratory infection and is one of the most important human pathogens , causing between 250 , 000 and 500 , 000 deaths every year around the world [1] . IAV are enveloped viruses belonging to the Orthomyxoviridae family . Their negative strand RNA genome is composed of 8 segments encoding up to 12 proteins . PB1-F2 is a virulence factor first described 10 years ago [2] . This 75–90 amino acid long accessory protein is encoded by an alternative +1 reading frame on segment 2 which also encodes the RNA polymerase basic protein 1 ( PB1 ) and N40 , an N-terminally truncated version of PB1 lacking transcriptase function [2] , [3] . PB1-F2 is expressed by most IAV strains and has been shown to be associated with immunopathological processes observed during infection [4] , [5] , [6] . During IAV infection , production of pro-inflammatory cytokines generally results in an innate host response that controls the virus propagation until it is eliminated by the adaptive immune system . However , in some cases , excessive host inflammatory response contributes to disease severity , especially with highly pathogenic strains such as avian H5N1 or the 1918 pandemic H1N1 strain ( “Spanish flu” ) [7] , [8] . PB1-F2 is suspected to contribute to this disproportional response which frequently leads to vital respiratory tissue damage and death of the infected person [5] . Several in vivo studies allowed determining the pathophysiological consequences of PB1-F2 expression during IAV infection . Using an attenuated model of mouse infection , Zamarin et al . demonstrated that wild type ( wt ) IAV displays a higher pathogenicity than its PB1-F2 knockout counterpart [9] . A single N66S mutation present in the PB1-F2 of the 1918 pandemic strain is sufficient to transform a strain of moderate virulence into a highly pathogenic virus in mice [10] . More recently , a study analyzing the effects of PB1-F2 in the IAV-induced pathogenesis of avian hosts also revealed that amino acid changes within the PB1-F2 reading frame of a highly pathogenic avian IAV decrease lethality in ducks [11] . PB1-F2 was also described as facilitating secondary infection with Streptococcus pneumoniae [12] . In contrast , when introducing a functional PB1-F2 in the 2009 pandemic H1N1 IAV that does not encode this accessory protein , only minimal immune response modulation was observed , underlying the complex contribution of PB1-F2 in virulence [13] . We recently showed that PB1-F2 exacerbates IFN-β production through the activation of the NF-κB pathway during IAV infection of the human respiratory epithelial cell line A549 , but not during infection of immune cells [6] . Despite its apoptotic functions , pathways related to cell death were not activated differently in wt and PB1-F2 knockout virus-infected A549 cells . In the present study , we analyzed the global transcriptional response of the mouse respiratory tract associated to PB1-F2 during IAV infection . We report that PB1-F2 expression during IAV infection increases the inflammatory response of the host . PB1-F2 plays a major role in NF-κB pathway activation , chemotaxis of granulocyte and apoptosis of recruited immune cells . Furthermore , we determined that IFN-γ expression is enhanced when PB1-F2 is expressed , and appears to have a central position in the gene network responsible for the respiratory disease provoked by IAV .
In order to analyze the impact of PB1-F2 expression in pathogenicity , we first examined the kinetics of PB1-F2 expression in a mouse infection model using the mouse-adapted A/WSN/1933 ( H1N1 ) strain of IAV . The time course of PB1-F2 expression in the lungs of infected mice was analyzed . PB1-F2 was efficiently expressed at day 2 post-infection ( pi ) and expression increased at day 3 and 4 pi ( Fig . 1A ) . Densitometric measurement of PB1-F2 levels standardized to β-actin levels on distinct infected mice confirmed the increase of PB1-F2 expression at days 3 and 4 pi ( Fig . S1 ) . Using a sensorchip assay ( Biacore ) with an anti-PB1-F2 monoclonal antibody [14] , the PB1-F2 concentration was estimated to range around 4 pmol per 100 pmol of NP at day 4 pi , corresponding to 26 pmol of PB1-F2 per mg of infected lung tissue . We then aimed to determine the specific effect of PB1-F2 expression on the mouse mortality rate using the wt virus and a mutant knocked-out for PB1-F2 expression . To this end , and because the introduction of mutations at the initiation and internal AUG codons of the PB1-F2 open reading frame ( ORF ) may result in the generation of truncated PB1 products [2] , [3] , we generated mutants unable to reinitiate translation in the two ORFs: PB1-F2-null and N40-null mutants ( Fig . S2A ) . We then compared the effect of PB1-F2- and N40-null mutations on PB1 expression in the infected mouse epithelial cell line MLE15 in order to verify that all viruses expressed similar amounts of PB1 ( Fig . S2B ) . Furthermore , to determine if N40 expression may modify the host immune response , infections of MLE15 cells by the wt virus and its mutants were carried out and IFN-β synthesis induction analyzed . As expected , the N40 null-mutants induced similar amounts of IFN-β than the wt virus , allowing us to exclude the involvement of N40 in host-response exacerbation processes ( Fig . S2C ) . Therefore , we retained the fully knocked-out mutant ΔF2 for further mice infections and subsequent analyses [6] . While 100% mortality was reached when mice were inoculated with 1×106 PFU with the wt and the ΔF2 viruses , the 50% Lethal Doses ( LD50 ) were estimated to be 1 . 48×105 PFU [95% confidence interval ( CI ) , 0 . 97×105 - 1 . 9×105] with the wt virus and 2 . 81×105 PFU [CI , 2 . 6×105 - 2 . 99×105] with the ΔF2 virus . No mortality was observed when mice were inoculated with 5×104 PFU of the two viruses . A survival curve after a viral challenge using 2 . 5×105 PFU is shown in Fig . 1B . A lower mortality rate was observed in the ΔF2 infected mice when compared to the wt infected mice ( 40% vs . 80% ) . Immunohisto-labelling of NS1 in mouse-infected lungs showed equivalent signal intensities for the two viruses indicating an identical viral infectivity ( Fig . S3 ) . These data are indicative of an impact of PB1-F2 expression in IAV-induced pathogenicity . To investigate the contribution of PB1-F2 in virus pathogenicity , we infected groups of mice at high and low infectious doses to measure and compare morbidities , replication levels and IFN-β inductions . Measurement of IFN-β was carried out since we previously observed that PB1-F2 exacerbates IFN-β expression within in vitro infected human pulmonary cells [6] . In response to a high viral challenge ( 1×106 PFU ) , the two groups of mice ( wt- and ΔF2-infected mice ) were as sensitive ( 0% survival for both groups ) and lost the same weight ( Fig . 2A ) . When infections were carried out with lower sub-lethal doses of virus ( 5×104 PFU ) , wt-infected mice lost a greater percentage of body weight than their ΔF2-infected counterparts: 29% vs . 14% at day 6 pi ( Fig . 2B ) . The differential weight loss persisted during the viral infection . To determine whether the morbidity associated to PB1-F2 expression could be attributed to a reduced viral replication in ΔF2 infected mice , we assessed viral RNA in the lungs by qRT-PCR . As shown in Fig . 2C and 2D , at day 3 pi , both viruses replicate with the same efficiencies when mice were infected with a high or a low viral dose . Wt- and ΔF2-virus titers were also identical at days 2 and 4 pi ( data not shown ) . These observations are consistent with previous studies [6] , [9] . We then measured the level of induction of IFN-β gene transcription at day 3 pi in the two conditions of viral challenge . When mice were infected with 1×106 PFU , both viral infections induced a strong IFN-β expression with the same range of magnitude ( Fig . 2E ) . At the lower infection dose , while IFN-β gene transcription is still strongly induced in wt-infected mice , a reduced fold of induction was observed in ΔF2-infected mice ( Fig . 2F ) . Collectively our results indicate that , as we previously observed in cultured human epithelial pulmonary cells , PB1-F2 is able to exacerbate IFN-β expression in infected mice . To provide additional insights on the pathogenicity associated to PB1-F2 expression during IAV infection , we investigated transcriptional profiles of the host genes in the lungs of mice infected by the wt- and the ΔF2-virus by microarray analyses . Mice were infected intranasally with 1 . 5×105 PFU and total RNA were extracted from mice lungs at day 2 , 3 , and 4 pi and analyzed by microarrays containing more than 44 , 000 oligonucleotides representative of the whole murine transcriptome . The transcriptomic profiles obtained at day 2 pi showed the most relevant differences between the two infection conditions . A Venn diagram shown in Fig . 3A summarizes the differentially expressed genes common or unique to wt and ΔF2 virus infections at day 2 pi . A total of 6025 genes showed a greater-than-2-fold up- or down-regulation in wt-infected lungs relative to mock-infected lung ( p<0 . 05 ) . By contrast , when looking at the regulated-genes during the infection with the ΔF2 virus , only 937 genes were regulated , demonstrating that PB1-F2 severely influence the early host response during IAV infection of mice . The most significant canonical pathways specifically associated with wt and ΔF2 virus infections according to Ingenuity Pathway Analysis ( IPA ) software are listed in Fig . 3B . The analysis revealed that both viruses induced genes involved in “Respiratory disease” showing that wt and ΔF2 viruses are both able to establish a deleterious infection in C57Bl/6 mice . Genes involved in the “Cell death” pathway are expressed at high levels in mice infected by the wt virus while it is not statistically regulated in ΔF2-infected mice . This is consistent with the fact that PB1-F2 is described as a proapoptotic factor in immune cells [2] . Another main functional consequence of the PB1-F2 deletion observed is the drastically reduced inflammatory response developed during the ΔF2 virus infection . This suggests a potent pro-inflammatory activity of PB1-F2 that could be an important contributor to severe immunopathology associated with highly pathogenic IAV strains . We also observed a correlation between PB1-F2 expression and the pathways “Tissue Morphology” and “Hematological System Development and Function” , which are direct consequences of the inflammatory response . As shown in Fig . 3C , genes implicated in inflammatory processes were shown to be preferentially regulated during the wt IAV infection , including genes involved in chemotaxis and activation of granulocytes ( Cxcl2 , Cxcl3 , Csf3 , Trem1 ) , genes encoding acute-phase proteins ( Saa1 , Saa3 and Saa4 ) and multiple other genes implicated in defense response . In order to validate the microarray expression data , we assayed several genes presented in Fig . 3C by qRT-PCR ( Ccr1 , Cxcl1 , Csf3 , Ptx3 , Tnf-α and Trem1 ) . These experiments performed on independent mice confirmed the PB1-F2 effect on inflammatory genes ( Fig . S4 ) . Collectively , these results suggest that PB1-F2 amplifies the inflammation observed during IAV infection , a feature that was partially identified in our previous analysis using the A549 cell infection model . To further explore the host responses associated with PB1-F2 expression , we characterized the functional relationship between genes that were induced or inhibited more than two fold ( p<0 . 05 ) between wt- and ΔF2-infected mice . Fig . 4A shows a representation of the significant gene network associated with PB1-F2 expression during mice infection . Top functions associated with this gene network were related to immune cell trafficking , inflammatory response and cell death . Of particular interest is the central position of IFN-γ in this network . IFN-γ is linked to multiple genes including peptidases , apoptotic activators , chemokines and cytokines . Most of these genes are involved in immune cell functions or in recruitment of monocytes and granulocytes towards sites of tissue infection . To confirm the ability of PB1-F2 to promote an increase in inflammatory mediator secretion during infection , we dosed by ELISA the amount of several cytokines involved in the inflammation process in the bronchoalveolar lavage ( BAL ) fluid of infected animals . Thus , IFNγ was quantified due to its central position within the network , TNFα because of its potent role in inflammatory processes , CXCL1/KC for its important role in neutrophil chemoattraction and IL6 as the most important mediator of fever and for its role in the acute phase response . As shown in Fig . 4B , these four cytokines were present in lower amounts in the BAL of mice infected by the ΔF2 virus when compared with BAL obtained from animals infected with the wt virus . These statistically relevant observations further confirmed the pro-inflammatory capacities of PB1-F2 during IAV infection . Altogether , our results indicate that upon influenza infection , the respiratory tract exhibits increased lysosomal proteases , leucocyte chemoattractants , and inflammatory signaling mRNA transcripts when PB1-F2 is expressed and suggest that this viral protein plays a major role in acute lung inflammatory processes and might be involved in severe immunopathology . We sought to verify the microarray results by assessing apoptosis in cells found in BAL fluids . BAL total cells were harvested at day 3 pi and the apoptotic index monitored . We opted for a method allowing the distinction between necrosis and apoptosis , and giving a result in the form of a ratio , excluding differences in terms of number of cells . As shown in Fig . 5A , the apoptotic index of cells recruited in the airways of wt-infected mice is significantly increased when compared to the mock-infected mice . Conversely , when apoptosis quantification was made in ΔF2-infected mice , we found that the apoptosis index was significantly reduced when compared to wt-infected mice . We next used the IPA software to identify the apoptosis-related signalling pathways activated during wt IAV infection of mice lungs . As shown in Fig . 5B , multiple genes involved in the pathway linked to TNF/FasL-mediated apoptosis are induced ( red ) when mice were infected with the wt virus , including numerous caspases . Apoptotic genes induced under these conditions suggest that the PB1-F2-mediated apoptosis response of the infected lung is mediated by activation of genes promoting cell shrinkage and membrane blebbing . NF-κB is the major pro-inflammatory transcriptional factor and regulates the induction of most proinflammatory cytokines . We recently observed that PB1-F2 mediates an upregulation of IFN-β by NF-κB activation in human pulmonary cells [6] . We thus analyzed the contribution of PB1-F2 in the NF-κB activation during an in vivo IAV infection . We used an adenovirus containing an NF-κB response element linked to a luciferase reporter gene ( Ad-NF-κB-luc ) [15] . As a positive control , we showed that E . Coli LPS induces NF-κB activation ( Fig . S5 ) . Control mice infected with Ad-NF-κB-luc alone ( PBS ) showed no NF-κB activity , attesting that the vector has no inflammatory activity by itself . Mice were co-infected with the Ad-NF-κB-luc and either the wt IAV or the ΔF2 IAV ( 1 . 5×105 PFU per mouse ) . Mice were then followed daily for luciferase activity measurement ( Fig . 6A ) . Wt-infected mice developed an important inflammatory reaction with a peak of activity after 3 days pi in contrast to animals infected by the ΔF2 virus that developed a moderate NF-κB activity ( Fig . 6B ) . Quantification of the luciferase activity showed that the difference is observable at day 3 pi , which corresponds to the viral replication peak ( data not shown ) . To determine whether PB1-F2 expression could be responsible for an increased recruitment of granulocytes during IAV infection , as suggested by the microarray data , we assessed cellular infiltration in the air spaces . Leukocytes were harvested from the BAL of mock- , wt- and ΔF2-infected mice 3 days pi , the number and phenotype of the cells were characterized by cytospin slide counting . Analysis of the BAL leukocyte populations demonstrated that IAV infections induced the accumulation of polynuclear neutrophils ( PNN ) in the airways . PNN constitute more than 80 % of the cells found in the BAL fluid of infected mice . The PNN accumulation peaked at day 3 pi and decreased thereafter ( Fig . 7A ) . When compared to ΔF2-infected mice , wt-infected mice PNN counts revealed a significantly more important afflux of PNN at day 3 pi ( Fig . 7A ) . To further investigate the PB1-F2 effect in the lung tissue , we quantified the myeloperoxidase ( MPO , a lysosomal protein specific for PNN ) signal at day 3 pi by immunohistology . Infected lung sections presented characteristic histological aspects related to acute inflammation . Staining of the viral NS1 protein showed a characteristic bronchial localization that was homogenous and equivalent in intensity for the two viruses ( Fig . 7C ) . Immunohistochemical localization of MPO showed an intense signal in bronchial tissue when mice were infected with the wt virus at day 3 pi . However , when ΔF2-infected mice were analyzed , a weaker MPO signal was observed in the bronchial tissue ( Fig . 7B , 7C ) highlighting the potent pro-inflammatory effect of PB1-F2 during IAV infection . Collectively , these histological analyses designate PB1-F2 as a virulence factor playing a harmful role in the pathogenesis of IAV infection .
Innate immune response is considered as an essential process required for efficient pathogen clearance . However , deregulation of this response can lead to severe damage for the host . The regulation of host defenses to IAV infections is particularly complex and leukocyte recruitment to the inflammatory lung , which is beneficial for clearing the virus , can become deleterious to the host tissues if the reaction is excessive [16] . In fact , although inflammation is essential for IAV clearance , it is becoming increasingly evident that a tight regulation of this process is necessary to avoid development of IAV-related immunopathological sequelae [8] , [17] . It illustrates the ambiguity of the host response to infections . The strength of the response has to be proportionate to the pathogen attack in order to eliminate it . If the response is too weak , the pathogen will proliferate , leading to infection . If the host response is too strong , uncontrolled inflammatory reaction will lead to tissue destruction . The present study demonstrates that PB1-F2 is endowed with pro-inflammatory activity and plays a role in the immunopathological consequences observed in severe impairment of influenza-infected lung function . By the use of reverse genetics , we obtained a mutant virus unable to express PB1-F2 and showed that this mutant was less lethal and virulent than the wt virus , consistent with a previous study [9] . These differences cannot be explained by reduced replication efficiency of the ΔF2 virus since no replication variation was observed between the two viruses . Besides , the use of mutants deleted for N40 or PB1-F2 exclude a potential role of N40 in this process . Our previous study identified IFN-β gene as a PB1-F2-mediated exacerbated gene during IAV infection of a human respiratory epithelial cell line [6] . Based on these previous results , we hypothesized that PB1-F2 expression may stimulate in vivo the host immune response , and as a consequence , be a key element that could explain the higher virulence of a virus expressing PB1-F2 . IFN-β measurement within infected-mice lungs confirmed our preliminary in vitro observation . To further study whether PB1-F2 could cause immunopathology , we compared the global gene expression patterns of wt- and ΔF2-infected mouse lungs . Our data demonstrate that expression of PB1-F2 during IAV infection enhances the host defense responses by strongly increasing the transcriptional profile of genes involved in inflammation , granulocyte migration and apoptotic pathways . No apoptotic pathways were identified when these two viruses were used to infect human pulmonary epithelial cells in vitro [6] . This may reflect the cell-specific behavior of PB1-F2 that only induces apoptotic process within immune cells and not in epithelial cells [2] . Thus , the use of the wt/ΔF2 virus couple showed the ability of PB1-F2 to modulate the host response , contributing to virulence in vivo . This is supported by a recent work by McAuley et al . characterizing the immunomodulatory properties of several variants of the PB1-F2 protein [5] . Using IPA software , we were able to analyze the functional relationships between the genes differentially regulated when PB1-F2 was expressed or not during infection . The analysis revealed the central role of IFN-γ , which could act as a relay between PB1-F2 and the genes implicated in respiratory disease , inflammatory response , immune cell trafficking , tissue remodeling and apoptosis induction . The PB1-F2-specific increase of IFN-γ expression was confirmed at the protein level in BAL fluids of infected mice . IFN-γ is a member of the type II interferon family and is a potent activator of macrophages and neutrophils . It is worth noting that in the IAV-infected pulmonary environment , IFN-γ represses the innate immunity developed by alveolar macrophages to enhance their adaptive antiviral response through increased MHC gene expression [18] . Consequently , the scavenger receptor MARCO is down-regulated , and innate defense against secondary bacterial infections is suppressed . The link that we identified between PB1-F2 and IFN-γ over-expression appears of first importance since PB1-F2 has been previously described as facilitating the development of opportunistic pneumococcal and staphylococcal infections [12] , [19] . Collectively , our data provide a rationale to understand how PB1-F2 exacerbate secondary bacterial infections and post-influenza pneumonia: PB1-F2 induces an apoptosis increase of recruited leucocytes together with a switch of their activation state from innate immunity to adaptive antiviral response through an exacerbation of the IFN-γ expression in the lung . The mechanism by which PB1-F2 mediates inflammatory increase by itself is still unknown . However , NF-κB pathway exacerbation is involved in this process as shown by the dramatic decrease of NF-κB activity in lungs of mice infected by the ΔF2 virus when compared to the wt infection ( Fig . 6 ) . The targeting of PB1-F2 to mitochondria could explain the NF-κB pathway exacerbation through a membrane destabilization of this organelle and activation of the RIG-I/MAVS signaling pathway . However PB1-F2 mitochondrial localization is a matter of debate since several studies described differential localization depending on the strain of the virus and the sequence polymorphism of the protein [20] , [21] . The capability of PB1-F2 to perforate or target other cellular membranes could also explain its inflammatory and apoptotic activities [14] , [22] , [23] . Considering the membrane affinity of PB1-F2 , it could form protein pores in endoplasmic reticulum and subsequently induce the release of Ca2+ within the cytosol . Such increase in concentration of Ca2+ in the cytosol is described to activate the NF-κB pathway and to play a pivotal role in inducing proinflammatory gene transcription in airway epithelial cells [24] . Another important characteristic of PB1-F2 is its propensity to form or promote the formation of amyloïdic fibers in infected monocytes [14] . These fibrillar aggregates are believed to associate to the membrane , disrupt its integrity and lead to perturbation of the cellular compartmentalization . This suggests that PB1-F2 could contribute to the pulmonary dysfunctions observed throughout IAV infections by several pathways . A striking feature of the inflammation exacerbated by PB1-F2 expression is the enhanced neutrophil recruitment within infected airways . There is evidence , based on their function , that neutrophils play an important role in mediating acute injury characteristic of highly pathogenic IAV infections . Circulating neutrophils migrate to the site of infection and participate in the destruction of pathogens . However this process has to be tightly regulated since the beneficial effect for eliminating microbes can become deleterious to host tissues through the development of lesions . Indeed neutrophils have the potential to damage airspaces by releasing serine proteases and by generating reactive oxygen species [25] . Neutrophil recruitment within lung tissue also increases protein permeability across the endothelial and epithelial barriers of the lung . This leads to the flooding of alveoli by plasma liquid and proteins and is characteristic of early lung injury [26] . However , depletion of neutrophils prior to influenza infection increased viral load and mortality compared to non-treated mice [27] , confirming the importance of this type of leucocyte during IAV infection , and underlying the complexity of the host-IAV interactions . A fine balance between inflammation and immunity is necessary to eliminate IAV . Our study shows that PB1-F2 is implicated in the dysfunction of this balance and that it induces a massive recruitment of neutrophils within airspaces through deregulation of the innate host defense . In summary , the present study demonstrated that PB1-F2 expression significantly increases the expression of genes associated with inflammation in the airways of IAV-infected mice . We identified IFN-γ as a pivotal host component implicated in this process , orchestrating immune cell apoptosis induction , granulocyte recruitment and tissue remodeling observed in infected lungs . The PB1-F2 specific NF-κB pathway exacerbation that we revealed is probably involved in this IFN-γ expression increase . Further studies are required to correlate PB1-F2 sequence variability and IAV virulence to help in the prediction of the pathogenicity of emerging virus strains in the human population .
This study was carried out in accordance with INRA guidelines in compliance with European animal welfare regulation . The protocol was approved by the Animal Care and Use Committee at CRJ under relevant institutional ( DSV , permit number: 7827 ) and INRA “Santé Animale” department guidelines . All experimental procedures were performed in a Biosafety level 2 facility . Influenza A/WSN/1933 ( H1N1 ) was used in this study . Wild type and PB1-F2 knockout viruses were produced using the 12 plasmid reverse genetics system [28] . The viruses were prepared as previously described [14] . Female C57Bl/6 and Balb/c mice were purchased from the Centre d'Elevage R . Janvier ( Le Genest Saint-Isle , France ) and were used around 8 weeks of age . Mice strains were bred in an animal facility in pathogen-free conditions . Mice were fed normal mouse chow and water ad libitum and were reared and housed under standard conditions with air filtration . For infection experiments , mice were housed in cages inside stainless steel isolation cabinets that were ventilated under negative pressure with HEPA-filtered air . Mice were anesthetized by a mixture of ketamine and xylazine ( 1 and 0 . 2 mg per mouse , respectively ) and infected intranasally with 50 µl of PBS containing 1×106 , 5×105 , 2 . 5×105 , 1 . 5×105 or 5×104 plaque forming units ( PFU ) of IAV . To determine the lethal dose 50% ( LD50 ) , groups of 5 mice were infected with different dilutions of virus and observed for signs of morbidity and death over 18 days . Alternatively , mice were killed at different time points , BAL fluids and lungs were then collected as previously described [29] . Total RNA isolated from mice lungs using Qiagen Rneasy kit ( Qiagen ) was treated with DNase I and reverse-transcribed with superscript reverse transcriptase ( Invitrogen ) using random hexamers ( Fermentas ) or the specific IAV M1 primer : 5′-TCT AAC CGA GGT CGA AAC GTA-3′ for virus quantification [29] , [30] . PCR was performed in 20 µl reactions with specific detection primer pairs for mouse IFN-β ( sense : 5′-CCC TAT GGA GAT GAC GGA GA-3′ ; antisense : 5′-CTG TCT GCT GGT GGA GTT CA-3′ ) and IAV M1 ( sense : 5′-AAG ACC AAT CCT GTC ACC TCT GA-3′ ; antisense : 5′-CAA AGC GTC TAC GCT GCA GTC C-3′ ) . The mRNA levels of IFN-β and vRNA levels of M1 were assayed using the Mastercycler realplex sequence detector ( Eppendorf ) and the double strand specific dye SYBR Green system ( Applied Biosystems ) . The PCR conditions and cycles were as follows: initial DNA denaturation 10 min at 95°C , followed by 40 cycles at 95°C for 15 sec , followed by an annealing step at 64°C for 15 sec , and then extension at 72°C during 30 sec . Each point was performed in triplicate . To ensure that the primers produced a single and specific PCR amplification product , a dissociation curve was performed at the end of the PCR cycle . Relative quantitative evaluation was performed by the comparative ΔΔCt method . The mean ΔCt obtained in non stimulated cells for each gene was used as calibrator , after normalization to endogenous control β-actin ( sense : 5′-TGT TAC CAA CTG GGA CGA CA-3′ ; antisense : 5′-GGG GTG TTG AAG GTC TCA AA-3′ ) . The results are presented as an n-fold difference relative to calibrator ( RQ = 2−ΔΔCt ) . Separate microarrays were run for each experimental sample ( one sample per mouse and three mice per time point ) . Transcriptional profiling was performed using Agilent's Whole Mouse Genome Microarray Kit , 4×44 K ( G4122F ) . Experiments were performed at the “Plateau d'Instrumentation et de Compétences en Transcriptomique” ( PiCT ) , INRA Jouy-en-Josas research center . Minimum Information about Microarray Experiment ( MIAME ) was deposited in ArrayExpress at EMBL ( http://www . ebi . ac . uk/microarray-as/ae ) . A dual color , balanced design was used to provide two direct comparisons: [uninfected/infected-by-wt-virus] and [uninfected/infected-by-ΔF2-virus] . Arrays were hybridized according to the manufacturer's instructions and as previously described [14] . For functional analysis the data files resulting from differential analysis were imported into GeneSpring GX 11 software ( Agilent Technologies , Massy , France ) . Hierarchical clustering analysis was performed to analyze cellular genes that were differentially expressed during infection ( Euclidian distance , average linkage ) . For further analysis , data files were uploaded into the Ingenuity Pathways Analysis ( IPA ) software ( Ingenuity Systems , Redwood City , CA; www . ingenuity . com ) . Right-tailed Fisher's exact test was used to calculate a p-value determining the probability that each biological function and disease assigned to that data set is due to chance alone . IL-6 , TNF-α , IFN-γ and CXCL1/KC concentrations in mice BAL were determined using DuoSet ELISA kits obtained from R&D Systems ( Minneapolis , Minnesota , United States ) . Apoptosis quantification in cells present in the BAL fluids was performed using the ApoGlow Assay Kit ( Lonza ) according to the manufacturer's procedures . The assay is based on the bioluminescent measurement of Adenylate Nucleotide Ratio ( ADP/ATP ) . Recombinant replication-deficient adenovirus ( purchased from “Gene Transfer Vector Core” , University of Iowa ) was used to transduce NF-κB-luciferase ( Ad-NF-κB-luc ) construct in the lung of mice . Ad-NF-κB-luc was prepared as previously described [15] . Mice received intranasal instillations of Ad-NF-κB-luc ( 2 . 5×108 PFU ) . Photon emission of the luminescent construct transducted in the lungs of mock-infected , wt- and ΔF2-IAV-infected mice ( 1 . 5×105 PFU per mouse ) was measured using the IVIS system ( Xenogen Biosciences ) . A digital false-color photon emission image of the mouse was generated , and photons were counted within a constant defined area corresponding to the surface of the chest encompassing the whole lung region . Photon emission was measured as photons emitted per second . Paraformaldehyde-fixed cryosections were permeabilized by 0 . 1% Triton X-100 . Endogenous peroxidase activity was inhibited with 3% H2O2 and nonspecific sites were blocked for 1 hr with 2% bovine serum albumin . Sections were incubated with the relevant primary antibody ( goat IgG anti-mMPO R&D Systems , goat anti-NS1 , santa cruz biotechnology ) for two hours at room temperature and with fluorochrome-conjugated secondary antibodies ( bovine anti-IgG goat Cy5 , rabbit anti-IgG goat Cy3 , Jakson Immuno Research Laboratories ) for 1 hr at room temperature . Nuclei were counterstained with DAPI . Double stainings were performed by mixing the primary antibodies and mixing fluorochrome-conjugated reagents , respectively . Quantification of fluorescence staining was made on 6 slides from 3 different mice in each group using Image J software . Cytokine levels , viral loads , apoptosis measurements , luminescence measurements and PMN counts are expressed as the mean ± standard deviation ( SD ) of at least three separate replicates , and statistical analyzes were performed using the unpaired Student t-test . Entrez accession number ( http://www . ncbi . nlm . nih . gov/Entrez ) of the Influenza A virus ( A/WSN/1933 ( H1N1 ) ) segment 2 complete sequence containing PB1-F2 ORF: CY034138 . Swiss Prot accession number ( http://expasy . org/sprot ) of the Influenza A virus ( A/WSN/1933 ( H1N1 ) ) protein PB1-F2 : B4URF6; and PB1 : B4URF5 . | Influenza A viruses may cause severe respiratory disease . PB1-F2 , a viral protein identified in 2001 is suspected to play a role in influenza-related pneumonia . In order to understand the impact of PB1-F2 in the pathogenesis underlying Influenza A virus infection , we engineered a mutant virus unable to express PB1-F2 . By the use of high-throughput gene expression assays , we compared the host responses of the wild-type-infected and the PB1-F2 mutant-infected mice . We identified that PB1-F2 expression enhances the immune cell death and inflammatory responses of mice . The inflammatory response mediated by the PB1-F2 expression leads to a massive recruitment of leukocytes within the air spaces , a feature that characterizes the influenza-mediated immunopathology . Our results suggest that PB1-F2 is a virulence factor implicated in the deregulation of the inflammatory response observed in acute influenza virus pneumonia . These data underlie the complexities of virus-host interactions and help us understand by which mechanisms Influenza viruses mediate severe respiratory diseases . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"animal",
"models",
"of",
"infection",
"inflammation",
"immunity",
"virology",
"immune",
"activation",
"innate",
"immunity",
"immune",
"defense",
"biology",
"microbiology",
"immunoregulation"
] | 2011 | Transcriptomic Analysis of Host Immune and Cell Death Responses Associated with the Influenza A Virus PB1-F2 Protein |
Uropathogenic E . coli ( UPEC ) is the primary cause of urinary tract infections ( UTI ) affecting approximately 150 million people worldwide . Here , we revealed the importance of transcriptional regulator hypoxia-inducible factor-1 α subunit ( HIF-1α ) in innate defense against UPEC-mediated UTI . The effects of AKB-4924 , a HIF-1α stabilizing agent , were studied using human uroepithelial cells ( 5637 ) and a murine UTI model . UPEC adherence and invasion were significantly reduced in 5637 cells when HIF-1α protein was allowed to accumulate . Uroepithelial cells treated with AKB-4924 also experienced reduced cell death and exfoliation upon UPEC challenge . In vivo , fewer UPEC were recovered from the urine , bladders and kidneys of mice treated transurethrally with AKB-4924 , whereas increased bacteria were recovered from bladders of mice with a HIF-1α deletion . Bladders and kidneys of AKB-4924 treated mice developed less inflammation as evidenced by decreased pro-inflammatory cytokine release and neutrophil activity . AKB-4924 impairs infection in uroepithelial cells and bladders , and could be correlated with enhanced production of nitric oxide and antimicrobial peptides cathelicidin and β-defensin-2 . We conclude that HIF-1α transcriptional regulation plays a key role in defense of the urinary tract against UPEC infection , and that pharmacological HIF-1α boosting could be explored further as an adjunctive therapy strategy for serious or recurrent UTI .
Urinary tract infection ( UTI ) is a very common and frequently recurrent bacterial disease accounting for approximately 8 million doctor visits per year , with an estimated total cost of $450 million annually in the United States [1 , 2] . More than 50% of women will experience at least one UTI during their lifetime , and about 30–40% of UTI recur within 6 months [3] . The invasive pathogen uropathogenic E . coli ( UPEC ) is a primary etiologic agent of UTI , causing severe bladder infection ( cystitis ) and acute kidney infections ( pyelonephritis ) [4 , 5] . To successfully establish infection , UPEC must first attach to superficial facet cells of bladder epithelium ( uroepithelium ) , followed by invasion ( entry ) into the cytosolic milieu of these cells . UPEC colonization and invasion triggers an acute inflammatory response in the bladder epithelium leading to release of multiple pro-inflammatory cytokines including interleukin 6 ( IL-6 ) and IL-1β , and chemokines such as IL-8 [6–10] , and recruitment of neutrophils , which appear in the urine [11 , 12] . The inflammatory consequences of this early innate immune response can promote structural damage and cell death , including rapid shedding of the superficial uroepithelial cell layer lining the surface of the bladder lumen , which is a hallmark of UPEC infection [13 , 14] . Although pro-inflammatory activation is an important first line of defense against pathogens , an excessive response can promote chronic cystitis and acute or chronic pyelonephritis [15–18] . Hypoxia inducible factor-1α ( HIF-1α ) is a transcriptional regulator that orchestrates the cellular response to low oxygen stress . HIF-1α is degraded at normoxia through a prolyl-hydroxylase ( PHD ) and proteasome-dependent pathway , but under low oxygen ( hypoxic ) conditions translocates to the nucleus , where it activates expression of multiple gene targets including glucose transporters , enzymes of glycolysis , erythropoietin , and vascular endothelial growth factor [19] . An emerging literature has revealed significant intersection between the hypoxic and innate immune responses , and HIF-1α is now recognized to play key role in modulating innate immune cell function [20–22] . Mice with targeted deletion of HIF-1α in myeloid cells ( neutrophils and macrophages ) and skin or corneal keratinocytes are more susceptible to bacterial infection , with diminished antimicrobial peptide ( AMP ) and nitric oxide ( NO ) production and impaired microbicidal capacity demonstrable in the corresponding HIF-1α-deficient cells [23–25] . Conversely , pharmacological stabilizers of HIF-1α ( PHD inhibitors ) are in clinical development for treatment of anemia [26 , 27] , and the potential for repositioning such drugs as “innate immune boosters” has been explored in animal models of bacterial infection . HIF-1α stabilizing agents increase the bactericidal capacity of phagocytic cells and epithelial cells [24 , 25 , 28 , 29] , show therapeutic benefit in mouse models of Staphylococcus aureus skin infection [28 , 29] , reduce intestinal tract inflammation and bacterial translocation in murine chemically-induced colitis [30] , and support host defense during early stages of mycobacterial infection in zebrafish [31] . The dynamics of HIF-1α expression has been studied in the context of bladder cancer [32]; however , its role during bladder infections has not been explored . In this study , we couple in vitro and in vivo models of UTI to examine the role of HIF-1α in host defense of the urinary tract against UPEC , with an eye toward pharmacological HIF-1α boosting as a novel therapeutic approach in this highly prevalent and difficult to manage infectious disease condition .
AKB-4924 ( Aerpio Therapeutics ) is a prolyl-hydroxylase inhibitor drug candidate that preferentially stabilizes HIF-1α and increases phagocyte killing of S . aureus in vitro and in a murine skin infection model [29] . We found that treatment with 20 μM AKB-4924 for 2 h significantly increased HIF-1α protein abundance in healthy human uroepithelial cell line 5637 ( ATCC HTB-9 ) , comparable to the effect of the hallmark HIF-1α agonist desferrioxamine mesylate ( DFO ) ( Fig 1A ) . This result confirms that AKB-4924 stabilizes HIF-1α to prevent its degradation; resulting in an increase in HIF-1α protein . HIF-1α expression in human 5637 cells is not significantly altered during UPEC CFT073 infection ( Fig 1B ) , but becomes upregulated upon AKB-4924 treatment ( Fig 1C ) . To corroborate this result , we examined the transcription level of the gene encoding vascular endothelial growth factor ( VEGF ) , a peptide cytokine classically induced by HIF-1α at the transcriptional level [25 , 33] . AKB-4924 treatment increased VEGF mRNA by approximately 6-fold ( Fig 1C ) . Thus , to examine how AKB-4924-mediated HIF-1α upregulation influenced UPEC survival during uroepitheilal cell infection , we recovered bacterial colony forming units ( CFU ) 2 h post-infection ( total bacterial survival ) , or after a subsequent 2 h of gentamicin treatment , a standard method frequently used to assess intracellular bacterial survival due to its poor ability to penetrate mammalian cell membranes [34–38] . At 100 μg/mL , gentamicin effectively kills extracellular UPEC , as no detectable colonies were recovered from the media containing gentamicin after 2 h treatment [39] . We observed a marked reduction ( ~40% ) in UPEC CFU recovery from 5637 cells pre-treated with AKB-4924 at both time points , compared to mock ( DMSO-treated cells ( Fig 1C ) . Likewise , we found infection of UPEC UTI89 , a hyper-invasive strain , was also significantly attenuated in AKB-4924 treated uroepithelial cells ( S1 Fig ) . UPEC are known to trigger rapid death and extensive exfoliation of the uroepithelial layer of the bladder [40 , 41] . Through Live/Dead cell staining ( Fig 2A ) , we documented a clear reduction ( ~40% ) in UPEC-induced cell death and detachment in AKB-4924 pre-treated cells ( Fig 2B and 2C ) . UPEC induce degradation of paxillin , a focal adhesion molecule important for cell attachment [42] . Western blot revealed that paxillin levels were partially restored in AKB-4924 treated bladder cells infected with UPEC ( Fig 2D ) , likely contributing to preservation of cell integrity and surface attachment . Treatment of AKB-4924 did not alter paxillin level in uninfected cells ( S2B Fig ) . Similar to our in vitro findings , mice pre-treated with AKB-4924 showed less paxillin degradation compared to mock treated animals after UPEC CFT073 infection ( S2A Fig ) . UPEC cytotoxicity to uroepithelial cells is mediated through the mitogen-activated protein kinase ( MAPK ) -dependent pathway , particularly via phospho-p65 and phospho-p38 MAPK activation [39 , 43–45] . While AKB-4924 treatment did not reduce UPEC-mediated p65 phosphorylation , it partially blocked UPEC-mediated p38 MAPK activation ( Fig 2D ) . Neither phospho-p65 nor phospho-p38 was altered by AKB-4924 in uninfected cells ( S2B Fig ) . Our in vivo study also revealed that AKB-4924 treated bladders with UPEC infection had diminished p38 activation ( S2A Fig ) . The p38 MAPK plays a key role in initiating cell death pathways and triggering multiple downstream pro-inflammatory cytokine responses in epithelial cells [46 , 47] . UPEC-infected cells released significantly less pro-inflammatory cytokines known to be regulated by p38 MAPK , including IL-6 , IL-1x and IL-8 [47 , 48] , in the presence of AKB-4924 ( Fig 2E ) . In contrast , AKB-4924 did not alter the cytokine production profile in uninfected cells . To examine the potential utility of AKB-4924 in preventing UPEC infection in vivo , we used a well-described murine UTI model [49] . Each mouse received 0 . 2 mg of AKB-4924 via transurethral injection for 1 h prior to initiation of an 18–24 h UPEC infection . AKB-4924 treated animals had an approximately 2-fold increase in HIF-1α mRNA recovered from their bladders upon infection ( Fig 3A ) . In contrast , UPEC infected animals without AKB-4924 treatment showed no changes in HIF-1α mRNA levels ( S3A Fig ) . AKB-4924 also elevated productions of VEGF mRNA and protein in the bladders at different infection time points ( Fig 3B and S4A Fig ) . Pharmacological HIF-1α activation remarkably decreased bacterial burden in the urine , bladder and kidneys compared to vehicle treated mice 18–24 h post-infection ( Fig 3C ) . Almost 50% of mice assayed had urine titers under 104 CFU/mL , which is a commonly applied cut-off for resolution of experimental bacteriuria in mice [18 , 50] . Thus AKB-4924 protects against development of localized UPEC bladder infection . Through a time-course analysis , AKB-4924 was shown to impair UPEC colonization in the bladder from 4 h to 8 h post-infection . Mock-treated mice had significantly higher bacterial burdens at 8 h compared to 4 h , whereas the bacterial burden in 4924-treated mice remained similar in this interval ( S4C Fig ) . Interestingly , many mice challenged with UPEC also displayed high levels of CFU recovery in kidneys 18–24 h post-infection; AKB-4924 treatment reduced bacterial recovery from kidneys ( Fig 3C ) , indicating that AKB-4924 impedes ascending UTI . One of the key events during UPEC infection is intracellular invasion of uroepithelial cells , upon which time the bacteria rapidly develop into intracellular bacterial communities ( IBCs ) [12] . To determine whether AKB-4924 prevents UPEC invasion into the bladder tissue , we treated bladders ex vivo with gentamicin to kill extracellular UPEC to assess level of intracellular bacteria within the tissue . Consistent with the total CFU recovered from the bladder and the in vitro gentamicin protection assay ( Fig 1B ) , we found significantly less intracellular ( gentamicin-protected ) bacteria in the bladder of mice prophylactically treated with AKB-4924 at 18 h post-infection ( Fig 3C ) . As a control , we excluded the possibility of a direct bactericidal effect of AKB-4924 on UPEC , as the drug did not influence UPEC growth in mouse urine or tissue culture media ( RPMI supplemented with fetal bovine serum ) over 20 h incubation at 37°C ( S3B Fig ) . In addition to growth , we also tested whether AKB-4924 has any direct impact on UPEC fitness or flagella-driven motility . Assessed by a swimming motility assay using soft agar ( 0 . 25% LB agar ) , AKB-4924 did not alter UPEC motility through 18 h of growth ( S3C Fig ) . A genetic approach was used to verify the importance of HIF-1α in uroepithelial defense UPEC infection . In mice with a Cre recombinase-mediated , keratinocyte-specific inactivation of HIF-1α , ( Hif1αflox/flox/K14-Cre+ , hereafter known as HIF-1-/- ) [24] , we confirmed an 80% reduction in bladder HIF-1α mRNA by real-time quantitative PCR ( qPCR ) compared to wild-type ( WT ) littermate controls ( Fig 4A ) . Upon UPEC challenge , these HIF-1-/- mice were more susceptible to bladder infection than WT mice ( Fig 4B ) . AKB-4924 pre-treatment did not alter infection severity or bacterial counts in the bladders of HIF-1-/- mice , in contrast to the protective effects observed in AKB-4924 treated WT mice ( Fig 4C ) . This confirms that AKB-4924-mediated protection against UPEC infections requires HIF-1α , and emphasizes the importance of HIF-1α stabilization in attenuating UTI . Gross morphology showed markedly reduced hemorrhagic inflammation in WT mice treated with AKB-4924 compared to untreated WT mice or AKB-4924 treated HIF-1-/- mice ( Fig 4D ) . These results confirm that the AKB-4924 therapeutic effect occurs via HIF-1 boosting and not an off-target activity , and provide further evidence that HIF-1α plays a key role in mitigating the development of UTI . To examine the inflammatory profile of UPEC-infected bladders , we measured the levels of secreted IL-1β , IL-6 and KC ( keratinocyte-derived chemokine , a murine ortholog of the human neutrophil chemokine CXCL1 ) in mouse bladder homogenate following 1 h pre-treatment with AKB-4924 or vehicle control prior to infectious challenge . Consistent with our in vitro data ( Fig 1B and Fig 2E ) , we observed a decrease in pro-inflammatory cytokine production in AKB-4924 pre-treated bladders ( Fig 5A ) . Reduced IL-1β and IL-6 levels were also detected in kidneys of AKB-4924 treated mice ( Fig 5B ) . Concomitant to these reduced cytokine levels , we observed a significant decrease in myeloperoxidase ( MPO ) activity , a marker of neutrophil recruitment [51] , in AKB-4924 treated bladders compared to those treated with vehicle control ( Fig 5C ) . Histology section from infected bladders with vehicle treatment displayed hyperplasia and edema , characterized by increases in crypt length within the uroepithelial lining; similar observations were previously made in other UPEC infected murine bladders [13 , 14 , 52 , 53] . In contrast , AKB-4924 treated mice exhibited decreased hyperplasia in the bladder epithelial layer ( S5 Fig ) . We also noted a decrease in bacterial population in the luminal region of the AKB-4924 treated mice ( S5 Fig ) . These observations reflect our previous cytokine measurement studies . As UPEC CFT073 caused only mild to moderate bladder inflammation , we repeated infection using the hyper-virulent UPEC strain UTI89 to trigger higher grade inflammation and further analyze the effects of AKB-4924 treatment . In the hyper-virulent UTI89 infection , MPO localization was monitored by immunofluorescent microscopy using a labeled anti-MPO antibody , revealing a significant reduction in MPO localization in the uroepithelial lining of AKB-4924 treated bladders ( Fig 5D ) . MPO expression is nearly undetectable level in uninfected cells ( S6B Fig ) , confirming the elevated level of MPO expression is not a pre-existing inflammatory condition but occurs specifically in response to UPEC infection . Together , these results indicate that HIF-1α stabilization via AKB-4924 treatment significantly dampens UPEC-mediated inflammatory responses to the bladder . Since AKB-4924 does not display direct bactericidal activity against UPEC ( S3B Fig ) , we hypothesized that the drug induced production of endogenous antimicrobial compounds by bladder epithelial cells to promote bacterial clearance . Patients with bladder infections are known to have increased inducible nitric oxide synthase ( iNOS ) and its product nitric oxide ( NO ) , which can exert antimicrobial activity against susceptible pathogens [10] . Because HIF-1α is a transcriptional activator of iNOS [25 , 31 , 54] , we assessed the functional output of iNOS activity by measuring levels of nitrite ( NO2- ) , a stable NO oxidation product . Like previous studies that showed significant nitrite production in urinary pathogenic E . coli infected mouse bladders [55 , 56] , we found UPEC also significantly induces nitrite production in human uroepithelial cells 5637 ( Fig 6A , left panel ) . Nitrite induction was further augmented by AKB-4924 pre-incubation , but the drug did not produce an observable increase in nitrite levels in uninfected cells ( Fig 6A , left panel ) . UPEC can reduce nitrate to nitrite and also derive nitrite from iNOS-mediated NO production . We normalized the nitrite production level to total CFU per well ( 1 mL ) , highlighting that AKB-4924 double nitrite release from 5637 cells during infection ( Fig 6A , right panel ) . Increased nitrite production in AKB-4924 treated cells correlated to increased iNOS transcription ( S7A Fig ) . To further analyze NO protection in UTI , we performed infection in a set of WT ( iNOS +/+ ) , iNOS heterozygous ( iNOS +/- ) and knockout ( iNOS-/- ) mice . iNOS knockout mice were significantly more susceptible to UPEC infection than either WT or iNOS heterozygous mice ( Fig 6C ) . The increased bacterial load in iNOS +/- bladders corresponded with lower nitrite levels in these mice ( Fig 6B ) , indicating NO contributes to effective bacterial clearance in the bladder . When WT and iNOS +/- mice were treated with AKB-4924 prior to UPEC CFT073 infection , we found that lacking a fully functional host iNOS system , the protective effect of AKB-4924 was significantly diminished; moreover , the effect of AKB-4924 was abolished in iNOS +/- mice ( Fig 6D ) . Overall , our results demonstrate that the protection generated by HIF-1 boosting via AKB-4924 is partially dependent on NO production . HIF-1α transcriptionally upregulates the expression of the cationic antimicrobial peptides cathelicidin ( human = LL-37 , mouse = mCRAMP ) and/or β-defensin in neutrophils [25 , 29] , keratinocytes [24] or cornea [23] to impede various pathogens including GAS , MRSA and P . aeruginosa . Uroepithelial cells secrete abundant quantities of peptides cathelicidin and β-defensin 2 in response to UPEC infections [57–59] . Cathelicidin has been shown to be important for UPEC clearance , since mCRAMP-deficient mice suffer significantly higher bacterial burdens during UTI [58] . Likewise , bladder epithelial cells that lack human β-defensin 2 ( hBD2 ) are more susceptible to E . coli infection [60] . Based on these studies , we examined whether the reduced bacterial burden in AKB-4924-treated epithelial cells and bladders could be correlated with an increase in antimicrobial peptide production . Real-time qPCR revealed hBD2 was transcriptionally upregulated by two-fold in 5637 uroepithelial cells pre-exposed to AKB-4924 , whereas UPEC infection alone did not alter hBD2 mRNA levels ( Fig 6E ) . Interestingly , we found UPEC significantly suppressed LL-37 transcription in 5637 cells 2h post-infection , which mirrored results from Chromek and colleagues , where they demonstrated UPEC induced a rapid decrease in LL-37 transcript in uroepithelial cells after more than 2 h infection [58] . Remarkably , AKB-4924 restored LL-37 mRNA back to the pre-infection levels in cells 2 h post UPEC infection , a 2-fold increase compared to infected and vehicle-treated cells ( Fig 6E , right panel ) . In correlation with these in vitro data , we found mRNA expression of mBD2 and mouse cathelicidin-related antimicrobial peptide ( mCRAMP ) was significantly augmented in AKB-4924 treated mice compared to vehicle-treated mice 18 h post-infection by approximately 4-fold ( Fig 6F ) . Immunofluorescence studies in UPEC UTI89-infected mice revealed a clear increase in cathelicidin protein distribution in the bladder uroepithelial layer compared to uninfected mice ( Fig 6G ) . A slight increase in cathelicidin expression was observed deeper into the bladder epithelium with AKB-4924 pre-treatment ( Fig 6G ) . A similar increase in cathelicidin distribution was illustrated in a previous study in which vitamin D treatment was used to stimulate bladder expression of the peptide [57] . We confirmed our immunohistopathology results by western blot and densitometry , which showed that AKB-4924 pre-treatment increased cathelicidin protein levels by approximately 1 . 5-fold compared to vehicle controlled in UPEC CFT073 infected bladders ( Fig 6H ) . This result is consistent with our prior study showing reduced cathelicidin expression in HIF-1-/- mouse skin keratinocytes [24] , emphasizing the key role of HIF-1α in regulating cathelicidin production . When we treated UPEC with filter-sterilized supernatants prepared from uroepithelial cells infected with UPEC , significantly fewer bacteria were recovered from supernatants from 4924-treated cells vs . controls ( S7B Fig ) . Since AKB-4924 is not bactericidal ( S3B Fig ) , HIF-1α stabilization increases the release of an antimicrobial factor from uroepithelial cells that contributes to bacterial killing . Having demonstrated that AKB-4924 pretreatment suppresses the establishment of UPEC infections in vivo , we assessed the value of AKB-4924 as a potential therapeutic agent for animals with established UPEC infection . Mice were first infected with UPEC and then 6 h later administered with 0 . 2 mg AKB-4924 transurethrally; this time point was selected based on previous studies indicating that it corresponded to the point of highest bacterial recovery [49] , at which time the bacteria have attached and invaded epithelium to form early IBCs [61] . Mice treated with AKB-4924 showed more than a 10-fold reduction in UPEC colonization of the bladder ( Fig 7A ) ; a finding that coincided with an increase in CRAMP and hBD2 mRNA levels ( Fig 7B and 7C ) . This finding suggests that AKB-4924 treatment can impede UPEC even in the face of established bacterial invasion and formation of early IBC in the bladder epithelium . Together , our combined data indicate that by stabilizing HIF-1α , AKB-4924 enhances production of antimicrobial effectors in both prophylactic and therapeutic settings , promoting bacterial clearance , and suggesting HIF-1α boosting as a potential adjunctive therapeutic strategy in UTI management .
Upon microbial infection , HIF-1α stabilization is part of the general myeloid cell innate immune response to elicit host protection [21 , 22 , 62 , 63] . HIF-1 is an oxygen-inducible master transcriptional activator that regulates several targets important for innate immune functions . Low oxygen levels inhibit ubiquitylation-mediated degradation of HIF-1 , resulting in activation of HIF-1 transcriptional response . HIF-1 transcription is also regulated through NF-κB signaling [64] . Previous studies have suggested that patients suffering from UPEC display lower oxygen levels in their bladders [65 , 66] , a phenomenon that could in theory further stabilize HIF-1 . Nevertheless , we demonstrated that HIF-1α post-transcriptional regulation plays a key role in protecting against UPEC infection and injury to bladder epithelium in vitro and in vivo . AKB-4924 effectively increases HIF-1α stability in cultured human uroepithelial cell line 5637 , paralleling studies which showed the drug can induce HIF-1 expression in murine fibroblasts , human monocytes [29] and intestinal epithelial cells [30] . AKB-4924 increases HIF-1α expression in UPEC-infected human uroepithelial cells in vitro and murine bladders in vivo , which may be potentiated by a positive feedback loop since HIF-1α positively auto-regulates its own expression by binding to the HIF-1α promoter [67] . In addition to HIF-1α , we found that AKB-4924 also stabilizes HIF-2α ( S2C Fig ) . During AKB-4924 treatment ( stabilizing both HIF-1 isoforms ) , HIF-1-/- animals experienced a significantly higher bacterial burden than WT animals , suggesting that HIF-1 plays the predominant role in regulating UPEC clearance . Exposing human uroepithelial cells or mouse bladders to AKB-4924 prior to infection significantly blocked bacterial attachment and invasion , and dampened the host pro-inflammatory response and p38 MAPK activation triggered by UPEC . We postulate that the observed reduction in cytotoxicity and inflammation in AKB-4924 treated cells is due to partial blockade of UPEC-mediated p38 MAPK activation . Reduced inflammation in AKB-4924-treated bladders was reflected in significant reduction of proinflammatory cytokines and chemokines including IL-1β , IL-6 , IL-8 ( CXCL8 ) and KC ( CXCL1 ) , and diminished neutrophil recruitment . Our histology studies further indicate that AKB-4924 treatment reduces level of epithelial hyperplasia in the crypt of bladder , which reflects reduced chronic inflammation in the bladder [18 , 52 , 53] . These changes are likely reflective of the reduced bacterial burden . Our findings share similarities with recent studies in colitis and corneal infection models , where HIF-1α ultimately acted to suppress inflammation in the affected tissues . In the first study , Keely et al . showed up-regulation by AKB-4924 suppressed gastrointestinal tract inflammation in mice with chemically-induced colitis , with reduced IL-6 , IL-1β and TNF-α production in both colons and blood sera [30] . The second study , silencing of HIF-1α transcription in the cornea led to greater bacterial burden , increased cytokine production and more MPO activity reflective of neutrophil infiltration [23] . Whereas intact pro-inflammatory signaling and neutrophil recruitment are critical for innate defense against bacterial pathogens , exaggerated levels of cytokines and recruited leukocytes can lead to persistent inflammation and recurrent injury to bladder epithelium , which could ultimately risk permanent renal scarring and functional impairment [17 , 18 , 68 , 69] . Thus , for the benefit of the host , it is important that the immune response remains tightly regulated and restores to homeostasis following infection . UPEC attach to the surface of uroepithelial layers , where they rapidly colonize and invade epithelial cells , and can reside within these cells in vesicles or by escape into the cytoplasm to form IBC [12 , 42] . During the early phase of acute infections , UPEC induces a broad pro-inflammatory cytokine release in uroepithelial cells [6–9] . In this current study , we suspect that the reduction in inflammatory markers observed in the AKB-4924 treated uroepithelial cells and bladders is reflective of a diminished bacterial burden . Contributions of NO and AMPs to host defense against bladder and other infections have been well documented in the literature [10 , 23 , 55 , 56 , 58 , 70 , 71] . CRAMP-deficient animals experience increased susceptibility to UPEC UTI [58] , and the robust killing effect of NO has also been clearly demonstrated in UTIs by various E . coli strains , including UPEC 1177 , UPEC J96 , ATCC25922 , RK4353 , and CM120 [56 , 72–74] . More recently , β-defensins have also been shown to play an antimicrobial role in protecting host against UTIs [75 , 76] . We found a significant difference in bacterial recovery between iNOS+/- and iNOS-/- mice 24 h post-infection . Although one study observed no overall difference in E . coli strain 1177 colonization between WT and iNOS-/- mice over a course of 7 d , they showed a 3-log fold increase in the bladders of iNOS-/- mice compared to WT mice 6h post-infection [56] . The discrepancy between our studies could also be due to different E . coli strains used , since NO susceptibility can vary depending on bacteria’s ability to detoxify NO [73 , 77] , and could also due to different infection dosage , since higher dosage of infection does not necessarily enhance the infection severity ( a lower infection dosage was used in our study ) . In this study we found AKB-4924 significantly boosted levels of NO , cathelicidins ( human LL-37 and mouse CRAMP ) and hBD2 in both UPEC infected human uroepithelial cells and murine bladders , coincides with studies that showed HIF-1α positively regulates NO and AMPs to support clearance of bacterial infections in skin keratinocytes and leukocytes [24 , 25 , 31 , 78 , 79]; thus we suggest that HIF-1α serves a similar role in uroepithelial cells . Future studies examining samples from human UTI patients will be required to precisely ascertain the antimicrobial role of NO and its relevance to human disease , since sometimes NO responses could be exaggerated in mice . Together , our results introduced HIF-1α as a key regulator of antimicrobial effectors providing host protection against UTI . In summary , we have uncovered an interplay that exists between HIF-1α and the host innate immune response in the urinary tract . By depositing HIF-1α boosting agent directly into the bladder through a catheter , we could prevent infections and limit the risk of bladder and renal damage caused by acute inflammation . Since there are no known effective prophylactic agents available ( other than antibiotics ) to prevent UTIs [80] , this application could serve as a prophylaxis to benefit certain high risk UTI patients , although administration of AKB-4924 to the urinary tract in an ascending fashion would likely be restricted to patients with indwelling catheters or those receiving daily catheterization . Therefore , a future goal of HIF boosting drugs would be to develop oral formulations that could effectively distribute into the urinary tract to reach wider patient populations . HIF-1α activation occurs as a result of loss of vHL in certain forms of renal cell carcinoma , and HIF-1α controls VEGF and other angiogenic factors that are essential in support of tumor growth , which some considered potential risks of HIF-1 boosting therapy , however recent data indicates that activation of HIF-1 does not increase intestinal tumorigenesis [81] . Moreover , thorough Food and Drug Administration review has allayed these concerns , and allowed a number prolyl-hydroxlase inhibitor ( HIF-1 boosting ) drugs to enter human clinical trials for chronic administration in anemia patients , including FG-4592 ( FibroGen , Astellas and AstraZeneca , ClinicalTrials . gov #NCT01887600 , Phase III ) , AKB-6548 ( Akebia Therapeutics , #NCT01235936 , Phase II ) , and GSK1278863 ( Glaxo-SmithKline , #NCT01977573 , Phase II ) [82] . We conclude that AKB-4924 and similar HIF-1α boosting agents could merit further exploration as novel therapeutic tools in the prevention and treatment UPEC-related UTIs .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of the University of California , San Diego ( Animal Welfare Assurance Number: A3033-01 ) . All efforts were made to minimize suffering of animals employed in this study . WT C57BL/6 mice aged 6–7 weeks were purchased from Jackson laboratory . HIF-1α knockout targeted to keratinocyte inactivation of HIF-1α ( Hif1αflox/flox/K14-Cre+ ) and WT littermates from the same breeding pair were used at 6–7 weeks [24]; iNOS+/- ( heterozygous ) or iNOS-/- ( null ) C57BL/6 mice ( Jackson Laboratory , strain B6 . 129P2-NOS2tm1Lau/J ) littermates from the same breeding pairs were used at 7–10 weeks [83] . Mice were handled by approved protocols of the UCSD Animal Care Committee . Wild-type uropathogenic E . coli ( UPEC ) strain CFT073 ( O6:K2:H1; ATCC 700928 ) and UTI89 were grown for at least 20 h in standing culture to stationary phase at 37°C in Luria-Bertani ( LB ) broth prior to infection . Human uroepithelial cell line 5637 ( ATCC# HTB-9 ) were cultured in RPMI-1640 ( Invitrogen ) media supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) at 37°C in humidified air with 5% CO2 . AKB-4924 ( Aerpio Therapeutics , Cincinnati , OH ) was manufactured as previously described [29] . For in vivo experiments , each mouse was treated transurethrally with 0 . 2 mg of AKB-4924 prepared in 40% 2-hydroxylpropyl-β-cyclodextran in 50 mM aqueous citrate buffer at pH 5 . For in vitro experiments , cells in RPMI-1640 supplemented with 10% FBS were treated for 2 h with 20 M of AKB-4924 resuspended in DMSO pH = 4 . 2–4 . 4 unless otherwise stated . Human uroepithelial cells 5637 were seeded at ~2 x 105 cells/mL in 24-well plates a day prior to AKB-4924 pre-treatment . Confluent 5637 cells were treated with 0 . 4% of DMSO ( control ) or 20 μM AKB-4924 prepared in DMSO ( Aerpio Therapeutics ) for 2–4 h and were infected with UPEC from a fresh overnight standing culture ( OD ~0 . 5 ) at a multiplicity of infection ( MOI ) ~20 . Plates were centrifuged at 650 x g for 5 min to facilitate bacterial contact with the host cell monolayer . The bacteria were then allowed to establish attachment on monolayers for 2 h in the presence of AKB-4924 or DMSO . After 2 h infection , cells were washed 3x with PBS lysed by adding 100 μL of 0 . 05% trypin ( 1 min ) followed by 900 μL of 0 . 025% triton X-100 in PBS to assess total surviving bacteria by CFU enumeration . To assess the level of internalized bacteria , cells were washed with fresh media 2 h post-infection and treated with 100 μg/mL gentamicin to kill extracellular bacteria for 2 h before CFU recovery as above [39] . An established mouse UTI protocol was used as previously described [49] . Urine is voided from the bladder of all animals prior to transurethral treatment with 2 mg/ml of AKB-4924 in 100 μL ( administered in ~4 sec ) for 1 h followed by infection with 50 μL of UPEC CFT073 ( administered in ~2–3 sec ) at ~2–5 x 108 colony forming unit ( CFU ) per mouse or UTI89 at ~6 x 107 CFU per mouse . Alternatively , mice were treated with AKB-4924 6h post-infection . Transurethral infection was achieved by inserting an UV-sterilized polyethylene tube ( inner dimension 0 . 28 mm , outer dimension 0 . 61 mm , Catalogue number 598321 Harvard Apparatus ) attached to a 30g hypodermic needle into the urethra . At ~20 h post-infection , urine samples were collected from each mouse . Bladders and both kidneys were removed and homogenized using MagNa Lyser ( Roche ) for CFU enumeration . For an ex vivo gentamicin protection assay , bladders were removed from mice ~18 h post-infection and washed three times with PBS followed by treatment with 100 μg/mL of gentamicin for 2 h to kill extracellular bacteria . Bladders were washed then three times , homogenization in 1 mL PBS + 0 . 025% triton , and serially diluted for CFU enumeration . All infection is performed using UPEC CFT073 unless otherwise stated . Confluent 5637 monolayers were grown on sterile no . 1 round coverslips ( Thermo Scientific ) in 24-well or 6-well plates . Cells were infected with WT UPEC at a MOI of 5–20 for 2 h . Cells were washed with PBS and treated with the viability assay mixture from the LIVE/DEAD Viability/Cytotoxicity Kit for mammalian cells ( Molecular Probes , Invitrogen ) for 30 min at 37°C , then mounted on glass slides for visualization and imaging using an Olympus BX51 fluorescent microscope fitted with appropriate filters as described previously [39] . Tissues for H&E staining were submerged in formalin overnight at 4°C and transferred to the Histopathology Core facility for processing . For Immunohistochemistry , immediately after euthanasia , the entire bladder was removed from the mouse and submerged in pre-warmed 3% paraformaldehyde for 3 h . Tissues were washed 3 times , 10 min each in 10 mL PBS and immediately frozen down in OCT compound with liquid nitrogen for 5 μM sectioning by the UCSD Mouse Histopathology Core facility ( N . Varki , Director ) . Tissue sections were immediately submerged into 0 . 2% Triton X-100 in PBS for 5 min and washed extensively with PBS before treatment with 5% normal goat serum ( NGS ) in TPBS/BSA for 20 min at room temperature . Tissues were incubated with rabbit anti-cathelicidin ( Catalog No . NB100-98689 , Novus Biological ) at 1:200 dilution , rabbit anti-MPO polyclonal antibody ( Catalogue No . RB373A , Thermo Scientific ) at 1:200 dilution or rabbit ( DA1E ) IgG monoclonal isotype control antibody ( Catalog No . 3900 , Cell Signaling Technology ) at 1:200 in 1% NGS in TPBS/BSA overnight at 4°C . On the next day , coverslips were washed with TPBS/BSA 3 times and incubated with a goat anti-rabbit Alexa 594 antibody ( Molecular probes ) for 1 h at 37°C before they were washed and mounted with ProLong Gold anti-fade reagent with DAPI ( Catalogue No . P-36931 Molecular Probes ) . Cells were visualized using an Olympus BX51 fluorescent microscope fitted with appropriate filters . The level of neutrophil migration into mouse bladders was determined by myeloperoxidase ( MPO ) assay . MPO activity released in bladder homogenates was quantified by the MPO colorimetric activity assay ( Catalogue No . MAK068 , Sigma Aldrich ) as previously described [33] . Greiss reagent ( Catalog No . G2930 , Promega ) was used to detect nitrite production in 5637 cells according to the manufacturer’s protocol . Confluent 5637 cells in T75 flask were washed with PBS once before treated with 20 mM 4924 or DMSO in 25 mL fresh media for 4 h at 37°C . Cells were harvested by 5 min treatment with 0 . 05% tryspin and 15 mL chilled media followed by centrifugation at 400 x g for 10 min at 4°C . Pellets were re-suspended in 1 mL PBS at 4°C and centrifuged for 5 min . The supernatant was collected for subsequent nuclear extraction using the NE-PER Nuclear and cytoplasmic extraction kit following manufacturer’s protocol ( Catalogue No . 78833 , Thermo Scientific ) . Whole cell lysates from 5637 were prepared for western blot as previously described [39] . The following antibodies were used: rabbit anti-HIF-1α antibody , rabbit anti-HIF-2α ( NB100-122 ) ( Novus Biological ) , mouse anti-P84 antibody ( Catalog No . GTX70220 ) ( GeneTex ) , rabbit anti-paxillin , rabbit anti-phospho-p65 ( Ser536 ) monoclonal antibody ( Catalogue No . 3033 ) and rabbit anti-phospho-p38 ( Thr180/Tyr182 ) polyclonal antibody ( Catalogue No . 92122 ) ( Cell Signaling Technologies ) . Bladder homogenates were prepared by lysing a bladder in 500 μL RIPA lysis buffer with beads using the MegaLyser ( Roche ) . Cathelicidin detection was made using the same anti-cathelicidin antibody as immunolocalization . Anti-mouse β-actin monoclonal antibody clone AC-74 ( Catalogue No . A5316 , Sigma-Aldrich ) was used as a loading control . Band intensity was measured using Image J software . Concentrations of cytokines were measured in supernatant from infected 5637 cells ( 2h post-infection ) or homogenized bladder or kidneys 1 d post infection in mice ( n > 3 ) . Levels of mouse or human ( 5637 cells ) cytokines IL-6 , and IL-1β were detected using ELISA kits following the manufacturer’s protocol ( R&D systems ) . Assays were performed in triplicates or quadruplicates for each experiment . For tissue cultures , 500 μL of TRIZOL reagent was added each well whereas for bladder and kidney isolates , 1 mL of TRIZOL was added . After RNA isolates were re-suspended in RNAase/DNase- free water and TURBO DNase ( Ambion , Invitrogen ) was added to eliminate potential DNase contamination in the RNA prep . To synthesize cDNA , approximately 100 ng of total RNA was used for iScript cDNA Synthesis kit ( Bio-Rad ) . Approximately 1 ng of cDNA was used in triplicates or quadruplicates for real-time qPCR using KAPA SYBR qPCR 2x master mix ( KAPA Biosystem , Catalog# KM4101 ) . The reaction was performed using the Biorad CFX96 Real-time C1000 Thermocycler . Primers were used at a final concentration of 200nmol . Primer sequences used in this study are listed in S1 Table . Mouse β-2-microglobulin ( β2M ) and human β-actin were used as control house keeping genes . Relative transcript level was normalized to endogenous house keeping genes using the 2-ΔΔCt method [84] . Overnight cultures of UPEC CFT073 was grown in LB and inoculated on swimming LB plates with 0 . 25% of agar supplemented with vehicle or 0 . 2mg/mL AKB-4924 ( used for in vivo study ) , 0 . 4% DMSO or 20 M AKB-4924 ( used for in vitro study ) . Bacteria were grown at 37°C over 18 h and swimming diameter was measured ( n = 3 ) . All experiments were performed in triplicates or quadruplicate , and repeated in at least two independent experiments . All data are presented as mean and error represents S . E . M . ( n > 3 ) from multiple independent experiments . Statistical analysis is performed using One-way ANOVA or Student’s unpaired two-tailed t-test ( Graph Pad Prism , version 5 . 03 ) . *P < 0 . 05 , **P < 0 . 01 and ***P < 0 . 001 represent statistical significance; P > 0 . 05 or n . s . is non significant . | Urinary tract infection ( UTI ) , commonly caused by uropathogenic E . coli ( UPEC ) , affects more than 150 million people worldwide , resulting in 14 million hospital visits per year and an estimated total cost of 6 billion dollars in direct health care . Due to the high prevalence of UTI and rapid emergence of antibiotic-resistant bacteria , new effective strategies to prevent and treat UTI are urgently needed . Here , we describe a global regulatory role of transcription factor hypoxia-inducible factor-1 ( HIF-1 ) in innate antimicrobial defense against UPEC . HIF-1 stabilization reduces UPEC attachment to and invasion of uroepithelial cells , and protects bladders from UPEC-mediated cytotoxicity in vivo . In the murine UTI model , we found normal bladder HIF-1 expression is required for efficient UPEC clearance , since HIF-1-deficient mice suffer more severe infection than normal mice . Further studies showed that key elements of host protection provided by HIF-1 regulation are uroepithelial cell nitric oxide and antimicrobial peptide production . This study provides valuable insight into the importance of HIF-1 in supporting host immunity during UTI and its potential as a therapeutic target . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Role of Hypoxia Inducible Factor-1α (HIF-1α) in Innate Defense against Uropathogenic Escherichia coli Infection |
A scientific ontology is a formal representation of knowledge within a domain , typically including central concepts , their properties , and relations . With the rise of computers and high-throughput data collection , ontologies have become essential to data mining and sharing across communities in the biomedical sciences . Powerful approaches exist for testing the internal consistency of an ontology , but not for assessing the fidelity of its domain representation . We introduce a family of metrics that describe the breadth and depth with which an ontology represents its knowledge domain . We then test these metrics using ( 1 ) four of the most common medical ontologies with respect to a corpus of medical documents and ( 2 ) seven of the most popular English thesauri with respect to three corpora that sample language from medicine , news , and novels . Here we show that our approach captures the quality of ontological representation and guides efforts to narrow the breach between ontology and collective discourse within a domain . Our results also demonstrate key features of medical ontologies , English thesauri , and discourse from different domains . Medical ontologies have a small intersection , as do English thesauri . Moreover , dialects characteristic of distinct domains vary strikingly as many of the same words are used quite differently in medicine , news , and novels . As ontologies are intended to mirror the state of knowledge , our methods to tighten the fit between ontology and domain will increase their relevance for new areas of biomedical science and improve the accuracy and power of inferences computed across them .
The word ontology historically represented the product of one person's philosophical inquiry into the structure of the real world: What entities exist ? What are their properties ? How are they grouped and hierarchically related ? While this original definition still holds in philosophy , the computational interpretation of an ontology is a data structure typically produced by a community of researchers through a procedure that resembles the work of a standards-setting committee or a business negotiation ( L . Hunter , 2010 , personal communication ) . To agree on the meaning of shared symbols , the process involves careful utility-oriented design . The collective ontologies that result are intended to be used as practical tools , such as to support the systematic annotation of biomedical data by a large number of researchers . A standard domain-specific ontology used in the sciences today includes a set of concepts representing external entities , a set of relations , typically defined as the predicates of statements linking two concepts ( such as cat is-an animal , cat has-a tail ) , and taxonomy or hierarchy defined over concepts , comprised by the union of relations . An ontology may also explicitly represent a set of properties associated with each concept and rules for these properties to be inherited from parent to child concept . Furthermore , formal ontologies sometimes incorporate explicit axioms or logical constraints that must hold in logical reasoning over ontology objects . In practice , what different research groups mean by the term ontology can range from unstructured terminologies , to sets of concepts and relations without complete connection into a hierarchy , to taxonomies , to consistent , formal ontologies with defined properties and logical constraints . An ontology developed by group represents a glimpse into the specific worldviews held within that group and its broader domain . By the same logic , we can consider the union of all published articles produced by a scientific community as a much more complete sample of scientific worldviews . While a research team that writes a joint paper agrees on its topic-specific worldview to some extent , its collective domain ontology is neither explicitly defined , nor free from redundancy and contradiction . Insofar as scientists communicate with each other and respond to prior published research , however , these worldviews spread and achieve substantial continuity and homogeneity [11] . A large collection of scientific documents therefore represents a mixture of partially consistent scientific worldviews . This picture is necessarily complicated by the flexibility and imprecision of natural language . Even when scientists agree on specific concepts and relations , their corresponding expressions often differ , as the same meaning can be expressed in many ways . Nevertheless , if we accept that the published scientific record constitutes the best available trace of collective scientific worldviews , we arrive at the following conclusion: Insofar as an ontology is intended to represent knowledge within a scientific domain , it should correspond with the scientific record . Moreover , an ontology would practically benefit from evaluation and improvement based on its match with a corpus of scientific prose that represents the distribution of its ( potential ) users' worldviews . Previously proposed metrics for ontology evaluation can be divided into four broad categories: Measures of an ontology's ( 1 ) internal consistency ( 2 ) usability ( or task-based performance ) , ( 3 ) comparison with other ontologies and ( 4 ) match to reality . While this review is necessarily abbreviated , we highlight the most significant approaches to ontology evaluation . Metrics of an ontology's internal consistency are nicely reviewed by Yu and colleagues [12] . They especially highlight: clarity , coherence , extendibility , minimal ontological commitment , and minimal encoding bias [4]; competency [13]; consistency , completeness , conciseness , expandability , and sensitiveness [14] . The names of these metrics suggest their purposes . For example , conciseness measures how many unique concepts and relations in an ontology have multiple names . Consistency quantifies the frequency with which an ontology includes concepts that share subconcepts and the number of circularity errors . Measurements of an ontology's usability [15]–[17] build on empirical tools from cognitive science that assess the ease with which ontologies can be understood and deployed in specific tasks [18] . Results from such studies provide concrete suggestions for improving individual ontologies , but they are also sometimes used to compare competing ontologies . For example , Gangemi and colleagues [19] described a number of usability-profiling measures , such as presence , amount , completeness , and reliability , that assess the degree to which parts of an ontology are updated by ontologists [19] . The authors also discuss an ontology's “cognitive ergonomics”: an ideal ontology should be easily understood , manipulated , and exploited by its intended users . Approaches to ontology comparison typically involve the 1 ) direct matching of ontology concepts and 2 ) the hierarchical arrangement of those concepts , often between an ontology computationally extracted and constructed from text and a reference or “gold standard” ontology built by experts . Concept comparison draws on the information retrieval measures of precision and recall [12] , [20] , [21] ( sometimes called term [22] or lexical precision and recall [22]; see Materials and Methods section below for precise definitions of precision and recall ) . Matching ontology terms , however , raises challenging questions about the ambiguity of natural language and the imperfect relationship between terms and the concepts that underlie them . Some ignore these challenges by simply assessing precision and recall on the perfect match between terms . Others deploy string similarity techniques like stemming or edit distance to establish a fuzzy match between similar ontology terms [23] , [24] . The second aspect of ontology matching involves a wide variety of structural comparisons . One approach is to measure the Taxonomic Overlap , or intersection between sets of super- and subconcepts associated with a concept shared in both ontologies , then averaged across all concepts to create a global measure [23]–[25] . Another uses these super and subconcept sets to construct asymmetric taxonomic precision and recall measures [26] , closely related to hierarchical precision and recall [27] , [28] . A similar approach creates an augmented precision and recall based on the shortest path between concepts [29] or other types of paths and a branching factor [30] . An alternate approach is the OntoRand index that uses a clustering logic to compare concept hierarchies containing shared concepts [31] . The relative closeness of concepts is assessed based on common ancestors or path distance , and then hierarchies are partitioned and concept partitions are compared . Approaches for matching an ontology to reality are more diverse and currently depend heavily on expert participation [12] . For example , Missikoff and colleagues [32] suggested that an ontology's match to reality be evaluated by measuring each ontology concept's “frequency of use” by experts in the community . Missikoff and colleagues' ultimate goal was to converge to a consensus ontology negotiated among virtual users via a web-interface . Smith [33] recommended an approach to ontology evolution which rests on explicitly aligning ontology terms to unique entities in the world studied by scientists . Ontology developers would then be required to employ a process of manual tracking , whereby new discoveries about tracked entities would guide corresponding changes to the ontology . In a related effort , Ceusters and Smith suggested studying the evolution of ontologies over time [34]: they defined an ontology benchmarking calculus that follows temporal changes in the ontology as concepts are added , dropped and re-defined . A converse approach to matching ontologies with domain knowledge appears in work that attempts to learn ontologies automatically ( or with moderate input from experts ) from a collection of documents [35]–[38] using machine learning and natural language processing . The best results ( F-measure around 0 . 3 ) indicate that the problem is extremely difficult . Brewster and colleagues [36] , [39] proposed ( but did not implement ) matching concepts of a deterministic ontology to a corpus by maximizing the posterior probability of the ontology given the corpus [36] , [39] . In this framework , alternative ontologies can be compared in terms of the posterior probability conditioned on the same corpus . Their central idea , which shares our purpose but diverges in detail , is that “the ontology can be penalized for terms present in the corpus and absent in ontology , and for terms present in the ontology but absent in the corpus” ( see also [19] ) . Each of these approaches to mapping ontologies to text face formidable challenges associated with the ambiguity of natural language . These include synonymy or multiple phrases with the same meaning; polysemy or identical expressions with different meanings; and other disjunctions between the structure of linguistic symbols and their conceptual referents . In summary , among the several approaches developed to evaluate an ontology's consistency , usability , comparison and match to reality , metrics that evaluate consistency are the most mature among the four and have inspired a number of practical applications [40]–[42] . The approach that we propose and implement here belongs to the less developed areas of matching ontologies to each other and to discourse in the world . When considering approaches that compare ontologies to each other and to discourse , metrics comparing ontologies to one another jump from the comparison of individual concepts to the comparison of entire concept hierarchies without considering intermediate concept-to-concept relationships . This is notable because discourse typically only expresses concepts and concept relationships , and so the measures we develop will focus on these two levels in mapping ontologies to text . Our purpose here is to formally define measures of an ontology's fit with respect to published knowledge . By doing this we attempt to move beyond the tradition of comparing ontologies by size and relying on expert intuitions . Our goal is to make the evaluation of an ontology computable and to capture both the breadth and depth of its domain representation—its conceptual coverage and the parsimony or efficiency of that coverage . This will allow us to compare and improve ontologies as knowledge representations . To test our approach , we initially analyzed four of the most commonly used medical ontologies against a large corpus of medical abstracts . To facilitate testing multiple ontologies in reference to multiple domains we also analyzed seven synonym dictionaries or thesauri—legitimate if unusual ontologies [43]—and compared their fit to three distinctive corpora: medical abstracts , news articles , and 19-century novels in English . Medical ontologies have become prominent in recent years , not only for medical researchers but also physicians , hospitals and insurance companies . Medical ontologies link disease concepts and properties together in a coherent system and are used to index the biomedical literature , classify patient disease , and facilitate the standardization of hospital records and the analysis of health risks and benefits . Terminologies and taxonomies characterized by hierarchical inclusion of one or a few relationship types ( e . g . , disease_conceptx is-a disease_concepty ) are often considered lightweight ontologies and are the most commonly used in medicine [44] , [45] . Heavyweight ontologies capture a broader range of biomedical connections and contain formal axioms and constraints to characterize entities and relationships distinctive to the domain . These are becoming more popular in biomedical research , including the Foundational Model of Anatomy [46] with its diverse physical relations between anatomical components . The first , widely used medical ontology was Jacques Bertillon's taxonomic Classification of Causes of Death , adopted in 1893 by the International Statistical Institute to track disease for public health purposes [47] . Five years later , at a meeting of the American Public Health Association in Ottawa , the Bertillon Classification was recommended for use by registrars throughout North America . It was simultaneously adopted by several Western European and South American countries and updated every ten years . In the wake of Bertillon's death in 1922 , the Statistics Institute and the health section of the League of Nations drafted proposals for new versions and the ontology was renamed the International List of Causes of Death ( ICD ) . In 1938 the ICD widened from mortality to morbidity [48] and was eventually taken up by hospitals and insurance companies for billing purposes . At roughly the same time , other ontologies emerged , including the Quarterly Cumulative Index Medicus Subject Headings , which eventually gave rise to the Medical Subject Headings ( MeSH ) that the NIH's National Library of Medicine uses to annotate biomedical research literature [49] , [50] . By 1986 several medical ontologies were in wide use and the National Library of Medicine began the Unified Medical Language System ( UMLS ) project in order to link many of them to facilitate information retrieval and integrative analysis [51] . By far the most frequently cited ontology today in biomedicine is the Gene Ontology ( GO ) , a structurally lightweight taxonomy begun in 1998 that now comprises over 22 , 000 entities biologists use to characterize gene products [52] . We propose to further test and evaluate our ontology metrics using the fit between a synonym dictionary or thesaurus and a corpus . A thesaurus is a set of words ( concepts ) connected by synonymy and occasionally antonymy . Because synonymy constitutes an is-equivalent-to relationship ( i . e . , wordx is-equivalent-to wordy ) , thesauri can be viewed as ontologies , albeit rudimentary ones . Moreover , because a given thesaurus is intended to describe the substitution of words in a domain of language , the relationship between a thesaurus and a corpus provides a powerful model for developing and testing general measures of the fit between ontology and knowledge domain . Most useful for our purposes , the balance between theoretical coverage and parsimony is captured with the thesauri model: A bloated 100 , 000 word thesaurus is clearly not superior to one with 20 , 000 entries efficiently tuned to its domain . A writer using the larger thesaurus would not only be inconvenienced by needing to leaf through more irrelevant headwords ( the word headings followed by lists of synonyms ) , but be challenged by needing to avoid inappropriate synonyms . Synonymy is transitive but not necessarily symmetric – the headword is sometimes more general than its substitute . Occasionally thesauri also include antonyms , i . e . , is-the-opposite-of , but fewer words have antonyms and for those that do , antonyms listed are far fewer than synonyms . A typical thesaurus differs from a typical scientific ontology . While ontologies often include many types of relations , thesauri contain only one or two . Thesauri capture the natural diversity of concepts but are not optimized for non-redundancy and frequently contain cycles . Any two exchangeable words , each the other's synonym , constitute a cycle . As such , thesauri are not consistent , rational structures across which strict , logical inference is possible . They instead represent a wide sample of conflicting linguistic choices that represent a combination of historical association and neural predisposition . Despite these differences , we believe thesauri are insightful models of modern , domain-specific ontologies . Working with thesauri also contributes practically to evaluating the match between ontologies and discourse . Because all of our measures depend on mapping concepts from ontology to text , assessment of the match between thesaurus and text can directly improve our identification of ontology concepts via synonymy .
Our proposed approach to benchmarking an ontology X with respect to a reference corpus T is outlined in Figure 1 . The essence of the approach requires mapping concepts and relations of the test ontology to their mentions in the corpus – a task as important as it is difficult [53] . Given this mapping , we show how to compute ontology-specific metrics , Breadth and Depth , defined at three levels of granularity ( see Materials and Methods ) . We also define another important concept – the perfect ontology with respect to corpus T . This ideal ontology represents all concepts and relations mentioned in T and can be directly compared to X . If corpus T is sufficiently large , the perfect ontology is much larger than the test ontology X . This allows us to identify a subset of the perfect ontology that constitutes the fittest ontology of the same size as test ontology X –the one with maximum Breadth and Depth . Finally , given knowledge about the fittest ontology of fixed size and metrics for the test ontology X , we can compute loss metrics , indicating how much ontology X can be improved in terms of its fit to the corpus . All definitions are provided in the Materials and Methods section . To demonstrate our approach to the comparison of biomedical ontologies , we identified concepts associated with disease phenotypes and relations in four medical ontologies: ICD9-CM [48] , [54] , CCPSS [55] , SNOMED CT [56] and MeSH ( see Table 1 and Figure 2 ) . Comparing each medical ontology concept-by-concept ( as assessed with UMLS MetaMap—see Materials and Methods ) , we found that despite a reasonable overlap in biomedical terms and concepts , different ontologies intersect little in their relations ( see Figure 2 A and B ) . This suggests that each ontology covers only a small subset of the full range of possible human disease concepts and circumstances . This likely results from the different ways in which each ontology is used in biomedicine . To evaluate the fit between an ontology and a corpus , we first estimated the frequency of ontology-specific concepts and relations in the corpus . We mapped ontology concepts to the biomedical literature and then estimated their frequency using MetaMap , which draws on a variety of natural language processing techniques , including tokenization , part-of-speech tagging , shallow parsing and word-sense disambiguation [57] . We then estimated the frequency of concept relations in the literature ( see Materials and Methods ) . We parameterized these relation frequencies as the probability that two concepts co-occur within a statement in our medical corpus ( see Table 2 , Materials and Methods ) . Our measures of ontology representation build on established metrics from information retrieval ( IR ) , which have been previously used in ontology comparison . IR tallies the correspondence between a user's query and relevant documents in a collection: When the subset of relevant documents in a collection is known , one can compute IR metrics such as recall , precision and their harmonic mean , the F-measure , that capture the quality of a query in context ( see Materials and Methods ) . We compute these measures as first-order comparisons between ontologies in terms of whether concept-concept pairs “retrieve” contents from the corpus . The major rift between IR metrics and the nature of ontologies lies in the binary character of IR definitions: IR measures weight all relations in an ontology equally , but concepts and relations from an ontology vary widely in their frequency of usage within the underlying domain . Further , unlike IR documents retrieved from a query , concepts and relations present in an ontology but not a corpora should not be considered “false positives” or nonexistent in scientific discourse . Unless the ontology contains explicit errors , it is reasonable to assume that by expanding the corpus , one could eventually account for every ontology relation . Formulated differently , we cannot justifiably classify any ontology relation as false , but only improbable . This logic recommends we avoid IR measures that rely on false-positives ( e . g . , precision ) and augment the remaining metrics to model theoretical coverage and parsimony as functions of concept and relation importance rather than mere existence in the domain of interest . To do this , we first define the complete ontology that incorporates every concept and relation encountered in a corpus . In our implementation , we approximate this with all of the concepts and relations that appear in the corpus and are identified by UMLS MetaMap with the semantic type “disease or syndrome . ” We then define two measures , breadth and depth , to describe the fit between an ontology and a corpus . Breadth2 ( see Materials and Methods for definition of several versions of Breadth and Depth ) is a generalization of recall that substitutes true-positives and false-negatives with real-valued weights corresponding to the frequency of concepts and the probability of relations in text . Depth2 normalizes breadth by the number of relations in the ontology ( see Materials and Methods ) and so captures the average probability mass for each ontology relation in the corpus . Large ontologies tend to have better breadth of coverage relative to a corpus , but not necessarily more depth: They may be padded with rare concepts lowering their corpus fit compared with small , efficient ontologies containing only the most frequent ones . Breadth and depth allow us to compare ontologies of different size , but do not account for the fact that as ontologies grow , each incremental concept and relation necessarily accounts for less of the usage probability in a corpus . To address this challenge , we define the fittest ontology of fixed size ( with a predetermined number of relations ) such that depth is maximized over all possible concepts and relations . Furthermore , for an arbitrary ontology we can compute its depth loss relative to the fittest ontology of same size ( see Materials and Methods ) . This approach allows us to more powerfully control for size in comparing ontologies . Our analysis of the disease-relevant subsets of four medical ontologies indicates that CCPSS , despite having the smallest number of concepts and a moderate number of relations , performs comparably or better with respect to our clinical corpus than its larger competitors . When we consider concepts and relations jointly ( see Table 3 ) , CCPSS outperforms the three other terminologies in terms of Breadth2 and Relative Depth2 , while being second only to MeSH in Depth2 . ICD9-CM and SNOMED rank last in Breadth2 and Depth2 , respectively . When only concepts ( but not relations ) are considered ( Table 3 ) , SNOMED CT has the greatest Breadth1 and Relative Depth1 but the worst Depth1 , whereas MeSH and CCPSS lead in terms of Depth1 . It is striking that the relatively small CCPSS matches clinical text equally or better than the three other ontologies . Table 3 also indicates that Depth2 Loss is smallest for the largest ontology , SNOMED CT and that CCPSS is next . Given its small size , CCPSS is still less likely to miss an important disease relation than MeSH or ICD9-CM . ICD9-CM , with the highest Relative Depth1 , 2 Loss , would benefit most by substituting its lowest probability concepts with the highest probability ones missed . In order to demonstrate the power of our metrics to capture different dimensions of the fit between ontology and knowledge domain , we compared 7 of the most common English thesauri ( see Table 1 and Figure 2 ) against three corpora that sampled published text from the domains of medicine , news and novels ( see Table 2 ) . Our thesauri included ( 1 ) The Synonym Finder , ( 2 ) Webster's New World Roget's A–Z Thesaurus , ( 3 ) 21st Century Synonym and Antonym Finder , ( 4 ) The Oxford Dictionary of Synonyms and Antonyms , ( 5 ) A Dictionary of Synonyms and Antonyms , ( 6 ) Scholastic Dictionary of Synonym , Antonyms and Homonyms , and ( 7 ) WordNet ( see Materials and Methods ) . While comparing multiple thesauri word-by-word , we found a pattern similar to our medical ontologies . Despite a larger overlap in headwords than medical ontology concepts , different dictionaries intersect little in their relations . ( A headword in a thesaurus is a word or phrase appearing as the heading of a list of synonyms and antonyms . Not every word or phrase that is listed as a synonym in a thesaurus also occurs as a separate headword . ) On average , only one relation per headword is found in all three of the largest dictionaries ( see Figures 2 C and D ) . This trend persists as we consider a longer list of thesauri ( see Table 2 in Text S1 ) and indicates that any single dictionary covers only a small portion of synonyms used in the body of English . But some dictionaries are better than others . To evaluate the fit between thesaurus and corpus , we estimated the frequencies of thesauri headwords and synonyms in the corpus . We assessed headword frequency as we did with medical ontology concepts . In the case of synonymy relations , we parameterize the synonym frequencies as the probability that a headword is substituted with each of its synonyms within a specific four-word context ( see Materials and Methods ) . While thesauri typically aim to capture universal properties of language , corpora can be surprisingly dissimilar and sometimes disjoint in their use of words and synonym substitutions . Figures 3 and 4 visualize ten words whose synonym substitution probabilities are most unlike one another across the medicine , news and novels corpora . Some words carry a different semantic sense in each corpus ( e . g . , cat as feline versus CT scan versus Caterpillar construction equipment ) , while other words have very different distributions of common senses . It is illuminating to consider the dominant substitutions for the three corpora: The noun insult translates most frequently to injury in Medicine , slur in News , and shame in Novels; the verb degrade to impair , demean , and depress in the same respective corpora ( see Figures 3 and 4 ) ; the adjective futile to small , fruitless and vain . In some contexts words are used literally and consistently , while in others , metaphorically and widely varying . The meaning of the noun headache in our medical corpus is always literal: the closest synonyms here are migraine and neuralgia – with no other synonyms used . In novels and news the predominant meaning of headache is metaphorical . Novels are replete with headache's synonym mess , a disordered and problematic situation ( i . e . , headache-inducing ) . The news corpus also predominantly uses headache to mean problem , but the most frequent synonyms are more precise and literal ( problem , concern , worry , trouble ) . The metaphorical mess and hassle are also present , but at far lower frequencies than in novels . The verb stretch is treated as equivalent to develop , increase , prolong , and enlarge in the medical corpus . In novels it means open , spread , and draw . The news corpus hosts dozens of distinct synonyms for stretch , the most frequent three being extend , widen , and sprawl . Figure 5 , a–i and table 2 in the Supplement compare all metrics discussed for all seven thesauri and three corpora . From Figure 5 d and g , we observe that our importance-based breadth corresponds to counts-based recall ( a ) . The correspondence is not perfect , however: Oxford and WordNet have greater breadth than 21st Century , but this is reversed in recall . On the other hand , larger thesauri tend to lead in both recall and breadth , but small thesauri excel in precision and depth , as shown in Figure 5 e and h . The rankings of depth across all seven thesauri on three corpora , however , are very different from those of precision , which suggests that depth captures a different internal characteristic of ontology . For fixed precision and recall , we can define multiple equal-sized corpus-matched ontologies with widely varying depth and breadth by sampling from the complete ontology . The converse , however , is not true: Our breadth and depth metrics uniquely define an ontology's precision and recall . Figure 5 f and i indicate that depth loss is negatively correlated with the size of our seven thesauri ( see Discussion ) . This is likely because a large thesaurus nearly exhausts the common relations in all domains by including synonyms that are rare in one context but common in another . Small dictionaries must focus . Unless explicitly tuned to a domain , they are more likely to miss important words in it . Finally , we can compare corpora to each other with respect to all thesauri . As clearly shown in Figure 5 , our three corpora map onto the seven thesauri non-uniformly . Precision , for example , is significantly lower across all thesauri for the medical corpus than for news or novels . This is likely due to the specialized and precise medical sublanguage , which renders a large portion of common synonyms irrelevant .
We introduced novel measures that assess the match between an ontology and discourse . These differ from former approaches to ontology comparison by focusing on concept and concept-to-concept relations , as these are the ontology elements present in textual statements . Moreover , our measures account for conceptual distinctions between comparing ontologies to one another versus to the discourse associated with a knowledge domain . In the latter comparison , the notion of a false positive , or a concept that appears in ontology but not in text is misleading , as it does not necessarily indicate the concept was not in discourse , but that the discourse was insufficiently sampled . Building on these insights , we introduce novel measures that capture the Breadth and Depth of an ontology's match to its domain with three versions of increasing complexity . Breadth is the total probability mass behind an ontology's concepts and relations with respect to the reference corpus . Depth , in contrast , is its average probability mass per concept and relation . Metaphorically , if breadth is “national income , ” then depth is “income-per-capita . ” An ontology with greater breadth captures more concepts and relations; an ontology with greater depth better captures its most important ones . By measuring the match between a medical ontology and a corpus of medical documents , we are also assessing the utility of each ontology's terms and relations for annotating that corpus . In this sense , breadth measures the overall utility of a given ontology in annotation , whereas depth measures the average annotation utility per ontology constituent . We also defined the fittest ontology of fixed size such that depth is maximized over all concepts and relations in order to more carefully compare ontologies of different sizes . For an arbitrary ontology we also computed its depth loss relative to the fittest ontology of same size ( see Materials and Methods ) . This approach not only allows us to control for size in comparing ontologies , but also has direct application for pruning an ontology of its most improbable parts . To illustrate the meaning and relation of depth loss to depth and breadth , imagine a casino with an enormous roulette wheel on which numbers may appear more than once , and some much more frequently than others . A gambler has limited time to observe the wheel before picking a set of numbers on which to bet . In this analogy , the numbers correspond to concepts and relations in science , the gambler to an ontologist , and a win to an efficient representation of science . The probability of winning or achieving a good scientific representation given a set of bets maps to breadth and the probability of winning normalized by number of bets to depth . The fittest ontology of given size is an optimal bundle of bets: the gambling ontologist can still lose by missing any particular concept or relation , but her risk is minimized . Depth loss , then , is the unnecessary risk of losing a gamble beyond that required by the constrained number of bets . As an ontology grows in size , the overall probability of missing an important scientific concept or relation shrinks . Therefore , depth loss will usually decrease as ontologies grow , even if the smaller ontology has greater depth . By capturing the breadth and depth of an ontology's coverage , our measures suggest precisely what the analyst gains by assessing the direct match between ontology and discourse , rather than attempting to extract or “learn” an ontology from discourse and subsequently compare it with a reference ontology . When an ontology is developed from discourse , all information about the relative frequency with which concepts and relations occur in the domain is lost . Consequently , a match with such an ontology can only grossly capture the representativeness of relations in the reference ontology . The larger difference between these approaches , however , is in the position of authority . Our measures suggest that discourse is the authoritative source of a community's scientific knowledge and should be the reference against which most scientific ontologies are judged . Measures that assess “learned ontologies” with a gold standard , by contrast , assume that ontologists and their constructions are the ultimate reference . Our approach to ontology evaluation has several limitations . It may be viewed as restrictive due to its reliance on the availability of a large corpus related to the domain of interest . This is usually not a problem for biomedical ontologies as the amount of biomedical text is typically overwhelming . For esoteric ontologies , however , it may be difficult to locate and sufficiently sample the textual domain they are intended to map . At the extreme , consider a hypothetical ontology configuring entities corresponding to a novel theory . Further , one can imagine ontologies for which any degree of match to an external domain is meaningless . For example , a hypothetical mathematical ontology should be , first and foremost , clear and internally consistent . As is common in mathematics , relevance to external research may not be required . This level of abstraction and invariance to reality , however , is atypical for biomedicine and other areas of science where the corpus of published research indicates much of what is known . Our approach addresses only one dimension of ontology quality: its match to collective discourse . Other quality dimensions such as consistency and usability are also clearly important . We do not advocate retiring other views of ontology quality: our measures of external validity can be used synergistically with assessments of internal validity to expand the overall utility of an ontology . Another limitation of our method is that we assume that formal relations among ontology concepts are represented explicitly in text , like the concepts themselves . As Brewster and colleagues have pointed out [36] , this is often not the case . More advanced methods are needed to improve on our use of concept co-occurrence . Our approach depends heavily on the advancement of parsing and mapping technologies to enable linkages between ontology concepts and their textual instances . It is particularly dependent on quality in the part-of-speech tagging , recognition of verb nominalization [58] and the association of inflectional and morphological variations in vocabulary . In this way , proper application of our proposed method demands that users surmount significant technical hurdles . It is not trivial to map concepts and relations from an ontology to a real corpus considering the ambiguities and complexities of unstructured discourse . Although we believe that these technical problems can be resolved with a reasonable degree of accuracy , there remains a lingering concern that ontology evaluation is confounded by imperfections in the analysis of text . To address this concern , our analysis of synonym substitution probabilities suggests a practical approach for generating probabilistic domain-specific thesauri that can be immediately used in more closely mapping arbitrary ontologies to text . These substitution probabilities can also be deployed to improve the cross-mapping of ontologies , expanding database queries , and text mining . Several previous approaches to ontology comparison involve explicit comparison of the entire taxonomy of relations . Our approach instead emphasizes comparison of ontology relationships individually . This is because metrics of taxonomic distance between two ontologies [23]–[28] are not easily transplanted to the comparison of ontology with text . Ontology comparisons often weight the match between concepts by the centrality of those concepts in each ontology's hierarchy [26] . The upper-level – the most central and abstract – relations in an ontology , however , are rarely mentioned explicitly in prose . This is partly because of the indexical power of context: an article published in the journal Metabolism does not need to mention or describe metabolism to its audience . The publication alone signals it . In contrast , specific concepts that are taxonomically close to the bottom of the hierarchy – the “leaves” of the tree – are often mentioned in text with disproportionate frequency . In short , while centrality denotes importance within an ontology , and ontology importance should correlate with frequency in discourse , we expect that this relationship is confounded in scientific domains where the most central “branching” concepts are likely so conditioned by context ( e . g . , a biology journal ) that they remain unspoken . In summary , our measures provide a reliable assessment of ontologies as representations of knowledge . We demonstrated their utility using biomedical ontologies , English thesauri and corpora , and we showed that different corpora call for different representations . We believe our straightforward approach can be extended to arbitrary ontologies and knowledge embedded in the literature of their communities . For example , our approach can directly assess the degree to which other popular ontologies represent published knowledge in their respective domains . Our approach would also recommend how these ontologies could be made more efficient or parsimonious . Finally , our measures facilitate comparison between competing ontologies . In conjunction with efforts to make ontologies logically consistent , greater external validity will insure that ontological inferences anchor to the most salient concepts and relations used by the community of science .
We used four medical ontologies , seven English thesauri ( Table 1 ) , and three corpora ( Table 2 ) from the areas of medicine , news , and novels . The four biomedical ontologies we used were ICD9-CM , CCPSS , SNOMED-CT , and MeSH each described in the following paragraphs . ICD9-CM [48] , [54] , the International Statistical Classification of Diseases and Related Health Problems , is a taxonomy of signs , symptoms , abnormal findings , complaints , social circumstances , and external causes of injury or disease . It uses predominantly one type of relation ( is-a ) , whereas CCPSS and SNOMED CT employ richer repertoires of relation types . The International Classification of Diseases is published by the World Health Organization ( WHO ) and is used worldwide for morbidity and mortality statistics , reimbursement systems , and automated decision support in medicine . The ICD9-CM version was created by the U . S . National Center for Health Statistics as an extension of the ICD9 system to include diagnostic and operative procedures – the CM referring to clinically modified . Here we use the 2009 version of ICD9-CM . A typical relation between two concepts in ICD9-CM looks as follows: CCPSS , the Canonical Clinical Problem Statement System [55] , is a knowledge base that encodes clinical problems encountered by ailing humans . It is specifically designed to encode clinical knowledge regarding relations between medical conditions . Typical relations encoded in CCPSS look as follows: SNOMED CT , Systematized Nomenclature of Medicine – Clinical Terms [56] , is a synthesis of terminologies produced by the College of American Pathologists and by the National Health Service of the United Kingdom . The American component is called SNOMED Reference Terminology , and the British one is referred to both as Clinical Terms and Read Codes . SNOMED CT is the most comprehensive clinical terminology in existence and includes ∼350 , 000 concepts . A typical relation in SNOMED CT looks as follows: Medical Subject Headings ( MeSH ) [49] is a comprehensive controlled vocabulary designed by the United States National Library of Medicine ( NLM ) . Its intended use is information retrieval; MeSH was not designed as a formal ontology . The 2009 version contains a total of 25 , 186 subject headings spanning anatomy; organism classification; diseases; chemicals and drugs; food and beverages; analytical , diagnostic and therapeutic techniques and equipment; health care , psychiatry and psychology; biological and physical sciences; anthropology , education , sociology and social phenomena; persons; technology and information science; humanities; publication characteristics and geographic locations . It is mainly used by the MEDLINE/PubMed article database for indexing journal articles and books . A typical relation present in the MeSH is-a hierarchy looks like We tested the medical ontologies against a corpora of modern medicine comprised of clinical journal article abstracts from the PubMed database . We limited ourselves only to English abstracts in the core clinical journals for the entire period covered by PubMed , 1945 through February of 2009 . The resulting corpus included 786 , 180 clinical medicine-related abstracts ( see Table 2 ) . Our broader analysis of synonym dictionaries included seven of the most common , sampling from very different kinds of thesauri . These include the large thesauri ( 1 ) The Synonym Finder and ( 2 ) Webster's New World Roget's A–Z Thesaurus; moderately-sized thesauri ( 3 ) 21st Century Synonym and Antonym Finder and ( 4 ) The Oxford Dictionary of Synonyms and Antonyms; and portable , compact thesauri ( 5 ) A Dictionary of Synonyms and Antonyms and ( 6 ) Scholastic Dictionary of Synonym , Antonyms and Homonyms . Each thesaurus shared a common layout involving alphabetically arranged headwords followed by synonyms ( and antonyms ) . Finally , we included the electronic dictionary ( 7 ) WordNet , which arranges its words asymmetrically into sets of synonyms or “synsets . ” To evaluate the match between these thesauri and a variety of text corpora , we added English news and novels to our sample of clinical medicine ( see Table 2 ) . The news corpus covered all Reuters news stories between 08/20/1996 and 08/19/1997 . The novels corpus contained 50 of the most influential novels of the 19th Century , written or translated into English . Complete information regarding each of these data sources can be found in the supplement . To map biomedical concepts to our clinical corpus we used MetaMap . MetaMap [59] is a knowledge-intensive natural language processing program developed at the National Library of Medicine for mapping snippets of biomedical text to the UMLS Metathesaurus [60] , [61] . MetaMap uses the SPECIALIST minimal commitment parser [62] to conduct shallow syntactic parsing of text – using the Xerox part-of-speech tagger . For each identified phrase its variants are generated using the SPECIALIST lexicon and a supplementary database of synonyms . A phrase variant comprises the original phrase tokens , all its acronyms , abbreviations , synonyms , derivational variants , meaningful combinations of these , and inflectional and spelling variants . Given a collection of phrase variants , the system retrieves from the Metathesaurus a set of candidate strings each matching one of the variant constituents . Each Metathesaurus-derived string is evaluated against the input text by a linear combination of four metrics , called centrality , variation , coverage and cohesiveness . The first two metrics quantify matches of dictionary entries to the head of the phrase , and the mean inverse distance between dictionary and text phrases . The latter two metrics measure the extent and sparsity of matches between the textual and dictionary strings . The candidate matches are then ordered according to mapping strength , and the highest-rank candidate is assigned as the final match . We used MetaMap's Strict Model to filter matches in order to achieve the highest level of accuracy [57] . The UMLS ( Unified Medical Language System ) Metathesaurus is a rich terminological resource for the biomedical domain [63] , [64] . All concepts in the UMLS Metathesaurus are categorized into 135 semantic types ( or categories ) . In this work we focused on the semantic type of “Disease or Syndrome” . This is why the counts of concepts and relations in Table 3 are much less than the total number of concepts and relations from each of the four ontologies in Table 1 . We used the Stanford POS tagger [65] , [66] to parse the news and novels corpora comparable to MetaMap's parsing of medical texts . After parsing , we processed the inflectional and morphological variations of each word . For the medical corpus , we retrieved the base form of a word by querying the UMLS Specialist Lexicon based on its appearance in the text ( e . g . , singular or plural for a noun , different tenses for a verb ) . For the news and novels corpora , we converted all words to their base word form ( e . g . , translating nouns from plural to singular and verbs from past and future to present tense ) with a rich set of morphological rules . Then we used these base word forms , in addition to their part of speech , to indicate word context for the calculations below . We also used these base forms to match against thesaurus entries . In this section , we define several metrics for mapping an ontology to a corpus , arranging the metrics by increasing complexity . The simpler metrics do not distinguish between multiple predicate types in an ontology , summarizing all relations between the same pair of concepts , i and j , with a single association probability , pij . More general versions of our metrics account for multiple relation types that occur in more complex ontologies , but these involve numerous additional parameters that require estimation from real data and therefore are more challenging to implement . For this reason , we count relations represented in a test ontology X in two separate ways . |ℜX | is the number of ordered pairs of concepts with at least one relation defined between them in ontology X , while |RX| is the total number of all relations in the ontology . For predicate-poor ontologies such as thesauri , these two ways of counting relations are equivalent . In predicate-rich ontologies with more than one relation between the same pair of concepts , |RX|>|ℜX | . Suppose an ontology has N concepts and each concept i has relations with other Mi concepts ( each denoted as concept j where j = 1 , 2 , … , Mi ) . We practically infer the probability pij that concept i is associated with concept j through simple concept co-occurrence in text . Namely , we estimate: ( 1 ) where nij is the number of times concept i co-occurs with concept j in the same unit of text , such as a sentence or a paragraph ( the medical abstract in our implementation ) . Note that when concept i is unobserved in the corpus , we encounter a singularity ( zero divided by zero ) when applying equation 1 directly and pij violates the basic property of probability by not summing to 1 . For this study we pragmatically postulate that if concept i is not observed in the corpus , then the value of pij is set to 0 . Datasets S1 , S2 , and S3 contain complete sets of non-zero estimates of synonym substitution probabilities for our three reference corpora . The advantage of setting pij to 0 when i is unobserved is that the ontology will be punished for concepts and relations unobserved in the corpus . One could alternately make pij behave as a probability under all conditions ( for all values of nij ) and still punish the ontology by making pij very small for all unobserved i in the following manner: ( 2 ) where parameter α and are small positive constants ( 0≤α β 1 ) . This would require us to further add a pseudo-concept , that relates to every concept i with the following probability: ( 3 ) such that is close to 1 when i is not observed and every pij is approximately 0 . One can imagine the use of more advanced natural language processing techniques than co-occurrence to assess the precise semantic relation in text , but we use the probability estimate from equation 1 in our preliminary evaluation of four medical ontologies against our corpus of clinical abstracts . Consider further an arbitrary ontology that has multiple distinct relations defined for the same pair of concepts . In such a case , we could supplement pij with an additional set of parameters , πk|ij . These new parameters reflect the relative frequency ( importance ) of textual mentions of the kth relation between concepts i and j , where In the case of thesauri , in which the primary relation is synonymy , we are able to assess pij more precisely than with medical ontologies . An English thesaurus has N headwords and each headword ( denoted as wi where i = 1 , 2 , … , N ) has a list of M synonyms ( denoted as wi , j where j = 1 , 2 , … , Mi ) . We compute the probability of substituting word wi with its synonym wi , j through probabilistic conditioning on all contexts observed in a corpus in the following way . ( 4 ) where is a shorthand for “sum over all possible contexts of headword ” . Equation ( 4 ) is closely related to distributional similarity metrics explored by computational linguists , e . g . [67] . This notion , that words occurring in the same contexts tend to have similar meanings is called the Distributional Hypothesis and was introduced by Zellig Harris [68] , then popularized by Firth—“a word is characterized by the company it keeps” [69] . Some researchers prefer to induce word relationships like synonymy and antonymy from co-occurrence rather than substitution in order to capture lexical as well as semantic similarity [70] , [71] . In our analysis , however , we do not induce synonymy , but rather begin with established synonyms from a published Thesaurus . We then simply calculate their substitution frequencies based on shared context . In our practical implementation , we defined the context of word wi within a sentence as a list of k words immediately preceding and following it , enriched with positional and part-of-speech ( POS ) information: To increase the number of comparable four-token contexts for synonyms in our relatively small corpus , we only considered nouns , verbs , adjectives and adverbs in our analysis of context , disregarding tokens with other part-of-speech tags . That is , given a word wi in the text , we select the nouns , verbs , adjectives and adverbs around it within window size 2k = 4 ( two before and two after wi ) , providing a four-word context for all words except those at a sentence boundary . Because many contexts constructed in this way are unique or very rare , we generalize them by ignoring word order and binning words that appear uniquely in the corpus into part-of-speech pseudo-words ( e . g . , rare-noun , rare-verb , rare-adjective , and rare-adverb ) . Equation 4 suffers the same limitation as equation 1 for headwords i that do not occur in corpus . One could extend it in the same manner as equation 1 by adding the pseudo-concept such thatcollects the vast majority of the probability mass for unobserved headwords . In information retrieval ( IR ) , the goal is to identify documents from a large collection most relevant to a user's query . If the subset of relevant documents is known , we can calculate the quality of an information retrieval method with the metrics precision , recall , and the F-measure ( harmonic mean of precision and recall ) . ( 5 ) ( 6 ) ( 7 ) True positives ( tp ) , false positives ( fp ) , false negatives ( fn ) and true negatives ( tn ) are defined by the cross-tabulation between relevance and retrieval: True positives comprise documents that are both relevant to the query and retrieved by the method; false positives are documents retrieved but irrelevant; false negatives are relevant but not retrieved; and true negatives are irrelevant documents not retrieved . More measures , such as accuracy and fallout , are introduced and computed in Text S1 . For a given reference corpus T , we define the complete ontology , which incorporates all concepts and all relations encountered in corpus T . We also use the corpus to derive a frequency for each concept i in CT , the set of all concepts in T , and concept association probabilityfor each relation in RT , the set of all relations in T . In the special case of a thesaurus , we understand this probability to be the probability of appropriate substitutability , or “substitution probability” for short . It should be noted that our ability to estimate fi depends on mapping concepts from ontology to text . This is why we spent so much time and energy working with thesauri to facilitate the detection of concept synonyms in text . should be normalized in such a way that ( N is the total number of concepts in corpus T ) and , by definition , is normalized so that for any concept , ci , involved in at least one relationship . In our implementation , we approximate the complete ontology for our medical corpus with all “Disease or Syndrome” concepts in MetaMap , which includes the union of our four medical ontologies in addition to more than a hundred additional terminologies , such as the UK Clinical Terms , Logical Observation Identifiers Names and Codes ( LOINC ) that identifies medical laboratory observations , RxNorm that provides normalized names for clinical drugs , and the Online Mendelian Inheritance in Man ( OMIM ) database that catalogues diseases with a known genetic component . The complete ontology only retains those concepts and relations that appear in the corpus . For our thesauri , we approximated the complete ontology with the union of compared thesauri , excluding concepts and relations not found in the corpus . Consider that we are trying to evaluate arbitrary ontology X with respect to reference corpus T . We define CX and RX as sets of concepts and relations within X , and | CX | and | RX | the cardinalities of those sets . To evaluate X with respect to T , we identify sets CX ( true-positives—tp ) , CX ( false negatives—fn ) , RX ( tp ) , and RX ( fn ) such that CX ( tp ) = CX ∩ CT , RX ( tp ) = RX ∩ RT , CX ( fn ) = CT — CX ( tp ) , and RX ( fn ) = RT — RX ( tp ) , where “—” represents set difference . Then we derive the first ontology evaluation measure—Breadth—to capture the theoretical coverage of an ontology's concepts: ( 8 ) We derive a corollary version of breadth to capture the theoretical coverage of an ontology's concept and relations: ( 9 ) where pij equals 0 if there is no relation between them in X . Both Breadth metrics are defined on the interval [0 , 1] . By modifying these measures to account for the number of concepts and relations , we develop measures of Depth to capture theoretical parsimony: ( 10 ) ( 11 ) where |ℜX| is the number of ordered pairs of concepts with at least one relation defined between them in ontology X . This normalization thus ignores the number of different relations that X may catalog between concepts i and j . We can also compare an arbitrary ontology X with the fittest ontology of the same size O ( X ) by including the most representative CX concepts and RX relations from corpus T that maximize depth . In practice , to compute the fittest ontology of fixed size , we have to perform a numerical optimization over a set of concepts and relations where the size of the ontology being optimized is kept fixed , but the concepts and relations taken from the fittest ontology are added or removed to improve the breadth and depth of the optimized ontology . An estimate of the depth of the fittest ontology of fixed size , DepthO ( X ) ( T ) , allows us to define and compute a Loss measure . ( 12 ) The above measure can be called the Loss of Depth or Depth Loss . In a similar way we can compute an ontology's Loss of Breadth . ( In practice , our estimates of the fittest ontology of fixed size were constrained only by the total number of relations in the corresponding test ontology , so that the Depth Loss in Table 2 was computed using equation ( 19 ) in Text S1 . ) Note that unlike Breadth , Depth is not naturally defined on the interval [0 , 1] , but will rather tend to result in small positive numbers . Therefore , we define normalized versions of Depth and Depth Loss in the following way . ( 13 ) ( 14 ) If we consider an arbitrary ontology with multiple types of relations between concepts i and j , we can further extend Breadth2 and Depth2 measures in the following way: ( 15 ) ( 16 ) Note that this definition of Depth3 and Breadth3 involves three levels of ontology evaluation: parameter fi captures usage of the ith concept in the corpus; parameter pij reflects the relative importance of all relations between concepts i and j with respect to all relations involving concept i in the corpus; and parameter πk|ij measures the relative prevalence of the kth relation between concepts i and j in the corpus . Precise implementation of this task would require capturing mentions of every concept i – relation k – concept j triplet in the text using natural language processing tools . The parameter estimates would then be computed by normalizing counts of captured relations and concepts in an appropriate way . If , on average , there is only one type of relation per pair of concepts , use of metric Depth3 and Breadth3 would be overkill . For computational simplicity , we use only the first- and the second-level Breadth and Depth in our practical implementation . | An ontology represents the concepts and their interrelation within a knowledge domain . Several ontologies have been developed in biomedicine , which provide standardized vocabularies to describe diseases , genes and gene products , physiological phenotypes , anatomical structures , and many other phenomena . Scientists use them to encode the results of complex experiments and observations and to perform integrative analysis to discover new knowledge . A remaining challenge in ontology development is how to evaluate an ontology's representation of knowledge within its scientific domain . Building on classic measures from information retrieval , we introduce a family of metrics including breadth and depth that capture the conceptual coverage and parsimony of an ontology . We test these measures using ( 1 ) four commonly used medical ontologies in relation to a corpus of medical documents and ( 2 ) seven popular English thesauri ( ontologies of synonyms ) with respect to text from medicine , news , and novels . Results demonstrate that both medical ontologies and English thesauri have a small overlap in concepts and relations . Our methods suggest efforts to tighten the fit between ontologies and biomedical knowledge . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"computational",
"biology/literature",
"analysis",
"computational",
"biology/bio-ontology"
] | 2011 | Benchmarking Ontologies: Bigger or Better? |
Rift Valley fever ( RVF ) is a zoonotic disease caused by Rift Valley fever virus ( RVFV ) found in Africa and the Middle East . Outbreaks can cause extensive morbidity and mortality in humans and livestock . Following the diagnosis of two acute human RVF cases in Kabale district , Uganda , we conducted a serosurvey to estimate RVFV seroprevalence in humans and livestock and to identify associated risk factors . Humans and animals at abattoirs and villages in Kabale district were sampled . Persons were interviewed about RVFV exposure risk factors . Human blood was tested for anti-RVFV IgM and IgG , and animal blood for anti-RVFV IgG . 655 human and 1051 animal blood samples were collected . Anti-RVFV IgG was detected in 78 ( 12% ) human samples; 3 human samples ( 0 . 5% ) had detectable IgM only , and 7 ( 1% ) had both IgM and IgG . Of the 10 IgM-positive persons , 2 samples were positive for RVFV by PCR , confirming recent infection . Odds of RVFV seropositivity were greater in participants who were butchers ( odds ratio [OR] 5 . 1; 95% confidence interval [95% CI]: 1 . 7–15 . 1 ) and those who reported handling raw meat ( OR 3 . 4; 95% CI 1 . 2–9 . 8 ) . No persons under age 20 were RVFV seropositive . The overall animal seropositivity was 13% , with 27% of cattle , 7% of goats , and 4% of sheep seropositive . In a multivariate logistic regression , cattle species ( OR 9 . 1; 95% CI 4 . 1–20 . 5 ) , adult age ( OR 3 . 0; 95% CI 1 . 6–5 . 6 ) , and female sex ( OR 2 . 1; 95%CI 1 . 0–4 . 3 ) were significantly associated with animal seropositivity . Individual human seropositivity was significantly associated with animal seropositivity by subcounty after adjusting for sex , age , and occupation ( p < 0 . 05 ) . Although no RVF cases had been detected in Uganda from 1968 to March 2016 , our study suggests that RVFV has been circulating undetected in both humans and animals living in and around Kabale district . RVFV seropositivity in humans was associated with occupation , suggesting that the primary mode of RVFV transmission to humans in Kabale district could be through contact with animal blood or body fluids .
Rift Valley fever virus ( RVFV ) is a single-stranded RNA virus in the order Bunyavirales , and recently classified in the Phenuiviridae family [1] and the Phlebovirus genus . RVFV causes disease in humans and animals [2] , and is transmitted by mosquitoes to livestock such as sheep , goats , and cattle [3] . Competent mosquito vectors include species from the Aedes , Culex , Anopheles , Eretmapodites , Mansonia , and Coquillettidia genera [4 , 5] . Humans can become infected with RVFV when they come into contact with blood or body fluids of infected livestock while caring for sick animals , assisting with animal birth , or slaughtering livestock; or through bites of infected mosquitoes . Occupations at the greatest risk of RVFV infection include herdsmen and butchers [6–8] . Raw milk or meat consumption are potential sources of RVFV , although transmission via these routes has not been confirmed . In livestock , RVFV infection can cause increased abortions and stillbirths , and high mortality in neonates and juvenile animals . As a result , RVFV outbreaks can lead to significant economic losses [3 , 9] . In humans , infection can range from asymptomatic or a mild flu-like illness to more severe disease that includes hepatitis , retinitis , or encephalitis [10] . Approximately 1% of human cases develop hemorrhagic disease , and an estimated 1–2% of cases are fatal . However , during an outbreak in Saudi Arabia , case fatality was as high as 14% [10] . There is no specific treatment for RVFV infection in humans or animals , but supportive care may prevent complications and decrease mortality [2] . Currently , no RVFV vaccine is approved for use in either humans or animals in North America or Europe [3] . Some inactivated and live-attenuated vaccines have been developed and have been efficacious in animals [11 , 12] . Development of human RVFV vaccines has been challenging due to the safety of the vaccine . An experimental human vaccine , TSI-GSD200 , has shown some utility in laboratory workers , but has not been used extensively in other settings [11] . To better understand the utility of RVFV vaccines in a particular setting , the prevalence of disease in humans and animals must first be understood . Rift Valley fever ( RVF ) outbreaks were first reported in East Africa in the 1930s [13] . In 1997–1998 , large RVF outbreaks in northeastern Kenya were associated with El Niño rains and floods , resulting in many deaths in livestock and humans [14 , 15] . Infection in humans was highly associated with contact with livestock and animal body fluids . RVF outbreaks were not reported again in East Africa until 2006–2007 , when large numbers of humans and livestock were infected in Kenya , Somalia , Tanzania , and Sudan [16 , 17] . Studies following these outbreaks reported that being a herdsman and handling or consuming products from infected animals were major risk factors for human infection . Additionally , outbreaks in East African countries mainly occurred in areas with poor soil drainage and flat lowlands that are less than 500 m above sea level [18 , 19] . Also , RVF outbreaks in humans and animals following flooding have occurred in Sudan [20] , Saudi Arabia [21] , Yemen ( 2000–2001 ) , South Africa , and Egypt [22] . Uganda is an East African nation that shares borders with Kenya and Tanzania , and to date no large RVF outbreaks have been reported since 1968 , when 7 human cases occurred near Entebbe [23] . However , a 2013 serological survey of goats in Ssembabule , Mpigi , Masaka , and Mubende districts in Uganda showed a seroprevalence of 9 . 8% , suggesting RVFV circulation [24] . The Uganda Virus Research Institute ( UVRI ) in Entebbe has been implementing laboratory-based surveillance for viral hemorrhagic fevers ( VHF ) in Uganda since 2010 , which includes testing for RVFV [25] . In March 2016 , UVRI confirmed 2 acute human RVF cases in Kabale district in the southwestern region of Uganda [26] . These were the first human RVF cases identified in Uganda since 1968 [23] . Because not all human or animal RVF cases are symptomatic , RVFV infections are often undetected . Thus , UVRI , the Ugandan Ministry of Health ( MoH ) , Ugandan Ministry of Agriculture Animal Industry and Fisheries ( MAIF ) , and the United States Centers for Disease Control and Prevention ( CDC ) collaborated on a study to assess the seroprevalence of RVFV in humans and animals living in and around Kabale district . The objectives of the study were to determine the sero-prevalence of RVFV in both humans and animals in Kabale and surrounding districts , identify risk factors and high-risk areas for RVFV , determine if RVF is emerging or endemic , and identify unrecognized RVF cases that may be related to the 2016 outbreak .
Ethical approval for this study was granted from review by the UVRI Research Ethics Committee ( UVRI REC: GC/127/16/03/551 ) . Animal subjects work was conducted according to Uganda national guidelines and performed by officers from Kabale District and the Ministry of Agriculture , Animal Industries and Fisheries . The CDC National Center for Emerging and Zoonotic Infectious Diseases ( NCEZID ) Human Subjects office classified this project as non-research because the survey was a follow-up to the confirmed outbreak in Kabale district and the results of the study would assist the local health officials to target public health actions and interventions based on the serosurvey results ( NCEZID: 032316TS ) . Kabale district is located in the southwestern corner of Uganda . According to the 2014 Uganda census , it has an estimated population of 534 , 160 people , with the majority living in a rural setting ( 457 , 592 of 534 , 160; 86% ) [27] . The altitude of Kabale ranges from 1 , 219 m to 2 , 347 m above sea level . Agriculture is an important source of revenue in this region . Most families own goats , sheep , cattle , and pigs . In addition to agricultural lands , Kabale district also has areas with high-altitude rain forests and savannahs . From April 1–12 , 2016 , 34 locations in and near Kabale district were selected for inclusion in the serosurvey by a multidisciplinary team consisting of individuals from Kabale district , Uganda MoH , Uganda MAIF , UVRI , and CDC ( Fig 1 ) . Four categories of people and animals were targeted for sampling based on perceived risk of RVFV infection . These were: 1 ) Animal slaughter house ( abattoir ) workers and the animals ( cattle , goats , and sheep ) slaughtered at the abattoir; 2 ) persons and animals from villages that had confirmed or probable human RVF cases; 3 ) persons and animals from villages considered at-risk for RVF due to geographic conditions ( see below ) ; and 4 ) persons and animals from randomly selected villages with no reported RVF cases . Additionally , animals were sampled from neighboring districts of Ntungamo and Kisoro . Since several studies have demonstrated increased risk of RVFV exposure in butchers , we selected this as a high-risk group for sampling . The research team worked with Kabale district health and veterinary officials to select the study sites determined to be at risk for RVF; these sites were identified based on the terrain , propensity for flooding , human and animal population density , cooperation from the community , and sharing of international borders . When the team arrived at a selected location , local health workers assisted in recruiting participants for the serosurvey from the village by word of mouth . The questionnaire was written in English but was translated to the local language ( Rukiga ) and administered orally by local health workers ( S1 ) . Convenience sampling was used; all participants who presented to the sampling site were considered eligible for the study . All participation was voluntary and participants did not receive any compensation . Participants completed a consent form for the questionnaire , which was translated from English into local languages and explained by team members . Children older than 7 years were allowed to participate if a parent provided written consent . Blood samples were collected from all but 2 eligible participants identified . Goats , sheep , and cattle that were brought to the collection site were sampled in relative proportion to the size of the herd at a given study site . Due to limited grazing areas in the Kabale region , few livestock owners maintain herds of more than 15 animals per species of cattle , sheep and goats . In cases where a livestock owner had less than 15 animals in a herd , all animals would be sampled . If an owner had more than 15 animals in a herd , only 25% of the herd was sampled . Only 3 of the households we sampled reported to have more than 15 animals per herd . We sampled all animals whose owners provided consent and presented their animals to the survey team . Both animals with a history of previous abortions or reported as having symptoms compatible with RVF , and apparently healthy animals of varied ages and sexes were sampled . Information collected about each herd included the general health of the herd , size of the herd , and grazing patterns . Because some humans and animals traveled from other subcounties to the location where the study was being administered , we did not collect both human and animal samples from all targeted subcounties . One ~4 cc blood sample was collected from each human and animal participant for serological testing . Human specimens were tested by ELISA for anti-RVFV IgM and IgG , and animal specimens were tested for anti-RVFV IgG only . ELISA testing of both human and animal samples was performed at UVRI as previously described [28] . Human blood specimens that were IgM positive were subsequently tested by RT-PCR for RVFV-specific RNA targeting the L genome segment[29] . Briefly , following heat and detergent inactivation , specimens were tested by anti-RVFV-specific IgM and IgG ELISA using inactivated RVFV-infected Vero-E6 cell antigens , using 4 dilutions of each specimen ( 1/100 , 1/400 , 1/1600 , and 1/6400 ) . Titers and the cumulative sum optical densities of each dilution ( SUMOD ) minus the background absorbance of uninfected control antigen ( adjusted SUMOD ) were recorded . Samples were deemed positive if both the adjusted SUMOD and titer were above pre-established conservative cutoff values of ≥0 . 45 for IgM ELISA and ≥0 . 95 for IgG ELISA [28] . Questionnaire data were entered into a Microsoft Excel worksheet and analyzed using Stata 13 ( StataCorp LP , College Station , TX , USA ) . The significance of human risk factors associated with RVFV seropositivity was first determined using bivariate analysis based on Pearson’s χ2 . A p-value <0 . 05 was considered statistically significant . After completing the bivariate analysis , the significance of independent variables as predictors was further assessed using multivariate logistic regression . All risk factors found to be significantly associated with human RVFV seropositivity were included in the multivariate logistic regression model . A Pearson’s χ2 goodness of fit test was done on the final model [30] . To determine if animal seropositivity at the subcounty level was significantly associated with human seropositivity within that subcounty , a multivariate logistic regression was performed adjusting for risk factors found to be significant in the bivariate analysis .
A total of 655 persons participated in the serosurvey ( Table 1 ) . Participants were recruited at the Kabale town abattoir ( n = 117; 18% of participants ) , from villages where a recent acute RVFV case had been identified ( n = 237; 37% ) , and from villages with no recorded outbreaks ( n = 293; 45% ) . Most participants ( n = 396; 60% ) were aged 20–49 years and had completed primary education ( n = 360; 55% ) . The most common occupation listed was farmer or herdsman ( n = 335; 52% ) , and most individuals owned animals ( 60% ) . Contact with animals was common , with 78% ( n = 508 ) of participants reporting contact with animals in the past year . Of all persons tested , 13% ( 88/655 ) were RVFV seropositive . Three ( 0 . 5% ) persons had anti-RVFV IgM only , 78 ( 12% ) had IgG only , and 7 ( 1% ) had both IgM and IgG . Two individuals positive for RVFV IgM also tested positive for RVFV RNA by RT-PCR , suggesting active infection at the time of sampling . The 3 IgM-only positive individuals ( a trader , a housewife , and a farmer ) were all from the village in which one of the initial acute human RVFV cases was living but were not related to the initial acute RVF case . No persons under 20 years of age were RVFV seropositive , while 17% ( n = 66 ) of individuals aged 20–49 years were seropositive ( Table 2 ) . Of individuals 50 years and older , 11% ( n = 22 ) were seropositive . Butchers were the most likely to be RVFV seropositive , with 35% showing evidence of seropositivity . Other occupations evaluated for RVFV seropositive include farming at 10% , housewife at 8% ( 4/49 ) , teacher at 18% ( 2/11 ) and trader at 12 . 5% ( 3/24 ) . In the bivariate analysis , older age groups ( χ2 = 14 . 4; p = 0 . 001 ) , male sex ( χ2 = 11 . 9; p = 0 . 001 ) , occupation as a butcher ( χ2 = 54 . 7; p < 0 . 001 ) , history of slaughtering or butchering animals ( χ2 = 23; p < 0 . 001 ) , and preparing raw meat ( χ2 = 13; p < 0 . 001 ) were all significiently associated with an increased risk of RVFV seropositivity ( S1 Table ) . In the multivariate logistic regression , being a butcher and handling raw meat were significantly associated with RVFV seropositivity , with an adjusted OR of 5 . 1 ( 95% CI 1 . 7–15 . 1; p = 0 . 003 ) and 3 . 4 ( 95% CI 1 . 2–9 . 8; p = 0 . 024 ) , respectively ( Table 3 ) . Age , sex , slaughtering/butchering , and contact with animals through grazing were not significantly associated with RVFV seropositivity in the multivariate model . Of all animals tested , 13% ( 133/1051 ) were RVFV seropositive . Seropositivity varied by species , sex , and age group among animals both in a bivariate analysis and the multivariate logistic regression ( S2 Table; Table 4 ) . Cattle showed significantly higher odds of being seropositive even after adjusting for age and sex ( OR 9 . 1; 95% CI 4 . 1–20 . 5; p < 0 . 001 ) . Adult animals and females also had significantly higher odds of being RVFV seropositive , with OR 3 . 0 ( 95% CI 1 . 6–5 . 6; p = 0 . 001 ) and OR 2 . 1 ( 95% CI 1 . 0–4 . 3; p = 0 . 04 ) , respectively . Human and animal seropositivity varied by subcounty ( Figs 2 and 3 ) , ranging 0–36% ( standard deviation 11; mean 14% ) , whereas animal seropositivity ranged 4–28% ( standard deviation 7; mean 12% ) ( S3 Table ) . Human RVFV seropositivity was significantly associated with close contact with cattle ( P-value = 0 . 003 ) , but shown not to be significant for contact with small ruminants ( p-value = 0 . 06 ) ( S1 Table ) . The sub-counties with the highest seroprevalence included the Kabale town council , where the main abattoir is located , and subcounties near bodies of water or wetlands . The association between animal seropositivity and human seropositivity within a subcounty was examined using multivariate logistic regression , adjusting for contact with raw meat and occupation , because these were found to be significant risk factors in the univariate analysis . Human seropositivity within a subcounty was found to be associated with animal seropositivity , with OR of 1 . 1 ( 95% CI 1 . 0–1 . 1; p < 0 . 001 ) .
Although RVFV had not been detected in Uganda since 1968 , our study demonstrates that it has likely been enzootic in Kabale district for some time . Overall , we found evidence of RVFV seropositivity in 13% of humans and animals sampled . Our study also showed that butchers and those who handled raw meat were most likely to be RVFV seropositive . Similar risk factors for RVFV seropositivity have been reported in previous studies . In a 2015 study in Kenya , LaBeaud and colleagues found that male sex , increased age , history of slaughtering livestock , history of malaise , and poor measured visual acuity were all factors for increased seropositivity [8 , 31] . Although we did not find an association with sex and RVFV seropositivity after adjusting for other factors , such as occupation , we did find an association with being a butcher ( i . e . , someone who cuts meat either at home or at a slaughterhouse ) and RVFV seropositivity . Previous studies also found that drinking raw milk may be associated with RVFV seropositivity [32 , 33] , but we were not able to find this association , likely because few individuals ( 36; 5% ) reported drinking raw milk . Anecdotally , individuals reported not drinking raw milk due to concerns about Brucellosis infection . We did not find a significant association between age and seropositivity , but interestingly , no persons younger than 20 years were found to have evidence of RVFV infection . This may be because only 6 individuals below the age of 20 reported having close contact with livestock with the majority in this age group reporting to be attending school , greatly reducing their risk of exposure to potentially infected livestock . Figs 2 and 3 show high seroprevalence , both in humans and animals , in 2 subcounties: Buhara subcounty , bordering Rwanda , and Bubare subcounty , near Kabale town . Kabale town contains the main abattoir , a likely source of RVFV infection in humans . However , Buhara and Bubare subcounties are connected by the primary North-South highway between Kabale town and the Rwandan border , a transportation corridor that could have served as a possible source of introduction of RVFV into Kabale district through livestock trade . A serosurvey of livestock conducted in Rwanda showed overall seropositivity of 16 . 8% , with districts closest to Tanzania showing the highest seropositivity and the 2 districts closest to the Ugandan border having the lowest seropositivity [34] . In addition , evidence of previous RVFV circulation and infection of livestock from samples collected in 2009 in goats from the Southeastern districts of Ssembabule , Mpigi , Masaka and Mubende shows a total seroprevalence of approx . 10% [24] . Although the testing methodologies employed for these samples was different than the one employed in our study , evidence of seropositivity in livestock may begin at earliest in 2009 , or possibly earlier , and may suggest that RVFV was introduced into this region following outbreaks in neighboring countries and maintained through low level inter-epidemic transmission [34] . However , because there are no published reports of RVF seropositivity prior to 2009 , we cannot be certain there was no widespread circulation of RVFV before that time . Our study suggests that RVFV transmission to humans in Kabale district is primarily due to exposure to the blood and body fluids of infected livestock , given that butchers and those handling raw meat were most likely to be RVFV-seropositive . Additionally , we found that human seropositivity was significantly associated with livestock seropositivity in each subcounty . Cattle density has also been previously associated with RVF seropositivity in RVFV models [35] . In our study , significantly more cattle were seropositive ( 27% ) than goats ( 7% ) or sheep ( 4% ) . This difference could be due to mosquito feeding behavior , as mosquitoes tend to select large and ornamented species [36] . Also , cattle are kept longer compared to goats and sheep hence are available to exposure to mosquito bites increasing their chances of being seropositive , however other factors such as mosquitoes species involved could be playing a role in Kabale district . For example , mosquitoes may have a preference for biting cattle compared to other ruminants . Furthermore , sheep and goats are usually kept indoors , especially at night , while cattle are rarely sheltered in Kabale district and thus exposed to nighttime-biting mosquitoes . Mosquitoes collected after outbreak investigations were mostly animal-specific feeders rather than human-specific ( Julius Lutwama , personal communication ) , indicating that RVFV is primarily transmitted by mosquitoes within animal and human populations , however , humans are also infected from direct contact with infected animals . The mosquito species that were trapped in this region during the 2016 RVF outbreak include mainly Aedes and Culex species . Further investigations of the relative contribution of mosquitoes and livestock in RVFV transmission are needed . The high seroprevalence in livestock seen in Kabale town is likely due to the main abattoir , where a majority of the animals in our survey were sampled . Animals , primarily from the neighboring subcounties , but also from more distant subcounties , are brought there for sale and slaughter . The high seroprevalence in animals can be attributed to the convergence of these animals from throughout the district in one location . This high seroprevalence in abattoir animals may be related to beef production dynamics in Uganda where older animals that are no longer of reproduction value are sold off by farmers for slaughter , hence more likely to be seropositive as stated above . The corresponding high seroprevalence seen in the abattoir workers can also be attributed to this , as well as daily occupational exposure to infected animal blood and body fluids . Comparing our results with serological studies conducted in the neighboring countries of Kenya and Tanzania , where RVFV is endemic , provides some insight into our human serosurvey findings . In a study in coastal area of Kenya from 2009–2011 [37] , RVFV seroprevalence in humans was lower ( 1 . 8% ) than in our study . Similarly , the seroprevalence in humans was only 5 . 2% in Mbeya region in Tanzania [38] , compared to 13% in our study . We think this is mainly because we collected samples after a confirmed outbreak or uptick in inter-epidemic transmission , unlike in the Kenya study . However , seroprevalence was higher in domestic ruminants in another study in Garissa , Kenya ( 27 . 6% ) [39] than in our study ( 13% ) . Generally , seroprevalence in both animals and humans is expected to be higher in RVFV-endemic regions of Kenya and Tanzania than what we found in this study , but the risk factors identified were the same–mainly , contact with livestock . In a study in nearby Rwanda , the seroprevalence in cattle was 16 . 8% , slightly lower than 27% in our study [34] . However , these differences could be attributed to various factors , including the serological testing protocol used , which may differ in specificity and sensitivity . Additionally , our study identified local variations in RVFV seroprevalence , so the sampling strategy may also affect seropositivity results . The extent of RVFV infection in Kabale district was greater than anticipated ( as high as 36% in Bubare subcounty , and 28% in Buhara subcounty in animals ) considering that the acutely identified human cases were the first to be laboratory confirmed in Uganda since 1968 . The presence and percentage of RVFV-specific IgM , IgG and PCR-positive samples in persons in Kabale district indicates emergence of RVFV in Kabale is localized and that specific geographical locations may show varying levels of transmission or exposure to recently infected livestock . The level of IgG seropositivity indicates that RVFV has been circulating in the district for some time and further studies are needed to identify when this introduction may have occurred . Our study had some limitations . Prior to initiating the survey , no reliable estimates of animal and human RVFV seropositivity existed , so we do not know how representative our samples are of the general population in Kabale district . In analyzing the correlation between human seropositivity and animal seropositivity by subcounty , we assume that seropositive humans were most likely to be exposed to RVFV within the subcounty of their residence , and did not take into account possible exposure in other locations . Uganda has had a formal dedicated VHF surveillance program coordinated by UVRI and the MoH since 2011 . During the past several years , 11 independent VHF outbreaks have been detected and confirmed by the UVRI VHF laboratory , including an outbreak of Marburg virus in 2012 that was first detected from a case in Kabale district [40] . Because of the previous experience and enhanced training and awareness , surveillance for VHF cases has become a priority , and healthcare workers were able to quickly identify the initial RVFV cases and immediately send samples for laboratory confirmation . Enhanced VHF and RVFV surveillance and health education must be continued given the potential for future RVFV re-emergence . Alongside this serosurvey , a knowledge , attitude practice ( KAP ) study was conducted in Kabale district and the findings of this KAP study has been used to design health education materials targeting different stakeholders [41] . For example , the education materials targeting farmers and butchers emphasize reporting of any sick animals to veterinarians , washing hands after touching raw meat or milk , cooking meat and milk thoroughly , using of mosquito bed nets and wearing of more protective clothing when working in high risk areas . WHO recommends continued surveillance of RVFV during both outbreak and inter-epidemic periods [42] . Although RVFV rarely causes death in infected persons , the economic consequences due to loss of livestock and animal abortions can be devastating in agricultural communities [43] . Further studies are needed to fully understand the enzootic and endemic presence of RVFV in Kabale and surrounding districts . A longitudinal study is underway in Uganda to obtain samples from animals in Kabale district and throughout Uganda over several years to determine the endemicity and spread of RVFV over time . This will help determine the regions are at risk for RVFV outbreaks , so resources and surveillance efforts can be targeted to detect emergent cases and initiate any necessary control efforts . | Viral hemorrhagic fevers are known to cause high morbidity and mortality and pose a serious threat to human and animal populations in endemic countries . An outbreak of Rift Valley fever was detected in Kabale district in March , 2016 and identified the first human cases in Uganda since 1968 . There was a need to perform a rapid assessment of the burden of Rift valley fever in Kabale district , identify undetected acute cases , identify risk factors associated with human disease , identify areas at high-risk or future infections , and to determine if this was a newly emerging infection or an endemic disease . Our study found the seroprevalence to be as high as 28% in humans and 36% in animals within some subcounties of Kabale district . Human seropositivity correlated with animal seropositivity , suggesting that animal to human transmission may be the predominant mode of virus spread . Our findings also suggest that this virus may have been endemic for many years prior to these human cases being identified . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"livestock",
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"ruminants",
"pathogens",
"geographical",
"locations",
"microbiology",
"uganda",
"vertebrates",
"diet",
"tropical",
"diseases",
"animals",
"mammals",
"animal",
"products",
"viruses",
"rift",
"valley",
"fever",
"nutrition",
"rna",
"viruses",
"meat",
"neglected",
"tropical",
"diseases",
"insect",
"vectors",
"africa",
"bunyaviruses",
"infectious",
"diseases",
"food",
"zoonoses",
"sheep",
"medical",
"microbiology",
"microbial",
"pathogens",
"disease",
"vectors",
"insects",
"agriculture",
"goats",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"eukaryota",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"viral",
"diseases",
"amniotes",
"organisms"
] | 2018 | Prevalence and risk factors of Rift Valley fever in humans and animals from Kabale district in Southwestern Uganda, 2016 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.